Aphasia

Stem-, Spraak- en
Taalpathologie
Supplement, September 2013
14th International Science of Aphasia

Conference
This conference is financially supported by:
STEM-, SPRAAK- EN
TAALPATHOLOGIE 32.8310/01/1813-i

c Groningen University Press
Vol. 18, No. S01, 2013, pp. i-viii
PREFACE
Dear participants,
We are very pleased to welcome you to the 14th Science of Aphasia conference
which is held in Brussels, Belgium from September 20th to September 25th 2013.
The Science of Aphasia conferences are intended to bring together senior and
junior scientists working in the multidisciplinary field of the Neurocognition of
Language and it focuses on both the typical and atypical aspects of neurocognition.
The number of participants is restricted to about 120 in order to facilitate
interaction between the delegates. The focus of this year’s conference is on
Cognition, language and their impairments.
This year’s conference is organized by the members of the research group

“Clinical and Experimental Neurolinguistics” (CLIEN) of the Vrije Universiteit
Brussel in cooperation with the departments of Neurology of ZNA Middelheim
Hospital, Antwerp and the University Hospital of Brussels (UZ Brussel). The
primary aim of the research group CLIEN is to conduct innovative clinical and
experimental research in the multi-disciplinary field of brain-cognition-behaviour
relationships. A close cooperation between CLIEN and the related neurosciences
concentrates on a variety of clinical and experimental research topics, including
awake neurosurgery, cerebellar neurocognition and affective processing, atypical
cerebral organisation of linguistic and cognitive functions, aphasia, foreign accent
syndrome, chronic aphasia rehabilitation, aphasia in children, Landau-Kleffner
syndrome, the neurocognitive and neurobiological substrate of multilingualism
and multilingual education, neurolinguistic correlates of language attrition or
language loss.
The conference is held in the Academy Palace, which is the seat of the Royal
Flemish Academy of Belgium for Science and the Arts. This Royal Society aims
to stimulate independent and interdisciplinary reflection about societal and
scientific problems and developments in the arts. The Academy Palace was built
between 1823 and 1828 for Prince William of Orange: it is a fine example of
neoclassicist style which is characterized by harmonious geometrical proportions
and renassiance symmetry. Academy Palace is situated halfway between the Royal
Palace on one side of Warandepark and the Belgian Parliament on the other side.
Many Belgian ministers have their official residence in this neighbourhood. Take
the opportunity to wander around the modest garden of this building because it
has several interesting statues from well-known sculpturers such as Auguste Rodin.
Brussels is a city with many faces. It is well known as the Capital of Europe with
the European Parliament being situated just round the corner of this conference
venue. It is a historic city with origins going back to the 10th Century. Particularly
ii
spectacular is the historic market place with its magnificent 15th century Town
Hall and its beautiful Guild houses. Brussels is also the Capital of Art Nouveau with
many buildings that are recognized as World Heritage by UNESCO. The brilliant
designs of architects such as Victor Horta and Paul Hankar continue to amaze. But
first and foremost, Brussels is a city of the good life: excellent food, cosy pubs and
trendy restaurants, and of course Belgian beers some of which are unique in the
world.
We wish you an intellectually stimulating conference and a most enjoyable stay in

Brussels.
Jo Verhoeven & Peter Mariën
On behalf of the Local Organising Committee

iii
Organization
The 14th International Science of Aphasia Conference is held in Brussels, Belgium,

September 20 - 25, 2013
Local organizing committee 2013

Chair:
Prof. dr. Peter Mariën
Members:
Prof. dr. Raf Brouns
Dr. Roel Crols
Drs. Elke De Witte
Prof. dr. Sebastiaan Engelborghs
Prof. dr. Philippe Paquier
Drs. Esli Struys
Drs. Dorien Vandenborre
Prof. dr. Jo Verhoeven
Scientific committee
Chair:
Prof. dr. Roelien Bastiaanse
Members:
Dr. Wendy Best
Dr. Frank Burchert
Prof. dr. Ria De Bleser (Honoray Member)
Prof. dr. David Howard
Dr. Roel Jonkers
Prof. dr. Gabriele Miceli
Prof. dr. Lyndsey Nickels
Prof. dr. Brendan Weekes
Abstract Selection Committee

Dr. Wendy Best
Prof. dr. David Copland
Dr. Roel Jonkers
Prof. dr. Gabriele Miceli
Prof. dr. Isabell Wartenburger
Venue
ROYAL FLEMISH ACADEMY OF BELGIUM FOR SCIENCE AND THE ARTS
Paleis der Academiën
Hertogsstraat 1
B-1000 Brussel
Contact
Website: http://soa-online.com/
iv
Conference Program
Friday, September 20, 2013
Arrival, Registration & Welcome Reception

17:30 Registration
18:00 Welcome reception
Saturday, September 21, 2013
Conference Opening
Session 1: Language and Cognition

09:30 - 10:00 David Howard: Introduction
10:00 - 11:00 Dan Bub: Language and cognition
11:00 - 11:30 Coffee break
11:30 - 12:30 David Caplan: Language and cognition impairments
12:30 - 13:30 Lunch
13:30 - 14:30 Contributed Papers I

Djaina Satoer et al.: Long term cognitive functioning after glioma surgery in
eloquent areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Elke De Witte et al.: A standard neurolinguistic approach to awake brain surgery 4
Vânia de Aguiar et al.: Event related potentials of the processing of reflexives,
pronouns and referential violations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
14:30 - 15:30 Poster Session I

Annelies Aerts et al.: Gender differences in neurophysiological activation
patterns during phonological input processing: A contributory factor for
developing normative data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Brent E. Archer & Nicole Müller: Word retrieval in aphasic Sesotho-speakers:
Possible implications for current models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Katja Batens et al.: Clinical use of event-related potentials in diagnostic and
therapeutic evaluation of phonological input processes in the acute stage of
aphasia: a case study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
v
Cedric Boeckx et al.: On the theoretical characterization of agrammatism:

Resolving a paradox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Mattias De Coninck et al.: Bilateral language representation in a patient with
a large porencephalic cyst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Miet De Letter et al.: Phonological and semantic registration of the
subthalamic nucleus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Cécile De Somer et al.: Repetitions in the connected speech of a patient with
semantic dementia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Elke De Witte et al.: Atypical language dominance in a right-handed
patient: An anatomoclinical study with Direct Electrical Stimulation (DES) and
functional Magnetic Resonance Imaging (fMRI) . . . . . . . . . . . . . . . . . . . . . . . 34
Valantis Fyndanis: Subcortical lesions and agrammatic aphasia: A case study
in a highly inflected language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
15:30 - 16:00 Coffee Break
16:00 - 17:20 Contributed Papers II

Jennie Grassly et al.: Facilitating word retrieval in people with
aphasia: an exploration of the relationship between language and wider
neuropsychological processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Erminio Capitani et al.: Phonological similarity between target and semantic
errors in picture naming: Are aphasic patients a homogeneous group? A study
of 31 cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Irene Ablinger et al.: Eye movements tell us more about the underlying reading
strategy in lexical readers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Sandra Hanne et al.: The subject-object asymmetry in aphasic argument
question comprehension: Eye-tracking reveals the role of morphology . . . . . . . 55
18:00 - 19:00 Committee Meeting
Sunday, September 22, 2013

Session 2: Language Acquisition and Childhood Aphasia
09:30 - 10:00 Philippe Paquier: Introduction
10:00 - 11:00 Barbara Höhle: Language acquisition
11:00 - 11:30 Coffee break
11:30 - 12:30 Mieke van de Sandt: Childhood aphasia
12:30 - 13:30 Lunch
13:30 - 14:30 Contributed Papers III

Adrià Rofes et al.: Object naming may overestimate patient’s language
performance after neuro-oncological surgery: A case study . . . . . . . . . . . . . . . 59
Anastasiia Romanova et al.: Proper and common noun learning: Same or
different? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Dorien Vandenborre & Peter Mariën: Broca meets Wernicke in one single case . 67
vi
14:30 - 15:30 Poster Session II

Lucy Hughes et al.: The WORD project: a case series study on intervention for
developmental word-finding difficulties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Lívia Ivaskó et al.: First data from constraint induced aphasia therapy for
Hungarian patients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Fedor Christiaan Jalvingh & Roelien Bastiaanse: The influence of working
memory on the inflection of verbs in Alzheimer’s and Parkinson’s disease: a
case study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Bernard A. Jap et al.: Verb comprehension in aphasic speakers of Standard
Indonesian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Dörte de Kok et al.: VAST-App - testing verbs and sentences with the iPad . . . . 86
Polyxeni Konstantinopoulou et al.: Past tense in children with focal brain lesions 90
Mary H. Kosmidis et al.: Dichotic listening in professional simultaneous
interpreters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Elena G. Kozintseva et al.: Naming actions in non-fluent aphasia: an fMRI
study of compensatory reorganization of brain activity . . . . . . . . . . . . . . . . . . 98
Sam-Po Law et al.: An fMRI study of morphosyntactic processing in Chinese . 101
16:00 - 17:20 Contributed Papers IV

Ilona Damen et al.: The effect of static versus dynamic depictions of actions
in verb and sentence production in aphasia . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Sarah Vanhoutte et al.: Early and late semantic processing of action verbs:
evidence from fluent and stuttering speakers . . . . . . . . . . . . . . . . . . . . . . . . . 109
Reem S. W. Alyahya: Arabic-speaking aphasics: Analysis of naming errors . . . . 112
Seçkin Arslan et al.: A fragile category: Turkish evidential source markers in
agrammatism and bilingualism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Monday, September 23, 2013

Session III: Neurodegenerative Diseases
09:30 - 10:00 Evy Visch-Brink: Introduction
10:00 - 11:00 Rik Vandenberghe: Associative semantic network in
neurodegenerative disorder
11:30 - 12:30 Contributed Papers V

Rimke Groenewold et al.: The effects of (in)direct speech on aphasic discourse
comprehension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Chris Code et al.: Spreading the word ‘aphasia’. New international
comparisons of the public awareness of aphasia in Argentina, Canada, Croatia,
Greece, Norway and Slovenia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Stefanie Abel et al.: Construction and validation of a speech-systematic
aphasia screening (SAPS) and its appendent therapy regimen . . . . . . . . . . . . . 126
vii
12:30 - 13:30 Poster Session III

Dorota Leśniak: Language acquisition, learning and dissolution . . . . . . . . . . . 129
Christina Manouilidou et al.: Processing pseudo-words in mild cognitive
impairment: On-line and off-line evidence from Slovenian . . . . . . . . . . . . . . . 132
Anna Martínez-Álvarez: The role of the dorsal pathway in primary progressive
aphasia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Anna Martínez-Álvarez & Silvia Martínez-Ferreiro: Fronto-temporal pathways
and paraphasias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Verónica Moreno-Campos & Beatriz Gallardo-Paúls: Conversational turn
length and fluency measurement in aphasia . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Avanthi Paplikar et al.: Language mixing in discourse in bilinguals with aphasia145
Andrés Felipe Reyes& Roelien Bastiaanse: When object clitisation and climbing
happen alone, and when they dance cheek to cheek: Selective impairment in
Spanish agrammatism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Maria Varkanitsa et al.: Syntactic dependency resolution in Broca’s aphasia . . 153
Mile Vuković & Irena Vuković: The ability of verbal learning and memory in
patients with non-fluent aphasia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
13:30 - 14:30 Lunch
14:30 Afternoon Excursion and Conference Dinner
Tuesday, September 24, 2013

Session IV: The Cerebellum
09:30 - 10:00 Mario Manto: Introduction
10:00 - 11:00 Jeremy Schmahmann: Cognitive functions and the cerebellum
09:45 - 10:30 Peter Mariën: Language impairments and the cerebellum
12:30 - 13:30 Lunch
13:30 - 15:10 Contributed Papers VI

Laura S. Bos et al.: Understanding discourse-linked processes in agrammatic
and fluent aphasia: a threefold study in Russian . . . . . . . . . . . . . . . . . . . . . . . 160
Roberta Franceschet et al.: The process of diminutivization in patients with
language impairments and children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
Silvia Martínez-Ferreiro & Mireia Llinàs-Grau: Complex constructions across
aphasic syndromes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Francesca Franzon et al.: Exploring gender inflection: an insight from errors
in aphasia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
Tuba Yarbay Duman & İlknur Maviş: Comprehension of if-conditionals at the
morphosyntax-semantics interface in Turkish Broca’s aphasia . . . . . . . . . . . . . 174
15:30 - 16:30 Forward Looking Plenary Discussion & Floor Discussion

viii
Wednesday, September 25, 2013

Departure
STEM-, SPRAAK- EN
TAALPATHOLOGIE 32.8310/S01/1813-1

Vol. 18, No. S01, 2013, pp. 1-3
Long term cognitive functioning after glioma

surgery in eloquent areas
Djaina Satoer1 , Evy Visch-Brink1 , Marion Smits2 , Alfred Kloet3 ,
Clemens Dirven1 & Arnaud Vincent1
1 Department of Neurosurgery, Erasmus MC - Medical Center Rotterdam,
The Netherlands
2 Department of Radiology, Erasmus MC - Medical Center Rotterdam,
The Netherlands
3 Medical Center Haaglanden, the Hague, The Netherlands
Introduction
Cognitive performance is an important outcome measure in treatment of low-
grade gliomas (LGGs), since it is a crucial aspect of Quality of Life. LGGs
are slow growing brain tumours infiltrating the central nervous system, often
in the proximity of eloquent areas. During brain surgery, direct electrocortical
stimulation is nowadays used to identify individual functional boundaries to
prevent permanent neurological and/or cognitive damage1 . Previous studies have
shown that LGG patients have pre-operative deficits in one or more cognitive
domains, such as language, memory, attentional and executive functions which
may even deteriorate after glioma surgery. Most studies claim that these
impairments are transient and recover within 3 months2-4 . Our short follow-up at
3 months, however, still showed cognitive deterioration5 post surgery. Long term
follow-up is necessary to gain more insight into the course of recovery.
Methods
Cognitive functioning of 45 patients (mean age 39 y.) with presumed LGG in the left
hemisphere (apart from 3 patients) was assessed before awake craniotomy (T1) and
3 months (T2), and 1 year (T3) afterwards with an extensive neuropsychological
test-protocol: Aachener Aphasia Test-repetition, -reading aloud and -writing to
dictation; Boston Naming T est; Verbal (Category and Letter) Fluency; Verbal
Memory (15WT inprenting and recall); T rail Making T est A,B and Stroop Colour-
Word Test I-III. We compared pre- and post-operative mean scores of the patients
to normal population. Within the patient group, comparisons were made to
investigate the short term effect of surgery (T1-T2), the course of recovery (T2-T3)
and the long term effect of surgery (T1-T3). Correlation analyses were conducted
between significant change scores and tumour-characteristics, i.e. pathology (low-
high grade), volume and tumour localization (language or non-language).
2 SATOER ET AL.
Results
Compared to normal population, patients were impaired at T1 on BNT, Category
Fluency, Letter Fluency, 15WTinpr, 15WTrecall, TMTA, Stroop I,II, and III (p<.01,
Stroop III p<.05). At T2, mean performance was disturbed on the same tasks
in addition of TMTB (p<.05). At 1 year, Stroop interference was also impaired
(p<.05), whereas TMTA and TMTB recovered (p>.05). Within the patient group,
performance on 15WTrecall improved at T2, whereas deterioration was found
on Category Fluency (p<.05). Improvements were observed between T2 and T3
on BNT and Letter Fluency (p≤.05). There was no influence of tumour-related
variables on cognition, apart from a positive correlation between pre-operative
volume and 15WT (inprenting and recall) (Pearson r=-0.343, p=0.028; Spearman
r=-0.316, p=0.047, respectively).
Discussion
This is the first study that investigated the long term effects of glioma surgery on
cognition. Apart from deterioration on Category Fluency and improvement on
verbal memory (recall), surgery did not induce major cognitive changes. We found
that language recovery (naming and letter fluency) post surgery takes longer than 3
months, in contrast to what most studies have documented so far. The observed
improvement in phonological fluency at longer term may be accounted for by
bilateral frontal lobe compensation in tumour patients or by anterior/posterior
compensation (both LH and RH), whereas semantic fluency, which deteriorated,
may be more specific to left hemispheric functioning and thus less beneficial
of bilateral compensation6 . Selective improvement of memory and attentional
functions at both short and longer term after surgery was already observed2, 7 .
Short term improvement of verbal memory may be accounted for by the release of
mass effect, which remains stable due to the slow growth rate of LGGs (4 mm p/y)8 .
The long-term recovery in the executive domain could be mediated by a close
connection between verbal working memory neural networks (which improved)
and processes of selective attention9 . Tumour-characteristics and localization
were no additional risk-factors for cognitive change. These results underline
the importance of cognitive testing at longer term, with Category Fluency as an
essential task to assess, pre-, during, and post-operatively and also as a target for
rehabilitation. In addition, deterioration on the sensitive language tasks, BNT and
Letter Fluency, could possibly be a sign of tumour recurrence.
References
1. De Witt Hamer PC, Robles SG, Zwinderman AH, Duffau H, Berger MS. Impact of
intraoperative stimulation brain mapping on glioma surgery outcome: a meta-
analysis. J Clin Oncol. Jul 10 2012;30(20):2559-2565.
LONG TERM COGNITIVE FUNCTIONING AFTER GLIOMA SURGERY 3
2. Teixidor P, Gatignol P, Leroy M, Masuet-Aumatell C, Capelle L, Duffau H. Assessment

of verbal working memory before and after surgery for low-grade glioma. J
Neurooncol. Feb 2007;81(3):305-313.
3. Bello L, Gallucci M, Fava M, et al. Intraoperative subcortical language tract mapping
guides surgical removal of gliomas involving speech areas. Neurosurgery. Jan
2007;60(1):67-80; discussion 80-62.
4. Duffau H, Gatignol P, Mandonnet E, Capelle L, Taillandier L. Intraoperative
subcortical stimulation mapping of language pathways in a consecutive series of 115
patients with Grade II glioma in the left dominant hepisphere Journal of Neurosurgery.
2008;109:461-471.
5. Satoer D, Vork J, Visch-Brink E, Smits M, Dirven C, Vincent A. Cognitive functioning
early after surgery of gliomas in eloquent areas. J Neurosurg. 2012;117(5):831-838.
6. Goldstein B, Obrzut JE, John C, Hunter JV, Armstrong CL. The impact of low-
grade brain tumors on verbal fluency performance. J Clin Exp Neuropsychol. Sep
2004;26(6):750-758.
7. Gehring K, Sitskoorn MM, Gundy CM, et al. Cognitive rehabilitation in patients with
gliomas: a randomized, controlled trial. J Clin Oncol. Aug 1 2009;27(22):3712-3722.
8. Mandonnet E, Delattre JY, Tanguy ML, et al. Continuous growth of mean tumor
diameter in a subset of grade II gliomas. Ann Neurol. Apr 2003;53(4):524-528.
9. Gruber O, Goschke T. Executive control emerging from dynamic interactions between
brain systems mediating language, working memory and attentional processes. Acta
Psychol (Amst). Feb-Mar 2004;115(2-3):105-121.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 4-7
A standard neurolinguistic approach to awake

brain surgery
Elke De Witte1 , Djaina Satoer2 , Erik Robert3 , Henry Colle3 , Arnaud Vincent2 ,
Evy Visch-Brink2 & Peter Mariën1,4
1 Center for Linguistics, Vrije Universiteit Brussel, Belgium
2 Department of Neurosurgery, Erasmus MC - Medical Center Rotterdam,
The Netherlands
3 Department of Neurosurgery, AZ Sint-Lucas, Ghent, Belgium
4 Department of Neurologys, ZNA Middelheim, Antwerp, Belgium
Introduction
Intraoperative language mapping is increasingly used in patients operated on
for tumours in eloquent areas. Direct electrical stimulation (DES) enables to
identify critical cortical and subcortical language areas and pathways which cannot
be resected without permanent language deficits (Duffau, 2007). Although a
positive impact of DES on postoperative linguistic outcome is generally advocated,
the literature is only scantily documented with information about the linguistic
methods applied in awake surgery. Moreover, linguistic testing during DES is
generally limited to object naming and counting tasks (De Witte & Mariën, 2013;
De Witte et al., 2013). No studies exist in which in addition to spontaneous speech
(Satoer et al., 2013) a standardised linguistic protocol, assessing different linguistics
levels, is used to identify the critical language zones. For the first time we developed
a standardised linguistic test battery for awake surgery in critical language areas.
Methods
The test battery, called DuLIP (Dutch Linguistic Intraoperative Protocol) includes
phonological, semantic, syntactic and verbal motor production and perception
tests (see Table 1) (De Witte et al., 2013). For the composition of the linguistic tests,
the Dutch databases CELEX (Baayen et al., 1993) and SUBTLEX-NL (Keuleers et
al., 2010) were used and the test items were controlled for the variables frequency,
imageability, word length, morphological and phonological form.
A normative study was conducted in a control group of native Dutch-speaking
adults (N=250). Means and standard deviations were calculated per linguistic
test. In addition, DuLIP is used in a study group of patients with brain tumours
(N=10) in the pre-, intra- and postoperative phase of awake surgery. Intraoperative
anatomoclinical correlations were made and compared with preoperative fMRI
findings to identify eloquent language areas. Pre- and postoperative (6 weeks
postsurgery) linguistic test results will be compared in the patient group to study
early linguistic outcome.
Table 1: Intraoperative linguistic tests from DuLIP
Timing of Linguistic level Task Stimuli

assessment(s)
START DES Awakening phase - counting - counting from 1 to 10 (over and over again)
- automatic sequences - days of the week, months of the year
- orientation questions - Which day is it? Where are we?
DURING DES Phonological level - repetition of 3-syllabic words with alternating word - agénda, óliebol, hypothéék
(in 4 seconds) accents - koníjn, váder
- repetition of 2-syllabic words s - individu
- repetition of words with phonemic similarities - programma
- repetition of words with consonant clusters - De kok bakt een taart.
- repetition of sentences - rek, nek, mat, hek (answer mat)
- reading with phonological odd word out*
Semantic level - object naming, 100 objects* - black and white drawings of objects
- reading with semantic odd word out* - been, arm, raam, voet (answer raam)
- naming with semantic odd word out* - pictures of ‘borstel, hond, kat’ (answer borstel)
- semantic association task* - auto, fiets, ... (answer e.g. bus)
- sentence completion (semantically induced sentences)* - Hij snijdt met een ... (answer e.g. mes)
Syntactic level - verb generation* - bal -> gooien/werpen
- action naming (3rd person singular, transitive - de man ... (answer e.g. loopt) (e.g. picture of a
verbs), 60 actions* (RUG-VUB, Rofes A., Bastiaanse R.) man who is running)
Verbal motor - praxia - verbal diadochokinesis test - repeat /papapa/, /pataka/, /papopu/,
/pafpafpaf/, /dafnaflaf/, /pafpofpuf/, /pafpaspaf/
(5x)
NOT DURING Phonological level - phonological sentence judgment - De hokkel eet een gersie. (wrong)
DES - phonological fluency - Letters D, A, T
DURING
RESECTION
Semantic level - semantic sentence judgment - Het stoplicht wacht op de kameel. (wrong)
- semantic fluency - animals/jobs
- sentence comletion (less semantically induced sentences)* - Om 5 uur ...
Syntactic level - syntactic sentence jugdment - Hij koopte snoep. (wrong)
- verbal fluency (verbs) - verbs
* presented with a laptop screen and powerpoint + beep
A STANDARD NEUROLINGUISTIC APPROACH TO AWAKE BRAIN SURGERY 5
6 DE WITTE ET AL.
Subjects
The control group consisted of female and male adults with different educational
levels (primary, secondary, tertiary). Age distribution was selected on the basis of
current statistics of the prevalence of supratentorial tumours in the general adult
population (30% between 18-50 years, 55% between 50-75 years, 15% older than 75
years (Van Eycken, De Wever, 2006)). Control subjects were recruited from different
Flemish and Dutch provinces. The inclusion criteria were: Dutch as mother tongue,
no history of cardiovascular, neurologic, psychiatric, or developmental disorders,
no drug or alcohol abuse, normal vision, normal hearing, no excessive use of
medication and a Mini Mental State score higher than 24/30.
The study group included 10 patients with gliomas in the left hemisphere close
to language or motor areas. Exclusion criteria were as follows: history of a medical
or psychiatric condition known to affect cognitive functioning, permanent motor
or language deficits as a result of prior treatment, preexisting language deficits,
deafness or severe visual disorder, and mental retardation (Satoer et al., 2012).
Statistical methods
SPSS Statistics (v. 20) was used to analyse the data. All data were checked for
normality through a 1-sample Kolmogorov-Smirnov test and a Shapiro-Wilk test.
Non-parametric tests were used for data not normally distributed.
For the normative study, the data of 250 participants were investigated. To
assess the degree of association between two variables, a Pearson product-moment
correlation coefficient or Spearman correlation was calculated. The mean data
between two groups (gender) were compared using an independent samples t-
test or Mann-Whitney U test. The mean data among three or more groups
(age, education groups) were compared using analysis of variance (ANOVA).
Subsequently, Bonferroni post hoc analyses were conducted or the Kruskal Wallis
Test with Mann-Whitney U test. Finally, means, standard deviations, ranges and
cut-off scores were calculated for each linguistic subtest.
For the experimental study, the correlation between preoperative fMRI
activations and intraoperative positive stimulation points was calculated in
percentages. The pre- and postoperative data will be compared with the normative
data using a 1-sample t-test or the Wilcoxon signed-rank test to determine whether
they differ from the average of the normal group. Subsequently, the pre- and
postoperative data will be compared with paired-samples t-tests or the Wilcoxon
signed-rank test to evaluate early linguistic outcome.
Results
Analysis of control data revealed that performance on all linguistic subtests from
DuLIP is significantly affected by age and years of education, resulting in distinctive
age groups (18-49y; 50-74y, >75y) and education groups (primary, secondary,
tertiary). Means and standard deviations are provided for each age and education
A STANDARD NEUROLINGUISTIC APPROACH TO AWAKE BRAIN SURGERY 7
group per linguistic subtask. The intraoperative data of the study group revealed
a number of unexpected anatomoclinical findings that were not predicted by
preoperative fMRI findings or classic language-brain models. The correlation
between fMRI and DES was only 60%. Analyses of pre- and postoperative linguistic
results showed mainly impairment in fluency and naming tasks. The comparison
between pre-, intra- and postoperative linguistic results will be described by means
of some illustrative cases.
Discussion
With the development of a standardised linguistic test battery a valuable
instrument has now become available to reliably identify linguistic functions in
Dutch patients undergoing awake surgery in eloquent brain regions. Preliminary
patient data indicate that application of the test battery during DES and during
surgical resection of the tumour substantially increases intraoperative comfort
and preservation of linguistic function. In addition, this standardised linguistic
tool might be a valuable approach to enhance the scientific reliability of
the neurosurgical procedure. It allows a number of additional analyses and
comparisons of the data collected by means of the standardised protocols.
References
Baayen, R. H., Piepenbrock, R., & van Rijn, H. (1993). The CELEX lexical database (CD-ROM).
Philadelphia, PA: Linguistic Data Consortium, University of Pennsylvania.
Duffau, H. (2007). Contribution of cortical and subcortical electrostimulation in brain
gliomasurgery: Methodological and functional considerations. Neurophysiologie
Clinique/ Clinical Neurophysiology, 37, 373-82.
De Witte, E., & Mariën, P. (2013). The neurolinguistic approach to awake surgery reviewed.
Clinical Neurology and Neurosurgery, 115(2), 127-145.
De Witte, E., Satoer, D., Robert, E., Colle, H., Visch-Brink, E. G., & Mariën, P. (2013). Essentiële
taalzones detecteren tijdens wakkere neurochirurgie. In E. Robert, E. Visch-Brink, &
A.-S. Beeckman (Eds.), Het (voor)beeldig Brein. Taal & Interventionele Geneeskunde.
Garant.
Keuleers, E., Brysbaert, M., & New, B. (2010). SUBTLEX-NL: A new measure for Dutch word
frequency based on film subtitles. Behavior Research Methods, 42(3), 643-650.
Satoer, D., Vork, J., Visch-Brink, E., Smits, M., Dirven, C., & Vincent, A. (2012). Cognitive
functioning early after surgery of gliomas in eloquent areas. Journal of Neurosurgery,
1-8.
Satoer, D., Vincent, A., Smits, M., Dirven, C., Visch-Brink, E. (2013). Spontaneous speech
of patients with gliomas in eloquent areas before and early after surgery. Acta
Neurochirurgica, DOI 10.1007/s00701-013-1638-8.
Van Eycken, De Wever. (2006). Cancer Incidence and Survival in Flanders, 2000-2001.
Brussels: Flemish Cancer Registry Network (VLK).
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 8-12
Event related potentials of the processing of

reflexives, pronouns and referential violations
Vânia de Aguiar1,2,3,4 , Roelien Bastiaanse4,5 ,
Alexandra Reis6 & Olga Dragoy7,8
1 International Doctorate in Experimental Approaches to Language And the Brain
(IDEALAB), University of Potsdam, University of Groningen, University of Trento,

Macquarie University and Newcastle University
2 Center for Mind/Brain Sciences (CIMeC) and Center for Neurocognitive
Rehabilitation (CeRIN), University of Trento, Rovereto, Italy

3 Centro de Medicina de Reabilitação do Sul, São Brás de Alportel, Portugal
4 Center for Language and Cognition Groningen (CLCG), University of Groningen,
Groningen, The Netherlands

5 University Medical Center Groningen (UMCG), Groningen, The Netherlands
6 Cognitive Neuroscience Research Group, Departamento de Psicologia, Institute of
Biotechnology & Bioengineering, Centre for Molecular and Structural Biomedicine,

Universidade do Algarve, Faro, Portugal
7 National Research University Higher School of Economics, Moscow, Russia
8 Moscow Research Institute of Psychiatry, Moscow, Russia
Introduction
Discourse level processes are often challenging for individuals with agrammatic
aphasia (Avrutin, 2006; Bastiaanse, Bamyaci, Hsu, Lee, Yarbay Duman, &
Thompson, 2011). Research focused on these processes can help to explain the
impairment that underlies specific aphasic symptoms.
Several theories predict processing differences between reflexives and pronouns.
Whereas the processing of reflexives requires highly automatic processes that occur
within narrow syntax, the processing of pronouns requires the more expensive
process of linking to information previously mentioned in discourse (Pesetsky,
1987; Reinhart & Reuland, 1993; Reuland, 2001; Avrutin, 2006).
The establishment of discourse dependencies was compared with syntactic
dependencies using Event Related Potentials (ERPs), eliciting N400 effects
(Burkhardt; 2005; Burkhardt, 2007) and an Nref effect (Leitão, Branco, Piñango,
& Pires, 2009; Nref, Van Berkum, Brown and Hagoort, 1999). In these studies,
the negativities were interpreted as markers of the extra processing costs required
to establish discourse dependencies. A question remains: which are the extra
processes that make pronouns more costly to process?
From previous ERP research we also learned that syntactic information is used both
in co-indexation (number information, Harris, Wexler, & Holcomb, 2000) and co-
reference (gender information, Lamers, Jansma, Hammer, & Münte, 2006). Which
other sources of information are useful in the establishment of co-indexation and
coreference?
PROCESSING OF REFLEXIVES, PRONOUNS & REFERENTIAL VIOLATIONS 9
Methods
Participants
Twenty-eight right handed native speakers of European Portuguese with normal
or corrected to normal vision and audition (7 per list, 13 female) were included in
this dataset. Mean age was 42.59 (SD=11.31) and mean educational level 9.06 years
(SD=2.61).
Materials
We created 80 sentence pairs including verbs that can be used with a pronoun or a
reflexive in European Portuguese as in the example (1). Each participant heard 260
sentence pairs, in continuous natural speech (20 practice items, 80 experimental
items and 160 fillers).
(1) The carpenter[NP1] is with the client[NP2] . The carpenter[NP3] hurts
him/himself [NP4] with a hammer.
We used a 2x2 factorial design with the factor ‘sentence’ (levels: ‘pronominal’
or ‘reflexive’) and the factor ‘context’ (levels: ‘matching’ or ‘not mismatching’).
For the ‘not matching’ conditions, the ‘reflexive’ sentence was used with the
picture suitable for the ‘pronominal’ sentence, and vice-versa. The mismatching
conditions generated referential violations, given that the picture and the sentence
were incongruent in the identity of the antecedent for the reflexive and pronoun.
Procedure
For each experimental trial participants saw a picture and heard a sentence pair
containing either a reflexive or a pronoun. After sentence offset, participants
judged whether the sentence pair matched correctly the pictures. Continuous EEG
signal was recorded with 64 pin-type electrodes. External electrodes were used
for eye-movement artifact rejection and for offline referencing to the joint mastoid
average.
Results
Participants responded with high accuracy (87%) to sentences with reflexives and
pronouns, as well as to the referential violations. The comparison of pronouns
versus reflexives in sentences correctly matched to pictures revealed a greater
negativity for pronouns at anterior sites in the 200-300ms time window, larger at the
midline and right hemisphere electrodes (F(1, 27)=7.098, p=0.013). The referential
violation effect was similar for both reflexives and pronouns, consisting of a
negativity over central (F(1,27)= 6.751, p=0.015) and posterior (F(1,27)=10.772,
p=0.003) electrode sites from 300-500ms. For pronouns, the referential violation
effect started at posterior sites from 200-300ms, extending to central sites from
300-500ms (F(1, 27)=8.751,p=0.006). A comparison of referential violations for
reflexives and for pronouns revealed further negativity for the referential violation
10 DE AGUIAR ET AL.
of pronouns at anterior sites, from 500-800ms (F(1,27)=6.805, p=0.015). Figure 1

shows the ERP waveforms for each comparison.
Figure 1: Summary of ERP data for pronouns, reflexives and referential violation
Panel A: anterior electrode sites; RM = sentence with reflexive pronoun, matching the
picture; PM = sentence with personal pronoun, matching the picture. Panel B: central and
posterior sites; Matched = sentence with reflexive/pronoun, matching the picture; Violation
= sentence with reflexive/pronoun, not matching the picture. Panel C: anterior electrode
sites; RN = sentence with reflexive pronoun, not matching the picture; PN = sentence with
personal pronoun, not matching the picture.
Discussion
We found differences in the neurophysiological responses for pronouns and
reflexives when correctly matched with pictures, consistent with an N300 effect
(Barrett & Rugg, 1990). Such effect reflects the retrieval of image based
representations (Gunter & Bach, 2004) and the processing of global coherence
(West & Holcomb, 2002). Consider the example (1) above. Rule I (Grodzinsky
& Reinhart, 1993) states that coreference with discourse objects occurs if the use
of co-indexation changes the meaning of the sentence. Knowledge of this rule
can be used to exclude NP3 as an antecedent of the pronoun. At this point, NP1
and NP2 are both suitable antecedents considering grammar and meaning. The
retrieval of information based on context, as index by the N300 effect, may help to
disambiguate between them.
The referential violation effect found both for reflexives and pronouns matches the
characteristics of the N400 effect (Kutas & Hilliard, 1980). The referential violation
seems to be treated as a conceptual violation. This indicates that, even if syntactic
processes are used in co-indexation and co-reference, the acknowledgment of the
PROCESSING OF REFLEXIVES, PRONOUNS & REFERENTIAL VIOLATIONS 11
identity of an antecedent is achieved through conceptual and not syntactic steps.

Another relevant finding is that, even though context is important to interpret
pronouns, participants respond accurately when context provides misleading
information. We assert that participants here use a Semantic-Pragmatic Inference
that allows them to disambiguate between the possible antecedents available in
linguistic discourse.
(2) If NPx does not co-refer with NPy [a pronoun], then NPs that co-refer with NPx
do not co-refer with NPy.
A direct comparison of the two violation types originated a sustained anterior
negativity, from 500 to 800ms after the critical word, consistent with an Nref
effect (Van Berkum et al., 1999). This effect supports our claim that contextual
information is particularly relevant for the processing of pronouns. The Nref
denotes the controlled processes necessary to disambiguate between possible
antecedents, such as the application of a Semantic-Pragmatic Inference and the
selection of the correct discourse unit.
References
Avrutin, S. (2006). Weak syntax. In K. Amunts & Y. Grodzinsky (Eds.), Broca’s Region (pp.
49-62). New York: Oxford Press.
Bastiaanse, R., Bamyaci, E., Hsu, C.J., Lee, J., Yarbay Duman, T. Y., & Thompson, C. K.
(2011). Time reference in agrammatic aphasia: a cross-linguistic study. Journal of
Neurolinguistics, 24(6), 652-673. doi:10.1016/j.jneuroling.2011.07.001
Barrett, S. E. & Rugg, M. D. (1990). Event-related potentials and the semantic
matching of pictures. Brain and Cognition, 14(2), 201-12. Retrieved from
http://www.ncbi.nlm.nih.gov/pubmed/2285513
Brown, C. & Hagoort, P. (1993). The processing nature of the N400: evidence from masked
priming. Journal of Cognitive Neuroscience, 5(1), 34-44. doi:10.1162/jocn.1993.5.1.34
Burkhardt, P. (2005). Evidence from processing: event-related potentials. In P. Burkhardt
(Ed.), The Syntax-Discourse Interface: Representing and interpreting dependency (pp.
196-211). Linguistik Aktuell 80. Amsterdam/Philadelphia: John Benjamins.
Burkhardt, P. (2007). Reference assignment in the absence of sufficient semantic content.
In M. Schwarz-Friesel, M. Consten, M. Knees (Eds.), Anaphors in Text: Cognitive,
formal and applied approaches to anaphoric reference (pp. 241-258). John Benjamins
Publishing Company: Amsterdam/Philadelphia.
Grodzinsky, Y. & Reinhart, T. (1993). The innateness of binding and coreference. Linguistic
Inquiry, 24(1), 69-101. Retrieved from http://www.jstor.org/stable/4178802
Gunter, T. C. & Bach, P. (2004). Communicating hands: ERPs elicited by meaningful symbolic
hand postures. Neuroscience Letters, 372(1-2), 52-6. doi:10.1016/j.neulet.2004.09.011
Harris, T., Wexler, K. & Holcomb, P. (2000). An ERP investigation of binding and coreference.
Brain and Language, 75(3), 313-46. doi:10.1006/brln.2000.2318
Kutas, M. & Hillyard, S. A. (1980). Reading senseless sentences: brain potentials reflect
semantic incongruity. Science, 207(4427), 203-205. doi: 10.1126/science.7350657
Lamers, M. J. A., Jansma, B. M., Hammer, A., & Münte, T. F. (2006). Neural correlates of
semantic and syntactic processes in the comprehension of case marked pronouns:
12 DE AGUIAR ET AL.
evidence from German and Dutch. BMC Neuroscience, 7, 23. doi:10.1186/1471-2202-

7-23
Leitão, J., Branco, A., Piñango, M., & Pires, L. (2009). Pronoun resolution to commanders
and recessors: a view from event-related brain potentials. In S. Devi, A. Branco, & R.
Mitkov (Eds.), DAARC 2009, LNAI 5847 (pp. 107-120). Berlin Heidelberg: Springer-
Verlag.
Pesetsky, David (1987) Wh-in-Situ: movement and unselective binding. In E. Reuland, and A.
ter Meulen, (Eds.) The representation of (In)Definiteness, Cambridge, MA: MIT Press.
Reinhart, T., & Reuland, E. (1993). Reflexivity. Linguistic Inquiry, 24(4), 657-720. Retrieved
from http://www.jstor.org/stable/4178836
Reuland, E. (2001). Primitives of Binding. Linguistic Inquiry, 32(3), 439-492. Retrieved from
http://www.jstor.org/pss/4179157
Van Berkum, J., Brown, C. M., & Hagoort, P. (1999). Early referential context effects
in sentence processing: evidence from event-related brain potentials. Journal of
Memory and Language, 41(2), 147-182. doi:10.1006/jmla.1999.2641
West, W. C., & Holcomb, P. J. (2002). Event-related potentials during discourse-level semantic
integration of complex pictures. Cognitive Brain Research, 13(3), 363-75. Retrieved
from http://www.ncbi.nlm.nih.gov/pubmed/11919001
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 13-15
Gender differences in neurophysiological

activation patterns during phonological input
processing: A contributory factor for
developing normative data
Annelies Aerts1,2 , Pieter van Mierlo3 , Robert J. Hartsuiker4 ,
Patrick Santens1,2 & Miet De Letter2,5
1 Faculty of Medicine and Health Sciences, Ghent University, Belgium
2 Department of Neurology, Ghent University Hospital, Belgium
3 Medical Image and Signal Processing Group, Department of Electronics and
Information Systems, Ghent University - IMinds, Belgium

4 Department of Experimental Psychology, Ghent University, Belgium
5 Department of Speech, Language and Hearing Sciences, Ghent University, Belgium
Introduction
Differences between men and women with respect to language perception have
been a matter of debate for several decades (Wallentin, 2009). To date no
consensus has been reached regarding structural, functional or neurophysiological
brain research, showing inconsistencies in either laterality patterns or in
neurophysiological measures such as amplitude and latency. With the present
study we wanted to elaborate on potential gender effects in neurophysiological
activation patterns during phoneme discrimination and word recognition by
means of event-related potentials (ERPs). With this information we wanted to
ascertain whether gender has to be considered as a contributing factor when
developing normative data for use of cognitive ERPs in acquired language disorders
(aphasia).
Methods
Twenty-four women and twenty men were included in the study, with an
equivalent mean age in men and women (p = 0.785). Phoneme discrimination
was investigated by means of six oddball paradigms in which we differentiated
between three phonemic contrasts (place of articulation, voicing and manner of
articulation) in both an automatic (Mismatch Negativity; MMN) and controlled
(P300) condition. Word recognition was investigated by contrasting real words with
pseudowords, but only in an automatic setting.
During the administration of the above tasks an electroencephalogram (EEG) was
recorded through 24 Ag/AgCl-electrodes placed on the scalp according to the
international 10-20 system. Data was analyzed using BrainVision Analyzer (Brain
Products, Munich, Germany) to elicit the cognitive ERPs of interest (MMN, P300
during phoneme discrimination; N100, P200 and N400 during word recognition)
14 AERTS ET AL.
and with special attention to potential gender effects and possible interaction with
phonemic contrasts, attention level or word type.
Results
On the level of phoneme discrimination, women exhibited larger P300 and MMN
amplitudes (see figure 1) than men (p < 0.01), but only in the condition where
place of articulation was the phonemic contrast. Within women larger P300
amplitudes were found for place of articulation compared to voicing and manner
of articulation (p < 0.05) which did not occur in the MMN condition. Men did
not show differences in either phonemic contrast condition or the automatic or
controlled condition. Moreover, in the automatic condition women showed a
trend towards a more bilateral distribution whereas men displayed a left-lateralized
preference (p = 0.068). On the contrary, in the controlled condition a reverse
pattern emerged showing more left lateralization in women in posterior regions
for all three phonemic contrasts and more bilateral activation in men in posterior
regions for place and manner of articulation (p < 0.05).
Figure 1: Gender difference during controlled PoA phoneme discrimination displayed in

Pz electrode. (a) Women showed a larger P300 amplitude than men when PoA was the
discriminating phonemic contrast during controlled phoneme discrimination; men = black,
women = grey. (b) The interaction between Contrasts and Gender is presented graphically;
* = significant difference.
Concerning word recognition, an N400 pseudoword effect was detected around 500
ms (p < 0.001). However, the pseudowords already displayed larger amplitudes
100 ms post-stimulus (N100; p < 0.01), which continued in the P200 (p < 0.001)
and eventually N400 time window, in both men and women and with a bilateral
activation pattern throughout. A gender effect did occur when we examined
the processing speed (= efficiency, accuracy). In the P200 time window women
appeared to be slower in processing real words (higher latency) than pseudowords
(p < 0.01) whereas the opposite pattern arose in the N400 time window showing
faster real word processing (shorter latency) than pseudoword processing (p <
0.05). Men showed no differentiation in speed of processing real words and
GENDER DIFFERENCES IN PHONOLOGICAL PROCESSING 15
pseudowords, but in the N400 time window they processed real words much slower
than women (p < 0.01).
Discussion
The present study further contributed to the knowledge of gender-related
differences in brain activation patterns during language perception. Women
displayed a larger sensitivity to spectrotemporal differences related to the
phonemic contrasts during phoneme discrimination. This was evidenced by larger
responses to the PoA contrast compared to the other contrasts in the controlled
condition and larger responses than men in the controlled and automatic PoA
condition. Men did not demonstrate such sensitivity. The degree of attention
played an important role as well, which was even more ratified by the “switch” in
laterality patterns in posterior regions within men and women in the automatic and
controlled condition.
During word recognition the pseudoword effect was already established 100
ms after stimulus presentation, indicating that the lexical effect started early,
irrespective of gender status. The difference between men and women became
apparent in the processing accuracy and speed of real word-pseudoword
dissociation, showing more efficiency in women. Both men and women showed a
bilateral activation pattern during word recognition, arguing for a clear separation
between language levels when investigating gender effects, considering the current
gender differences in more basic phonological processes. In conclusion, gender
should definitely be looked upon as a contributory factor when developing
normative data. As such it is recommended to expand existing normative data for
age (Aerts et al., 2013) in order to create a justified distinction between men and
women.
References
Aerts, A., van Mierlo, P., Hartsuiker, R. J., Hallez, H., Santens, P. & De Letter, M.
(2013). Neurophysiological investigation of phonological input: Aging effects and
development of normative data. Brain & Language, 125(3), 253-263.
Wallentin, M. (2009). Putative sex differences in verbal abilities and language cortex: A
critical review. Brain & Language, 108, 175-183.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 16-18
Word retrieval in aphasic Sesotho-speakers:

Possible implications for current models
Brent E. Archer & Nicole Müller
University of Louisiana at Lafayette, USA
Many psycholinguistic studies of naming have been informed by two models

of word retrieval. The WEAVER++ and the Interactive Lexical Network models
have much in common, but there are many contrasts between these two
conceptualizations (Levelt, 1999). In particular, feedback is thought to operate
differently under the two models. WEAVER++ proponents argue that while
feedback may flow from the lexical level to the semantic conceptual level, reverse
activation from the phonological level to the lexical level is not possible (Dell,
Nozari and Oppenheim, in press). Conversely, the framers of the Interactive Lexical
Network view feedback as being more ubiquitous; under this model, feedback may
flow freely throughout the word retrieval system. Events at the phonological level
can influence activity at all other levels in the system (Dell, Nozari and Oppenheim,
in press).
The current study focuses on naming difficulties in two speakers of Sesotho, a
Bantu language spoken by about 4 million people in South Africa (Lewis, Simons
and Fennig, 2013). Sesotho is a noun class language; most nouns are members of a
series of classes (Doke and Mofokeng, 1967). This study focused on singular/plural
marking. In Sesotho, nouns are inflected for number through a system of prefixes
(eg. lehapu ‘watermelon’ mahapu ‘watermelons’). The two participants in this
study were first language speakers of Sesotho. Both participants developed more
phonologically-based anomias after CVAs. In the case of T. (Sesotho-adapted WAB
Aphasia Quotient:66), anomia appeared to be the hallmark symptom, while S.
(Sesotho-adapted WAB Aphasia Quotient:42) experienced anomic moments as part
of a broader expressive aphasic syndrome.
The aim of this study was to determine the effects of two cueing techniques for
facilitating naming in Sesotho speakers with anomia. Two cue types (treatments)
were studied: cues based on the initial phoneme of the full (prefix+root) target
item, and cues based on the initial phoneme of the uninflected root.
Initial phoneme cues are a widely described and used therapy technique for
anomia (for example Nettleton and Lesser 1991; DeDe Parris and Waters 2003 ;
Maher and Raymer 2004; Best, et al, 2002). Because of the morphosyntactic profile
of Sesotho, initial phoneme cues amount to prefix-based cues or PBCs (e.g. for the
target ‘lehapu’, PBC /l-/ would be based on the noun prefix, le-). Cues based on
the initial phoneme of the uninflected root represent a novel approach. Within the
parameters of Sesotho, these cues might be described as root-based cues or RBCs
(e.g. for the target ‘lehapu’, the RBC /h-/ would be based on the first sound of the
uninflected root -hapu).
WORD RETRIEVAL IN APHASIC SESOTHO-SPEAKERS 17
Methods
The researcher developed and balanced 2X 200 item word lists using a variety of
criteria. Each technique was allocated a word list. A confrontation picture naming
paradigm was used to deliver the treatments. Assessment of pre and post naming
abilities under the two conditions (PBC and RBC) yielded data. All participant
productions and errors during the study were recorded and coded according to
established guidelines found in the literature.
The Allison-MT procedure, which was developed specifically for use in single or
small group treatment studies (Brossart, et al, 2006), was used to analyse the
treatment data. A Chi-squared test for association procedure was used to analyse
the error data.
Results
The treatment data analysis suggests that although both techniques were
associated with an increase in naming ability, PBC is less effective at remediating
anomia in Sesotho speakers than the provision of a cue based on the first phoneme
of the uninflected form.
Discussion
The two models discussed above were examined, and the researcher attempted
to select the model with the greatest explanatory power. Of the two models
discussed, the Interactive Lexical Network appears to align more closely with the
current findings. Participants’ performance under the two treatment conditions
lend further credence to the notions of widespread interactive feedback featured in
this model.
The analysis of the error data was undertaken to determine if the error patterns
noted might provide further support for the suggestions emanating from the
treatment data. In essence, the researcher wished to ascertain if either cue
condition was associated with a signficant growth in the proportions of any error
type over the course of the experiment. In the case of T., PBCs appeared to be linked
to an increase in the percentage of circumlocution-type errors. For participant S.,
there seemed to be a connection between PBCs and a growth in the percentage of
semantic-paraphasia type errors. If the interpretive frameworks provided by the
models are applied to the error data, the Interactive Lexical Network appears to
provide a more plausible account of word retrieval in the two participants.
As the first study of this kind conducted in a non-Indo European language, this
experiment provides further information about the process of word retrieval. The
cumulative effect of such endeavours can help to gradually provide more detail in
currently underspecified models of word retrieval.
18 ARCHER & MÜLLER
References
Best, W.„ Herbert,R., Hickin, J., Osborne, F., and Howard, D. (2002) Phonological and
orthographic facilitation of word retrieval in aphasia: Immediate and delayed effects.
Aphasiology 16 (1-2):151-168.
Brossart, D., Parker, R., Olson, E., and Mahadevan, L. (2006) The Relationship between
visual analysis and five statistical analyses in a simple AB single-case research design.
Behaviour Modification 30(5): 531-563.
DeDe, G., Parris, D., and Waters, G. (2003) Teaching self-cues: A treatment approach for
verbal naming. Aphasiology 17 (5): 465-480.
Dell, G., Nozari, N., and Oppenheim, G. (in press) Word production: behavioral and
computational considerations. Centre for Research in Language. Web 24 March 2013
<http://crl.ucsd.edu/~goppenheim/pubs/
dellNozariOppenheim_accepted_oxfordLangProd.pdf>
Doke, C., and Mofokeng, S. (1967). Textbook of southern Sotho grammar. London:
Longmans.
Levelt, W. (1999) Models of word production. Trends in Cognitive Sciences 3(6): 223-232.
Lewis, P., Simons, G., and Fennig,C. (2013) Ethnologue: Languages of the World. Dallas: SIL
International.
Maher, L., and Raymer, A. (2004) Management of anomia. Topics in stroke rehabilitation 11
(1):10-21.
Nettleton, J., and Lesser, R. (1991) Application of a cognitive neuropsychological model to
therapy. Journal of Neurolinguistics 6 (2):139-157.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 19-21
Clinical use of event-related potentials in

diagnostic and therapeutic evaluation of
phonological input processes in the acute stage
of aphasia: a case study
Katja Batens1 , Miet De Letter2,3 , Robrecht Raedt3 , Annelies Aerts4 ,
Wouter Duyck5 , Dirk Van Roost6 & Patrick Santens3
1 Institute of Neurosciences, Department of Neurology, Ghent University Hospital,
Belgium
2 Department of Oto-Rhino-Laryngology and Logopaedic-Audiologic Sciences,
Ghent University, Belgium

4 Departement of Internal Medicine, Ghent University, Belgium
5 Department of Experimental Psychology, Ghent University Hospital, Belgium
6 Department of Neurosurgery, Ghent University Hospital, Belgium
Introduction
Neuroanatomical imaging and behavioural language testing cannot provide

enough information to disentangle underlying disturbed neurophysiological
language processes in aphasic patients. Moreover, they cannot provide insight
in the neurophysiological recovery patterns. Especially in the (sub) acute phase
of aphasia, it is important to have a good insight in the neuroplastic changes,
since different processes like spontaneous recovery and therapeutic interventions
(medical and paramedical) simultaneously influence the reorganisation after
stroke. Event related potentials (ERP) offer the opportunity to acquire information
about the timing and amplitude of neural activity. It has been reported that there
is a correlation between behavioural and electrophysiological testing (Pettigrew
et al., 2005), although ERP’s seem to be more sensitive (Elting et al., 2008).
Electrophysiology may contribute to the development of a disorder-oriented
rehabilitation approach in the acute stage and to follow-up.
Phonology is the most disturbed linguistic modality in aphasia, due to its

distributed networks and its involvement in both language production and
comprehension. In this study the behavioural and electrophysiological evolution
of the phonological input processes of a single subject are described during the
first four weeks after stroke. The electrophysiological results are compared with
normative data for the Flemish population (Aerts et al., 2013)
20 BATENS ET AL.
Methods
Patient
A 46-year-old right-handed male patient, who suffered an ischemic
cerebrovascular accident of the left middle cerebral artery, was included in
this study. There were no signs of previous speech or language disorders and
no hearing impairment. The initial language production can be described
as telegraphic speech with severe word finding difficulties and phonological
paraphasia. There were only mild comprehension problems in spontaneous
conversations.
Linguistic evaluation
The behavioural testing consisted of the Aachen Aphasia Test (AAT(Graetz,
De Bleser, & Willmes, 2005)and the Psycholinguistic Assessment of Language
Processing in Aphasia (PALPA; (Kay, Lesser, & Coltheart, 1996)of which three
phonological subtests (auditory discrimination of non-words, PALPA 1; auditory
discrimination of minimal pairs, PALPA 2; and auditory lexical decision, PALPA 5)
and the auditory memory for digits (PALPA 12) were administered.
Neurophysiologically, two different oddball paradigms were created to evaluate
attended (P300) and unattended (mismatch negativity; MMN) auditory
discrimination and unattended word recognition (real versus non-words).
The auditory discrimination paradigm was then subdivided according to the three
distinctive characteristics present in Dutch language (place of articulation (PoA),
manner of articulation (MoA) and voicing)
All behavioural and neurophysiological tests were carried out within the first week
after stroke and were repeated after completion of all therapy sessions.
Language therapy
An intensive tailor made training program focussing on the connection of
phonological and semantic processes was developed. 30 hours of therapy were
provided in a 3-week period. Each therapy session lasted 2 hours and took place
at the patient’s home. Therapy started one week after stroke.
Results
The results described below are preliminary results. There was substantial noise in
the P300 recordings, which hampered interpretation of results.
Before therapy
The ALLOC classification of the AAT suggested Broca’s aphasia without outliners in
the subtests. Auditory memory was severally impaired (25/60).The patient reached
maximum scores on the behavioural auditory discrimination tests. These were
only partially confirmed by the neurophysiological results, since deviant auditory
ERP IN DIAGNOSTIC AND THERAPEUTIC EVALUATION IN ACUTE APHASIA21
discrimination MMN’s were recorded for the distinctive characteristics MoA and
voicing when comparing to normative data (Aerts et al., 2013).
PALPA 5 only revealed difficulties recognizing non-words (real words: 79/80; non-
words 66/80). In the word recognition paradigm, the N400 was absent in the ERP
signal for real words. Furthermore, no ERP could be elicited for non-words.
After therapy
AAT analyses indicated an overall significant improvement up to the level that the
ALLOC classification could not assigns the problems to an aphasic syndrome.
The behavioural linguistic tests showed an overall significant improvement on the
tests that had not obtained a maximum score prior to therapy (PALPA 5 and PALPA
12)
Neurophysiologically, improvements on MMN of auditory discrimination could
be recorded; the latency decreased for voicing, where for MoA the amplitude
increased. In the word recognition paradigm, a N400 was detectable for words and
there was an ERP visible for non-words.
Discussion
The results confirm the hypothesis of a higher sensitivity of electrophysiological
examination as compared to behavioural testing. The electrophysiological results
are able to detect abnormalities in case of ceiling effects on behavioural testing or
when therapeutic progress is not behaviourally measurable.
On the other hand, the interpretation of ERP’s at single subject level remains
difficult, especially for paradigms that necessitate cooperation, like P300. Even
for unattended paradigms, enough trials are necessary in order to reduce noise to
acceptable levels.
When cautiously used, ERP’s can be clinically useful in diagnostic and therapeutic
evaluation of phonological input processes in single subjects with aphasia.
References
Aerts, A., van Mierlo, P., Hartsuiker, R. J., Hallez, H., Santens, P., & De Letter, M.
development of normative data. Brain and Language, 125(3), 253-263.
Elting, J. W., Maurits, N., van Weerden, T., Spikman, J., De Keyser, J., & van der Naalt, J. (2008).
P300 analysis techniques in cognitive impairment after brain injury: Comparison
with neuropsychological and imaging data. Brain Injury, 22(11), 870-881.
Graetz, P., De Bleser, R., & Willmes, K. (2005). Akense afasie test: Harcourt Test Publishers.
Kay, J., Lesser, R., & Coltheart, M. (1996). Psycholinguistic assessments of language
processing in aphasia (PALPA): An introduction. Aphasiology, 10(2), 159-180.
Pettigrew, C. M., Murdoch, B. E., Kei, J., Ponton, C. W., Alku, P., & Chenery, H. J. (2005).
The mismatch negativity (MMN) response to complex tones and spoken words in
individuals with aphasia. Aphasiology, 19(2), 131-163.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 22-24
On the theoretical characterization of

agrammatism: Resolving a paradox
Cedric Boeckx1,2 , Anna Martínez-Álvarez2 & Evelina Leivada2
1 ICREA, Spain
2 Universitat de Barcelona, Spain
Introduction
This paper proposes a novel way to resolve a paradox that emerges between
findings in the agrammatism literature and the cartographic hierarchy of
functional projections (e.g., (1)) on which many of the agrammatism studies
rely.
(1) [CP C’ [AgrP Agr’ [TP T’ [VP [V’]]]]] (Belletti 1990)
Findings from agrammatism suggest that there is a crosslinguistic pattern

according to which higher nodes are more affected than lower ones. According to
Friedmann & Grodzinsky’s (1997) Tree-Pruning Hypothesis (TPH), T(ense) shows
up as impaired while Agr(eement) is preserved. This argument has received
empirical support from several languages (e.g., Friedmann & Grodzinsky 1997,
2000 and Friedmann 1998 et seq. for Palestinian Arabic and Hebrew, Stavrakaki &
Kouvava 2003 for Greek, Martínez-Ferreiro 2009 for Catalan, Galician and Spanish).
Based on what these studies propose for the impairment of T and Agr, the hierarchy
of these two nodes that one expects to see is T >Agr (T higher than Agr, as in (2)).
(2) [CP C’ [TP T’ [AgrP Agr’ [VP [V’]]]]] (Pollock 1989)
The paradox we focus on lies in the fact that the hierarchy put forth in cartographic
studies (e.g., Belletti 1990) is Agr >T. Relating (1) and much work since then to the
TPH and to findings from studies on agrammatism, a clash is observed between
what the cartographic representation predicts as impaired and what what the
agrammatic literature has shown.
Methods
We bring together findings from various experiments in order to shed light to
the aforementioned paradox and we comparatively discuss previous explanations
that are offered for it in the literature. More specifically, Bastiaanse & Jonkers
(2012) pinpoint this conflict between theory and results from agrammatism studies
and discuss it also from the perspective of Wenzlaff & Clahsen’s (2005) Tense
Underspecification Hypothesis: Tense problems might be due to the fact that tense
morphology conveys extrasentential information. This implies that the reason
ON THE THEORETICAL CHARACTERIZATION OF AGRAMMATISM 23
for impaired production is related to time reference and not to the position that
T has in relation to Agr on the syntactic tree. Another answer to the paradox
is offered by Nanousi et al. (2006) along the lines of feature interpretability.
According to their Impaired Interpretable Feature Hypothesis, functional categories
that bear interpretable features, such as Tense, are impaired in agrammatic
production because their morphological realization is dysfunctional. On the
contrary, uninterpretable features are checked through Agree operations and
these functional categories show up intact. These two explanations might be in
conformity with the reports about the production of T and Agr in the agrammatic
literature, however they do not offer insights with respect to what differentiates
clitic pronouns from other Agr markers.
Results
Taking as its departure point the results generated by experimental studies that
examine the production of both T, Agr and clitics (e.g., Martínez-Ferreiro 2009) and
aiming to approach the T impairment/Agr preservation also in relation to the status
of clitics, the present paper proposes to resolve the aforementioned paradox in a
novel way by assuming feature inheritance from Agr to T in line with Chomsky
(2007) and Richards (2007). More concretely, we assume that Agr features start
off on C higher than T (consistent with (1)), but are below T at the time transfer
occurs, due to inheritance. The reported impaired production of T in agrammatic
literature entails that T is accessed post-syntactically after transfer and explains
why agrammatic production seems to correspond to a structure like (2).
Discussion
Our analysis receives empirical support from the clitics vs. T difference,
discussed in the agrammatic literature, along the lines of feature inheritance and
interpretability. More specifically, Martínez-Ferreiro (2009) reports significantly
higher percentages of impaired production of clitics compared to impaired
production of T for both mild and moderate agrammatic subjects. This is
expected according to our analysis: Clitics are hosted in a functional projection
residing between the CP and the IP, a projection that is higher than T (Raposo
and Uriagereka 2005), hence clitics are more susceptible to impairment. Being
agreement markers that bear interpretable features due to their argumental nature
(Roberts 2010), clitics do not lower for interpretability purposes. They remain
higher than T and therefore the dissociation observed between clitics and T in
the production of agrammatic subjects in terms of the former being more severely
impaired than the latter can be explained by assuming that no inheritance takes
place in the case of clitics precisely because of their status as interpretable markers.
24 BOECKX ET AL.
References
Bastiaanse, R. & R. Jonkers (2012). Linguistic accounts of agrammatic aphasia. In R.
Bastiaanse & C. Thompson (eds.), Perspectives on Agrammatism, 17-33. Hove:
Psychology Press.
Belletti, A. (1990). Generalized Verb Movement. Torino: Rosenberg and Sellier.
Chomksy, N. (2007). Approaching UG from below. In U.Sauerland & H.-M. Gärtner (eds.),
Interfaces + Recursion = Language? Chomsky’s Minimalism and the View from
Syntax-Semantics, 1-30. Berlin: Mouton de Gruyter.
Friedmann, N. & Y. Grodzinsky (1997). Tense and agreement in agrammatic production:
Pruning the syntactic tree. Brain and Language, 56, 397-425.
Friedmann, N. & Y. Grodzinsky (2000). Split inflection in neurolinguistics. In M. A.
Friedemann and L. Rizzi (eds.), The Acquisition of Syntax: Studies in Comparative
Developmental Linguistics, 84-104. New York: Longman.
Friedmann, N. (1998). Functional Categories in Agrammatic Production: A Crosslinguistic
Study. Doctoral dissertation. Tel Aviv University.
Martínez-Ferreiro, S. (2009). Towards a Characterization of Agrammatism in Ibero-
Romance. Doctoral dissertation. Universitat Autò’noma de Barcelona.
Nanousi, V., J. Masterson, J. Druks & M. Atkinson (2006). Interpretable vs. uninterpretable
features: Evidence from six Greek-speaking agrammatic patients. Journal of
Neurolinguistics, 19, 209-238.
Pollock, J. (1989). Verb movement, Universal Grammar and the structure of IP. Linguistic
Inquiry, 20, 365-424.
Raposo, E. & J. Uriagereka (2005). Clitic placement in Western Iberian: A Minimalist view. In
G. Cinque & R. Kayne (eds.), The Oxford Handbook of Comparative Syntax, 639-697.
Oxford: Oxford University Press.
Richards, M. (2007). On feature inheritance: An argument from the phase impenetrability
condition. Linguistic Inquiry, 38, 563-572.
Roberts, I. (2010). Agreement and Head Movement: Clitics, Incorporation and Defective Goals.
Cambridge, MA: MIT Press.
Stavrakaki, S. & S. Kouvava (2003). Functional categories in agrammatism: Evidence from
Greek. Brain and Language, 86, 129-141.
Wenzlaff, M. & H. Clahsen (2005). Finiteness and verb-second in German agrammatism.
Brain and Language, 92, 33-44.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 25-28
Bilateral language representation in a patient

with a large porencephalic cyst
Mattias De Coninck1 , Lise Van der Haegen2 , Wim Van Hecke3 ,
Qing Cai4,5 , Debby Van Dam1 , Peter P. De Deyn1,6,7,8 ,
Marc Brysbaert2 & Peter Mariën6,9,10
1 Laboratory of Neurochemistry & Behavior, Institute Born-Bunge, Dept. Biomedical
Sciences, University of Antwerp, Wilrijk, Belgium

2 Department of Experimental Psychology, Ghent University, Belgium
3 Icometrix, Leuven, Belgium
4 INSERM, Cognitive Neuroimaging Unit, Gif-sur-Yvette, France
5 Institute of Cognitive Neuroscience, East China Normal University,
Shanghai, China
6 Department of Neurology and Memory Clinic, Hospital Network Antwerp (ZNA)
Middelheim, Belgium
7 Department of Neurology and Alzheimer Research Center, University of Groningen,
University Medical Center Groningen (UMCG), The Netherlands

8 Biobank, Institute Born-Bunge, University of Antwerp, Wilrijk, Belgium
9 Flemish Academic Centre, Advanced Studies Institute of the Royal Flemish
Academy of Belgium for Sciences and the Arts, Brussels, Belgium

1 0Vrije Universiteit Brussel, Department of Clinical and Experimental
Neurolinguistics, Belgium
Introduction
A unique possibility to enrich insights in the lateralization and organisation
of language functions is to investigate individuals in whom an anomalous
organisation of brain functions is clinically evident or might be theoretically
expected, such as subjects with porencephalic cysts (PC). Naef (1958) defined
porencephaly as ‘cystic defects in the cerebrum communicating either with the
ventricle or with the subarachnoid space or with both, due to a developmental
malformation or to a destructive lesion’ (p. 136). Based on an interview, Naef
concluded that in more than 75% of the 32 cases neurological symptoms occur
perinatally. Danckert et al. (2004) for the first time reported neuropsychological
and neuroimaging findings of a left-handed man with a PC replacing a large part
of the left posterior hemisphere. Their findings suggest that intrahemispheric
reorganisation may take place in the presence of a PC and that language, motor
and somatosensory functions are not necessarily transferred to the undamaged
hemisphere.
This study investigates for the first time functional language organisation at the
supra- and infratentorial level in a neurologically and psychiatrically healthy
person identified with a PC. A Word Generation fMRI-paradigm was used to
26 DE CONINCK ET AL.
determine language lateralization at both the cerebral and cerebellar level.
Methods
Participants
After an extensive manual research of the MRI databases of two neuroradiology
departments, only one subject matched the selection criterion of “neurologically
and psychiatrically healthy individual with an isolated, congenital PC”. This subject
(CI) was a 50-year-old, right-handed (EHI = +100) woman with a large PC in the left
temporal lobe.
Neuropsychological assessment
Extensive neuropsychological examinations were performed to formally rule
out cognitive dysfunctions: intelligence (WAIS-III), memory (WMS-R), attention
(Trailmaking Test, D2 Test of Attention & Stroop Colour-Word Test) and language
(Boston Naming Test & Semantic Word Fluency Task) were investigated.
Neuroimaging
fMRI data acquisition consisted of a (1) word generation task (Cai et al. 2013)
and (2) as a control condition, the repetition of the letter string ‘dada’. A Siemens
Trio 3.0-Tesla scanner (Siemens Medical Systems) was used for anatomical (T1-
weighted) and functional (T2-weighted) imaging. Data analysis was performed by
means of SPM8 (www.fil.ion.ucl.ac.uk/spm/software/spm8).
Results
Neuropsychological assessment
An asymmetric distribution of IQ scores characterized the general cognitive profile.
A significant discrepancy of 17 IQ-points was found between the normal verbal IQ
(= 108) and superior performance IQ (= 125). Working memory (index = 134) as
well as delayed recall (index = 133) was within the superior range. A consistent and
symmetrical distribution of visual (index = 138) and verbal memory (index = 126)
indices was found. CI obtained normal to superior scores for general attentional
skills as measured by the WMS-R concentration index (= 112), visual search and
sequencing capacities (Trail Making Test), sustained visuo-motor attention (D2
Test of Attention), inhibition of a competing and more automatic response set
(Stroop Colour-Word Test), digit span and symbol substitution (WAIS-III). Visual
recognition and naming, as measured by the BNT, was normal, as were Word
fluency and the verbal subtests of the WAIS-III.
fMRI results
As demonstrated in figure 1, the fMRI language paradigm activated the
supplementary motor area, the precentral and inferior frontal gyrus, the insula
BILATERAL LANGUAGE REPRESENTATION IN A CYST PATIENT 27
and visual association cortex were bilaterally activated. Activation of the inferior
frontal gyrus was more pronounced in the left than in the right hemisphere.
Significant activation was found in the superior and inferior temporal gyrus, the
supramarginal gyrus, the putamen, the cuneus, the inferior parietal lobe and the
calcarine fissure of the left hemisphere. The superior parietal lobe, middle and
medial frontal gyri were activated in the right hemisphere.
At the cerebellar level, bilateral activation was found in the posterior lobe at the
level of the declive. The activation of the right declive is higher than the left,
which is in accordance with the more pronounced left frontal activity. Significant
activation of the left cerebellar culmen is also observed.
Figure 1: fMRI activations of CI showing bilateral cerebral and cerebellar activations during
word generation.
Numbers represent z-coordinates.
Discussion
In this patient with a large left temporal PC, neuropsychological examinations
revealed an asymmetric IQ-profile characterized by a significant discrepancy of 17
IQ points between the verbal (= 108) and performance (= 125) IQ level.
A typical pattern of fMRI language activations was not found in this patient.
Activations observed during Word Generation tasks are usually restricted to the
inferior and middle frontal gyri, the cingulate gyrus, the supplementary motor
area, the inferior parietal lobule of the language dominant hemisphere and the
contralateral posterior cerebellar lobe (Cai et al., 2013).
Due to a congenital anomaly, atypical organisation of language functions was
found as reflected by a bilateral distribution of activations at the cerebral and
cerebellar level. Although slightly more pronounced in the left hemisphere,
28 DE CONINCK ET AL.
bilateral activations were found in the middle and inferior frontal gyri. This pattern
of supratentorial activations was reflected at the infratentorial level by bilateral
activations in the posterior lobe of the cerebellum with slightly more activity
located in the right cerebellar hemisphere. Bilateral hemispheric organisation of
expressive language function is probably the result of a compensatory mechanism
for the lesion and might indicate operational inefficiency of the neural network
subserving language in the left hemisphere (Deary and Caryl, 1997). An illustration
of this inefficiency is obtained by the significant lower verbal IQ.
Lidzba et al. (2008) observed a functional language shift to the right cerebral
hemisphere in a population with congenital periventricular lesions in the
left hemisphere. Crossed cerebellar activation is observed as a result of
this supratentorial reorganisation. The bilateral cerebellar activation pattern
observed in CI confirms that the pattern of supratentorial language dominance is
intrinsically reflected at the level of the cerebellum.
References
Cai, Q., Van der Haegen, L., & Brysbaert, M. (2013). Complementary hemispheric
specialization for language production and visuospatial attention. Proceedings of the
National Academy of Sciences of the United States of America, 110, 4, E322-E330.
Danckert, J., Mirsattari, S.M., Danckert, S., Wiebe, S., Blume, W.T., Carey, D., et al. (2004).
Spared somatomotor and cognitive functions in a patient with a large porencephalic
cyst revealed by fMRI. Neuropsychologia, 42, 405-418.
Deary, I.J., & Caryl, P.G. (1997). Neuroscience and human intelligence differences. Trends in
NeuroScience, 20(8), 365-371.
Lidzba, K., Wilke, M., Staudt, M., Krägeloh-Mann, I., & Grodd, W. (2008). Reorganization of
the cerebro-cerebellar network of language production in patients with congenital
left-hemispheric brain lesions. Brain & Language, 106, 204-210.
Naef, R. (1958). Clinical features of porencephaly: a review of 32 cases. American Medical
Association Archives of Neurology and Psychiatry, 80, 2, 133-147.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 29-30
Phonological and semantic registration of the

subthalamic nucleus
Miet De Letter1,2 , Annelies Aerts1 , Sarah Vanhoutte1 , Robrecht Raedt1 ,
John Van Borsel1,3 , Dirk Van Roost4 & Patrick Santens1
2 Department of ORL & Logopaedic and Audiologic Sciences, Ghent University,
Belgium
3 Universidade Veiga de Almeida, Rio de Janeiro, Brazil
4 Department of Neurosurgery, Ghent University Hospital, Belgium
Introduction
Over the past two decades there has been an increasing awareness that subcortical
structures such as the subthalamic nucleus (STN) not only contribute to the
regulation and coordination of motor aspects of speech but are also important
components in the neural circuits that regulate cognitive and linguistic function.
Investigations of the impact of surgically induced functional inhibition of the
STN reveal a variability in language outcome and a larger influence on higher-
level language processes than on general language processes. This study aims to
investigate if the STN is involved in phonological and semantic input processing
on phoneme and word level. Local field potentials in the STN are obtained
by registration from the implanted STN electrodes of patients with Parkinson’s
Disease. Event-related potentials in the STN are used in order to measure responses
to linguistic paradigms . In Dutch normal controls these language paradigms evoke
an activation in the temporal area. Therefore, frontal cortex ERP’s are used as a
control in order to exclude non-specific electrical responses. The paradigms are
administered twice, with and without dopaminergic medication respectively.
Methods
Subjects
Until now seven patients with advanced Parkinson’s disease have been included
in this study. All patients were evaluated at one week post implantation of Deep
Brain Stimulation (DBS) electrodes. All subjects were right-handed native Dutch
speakers with normal hearing.
Method
During the first week after implantation of DBS-electrodes, patients are stimulated
using external leads. This allows connection of the electrodes to conventional
electrophysiological registration equipment in order to obtain local field potentials
from the STN. Following the internationally accepted CAPSIT-protocol the patients
30 DE LETTER ET AL.
are tested in a random order with and without dopaminergic medication. For
phonological and semantic evaluation the standardized Dutch language ERP-
paradigms developed by Aerts et al. (2013) are used. The language paradigms
consist of phonological tasks evaluating attended and unattended auditory
discrimination for phonetic contrasts, voicing and manner of articulation and of
semantic detection of action and non-action verbs.
Results
The data of all registrations are currently analyzed. Preliminary results suggest
an activation of the subthalamic nucleus when phonological as well as semantic
stimuli are presented in the condition without dopamine-administration. In
contrast to the language-induced electrical activity in the STN, no ERP is observed
in the frontal cortical area. Whether these language networks are dopamine-
sensitive will become clear when the data in the on-condition are analyzed.
Discussion
Results will be discussed using current models of basal ganglia involvement in
higher linguistic function. These can be related to known linguistic defects
occurring in PD patients and to the effects of dopaminergic treatment and DBS-
related changes in language.
References
Aerts, A., Van Mierlo, P., Hartsuiker, R.J., Hallez, H., Santens, P., & De Letter, M.
development of normative data. Brain & Language, 125, 253-263.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 31-33
Repetitions in the connected speech of a

patient with semantic dementia
Cécile De Somer1 , Dominiek Sandra1 ,
Sebastiaan Engelborghs2 & Peter Mariën2,3
1 University of Antwerp, Department of Linguistics; Centre for Computational
Linguistics and Psycholinguistics (CLiPS), Belgium

2 ZNA Middelheim Hospital, Department of Neurology, Antwerp, Belgium
3 Vrije Universiteit Brussel, Department of Linguistics, Clinical and Experimental
Neurolinguistics, Belgium
Introduction
Semantic dementia (SD) is a neurodegenerative disorder characterized by a
progressive loss of semantic knowledge in both the verbal and non-verbal domain
(Neary et al., 1998). In this condition, disproportionately affected naming and
single word comprehension, especially for low-familiarity words, constitute early
hallmark features at the linguistic level (Gorno-Tempini et al., 2011). This paper
reports longitudinal follow-up findings of connected speech in a patient with
histopathologically confirmed SD. In addition to typical SD features, a unique
profile of repetitions and perseverations was found in connected speech. These
neurolinguistic phenomena have not been reported before in SD.
Methods
Participants
Video-taped conversational speech samples of a male patient with a diagnosis of
SD were collected at the age of 55, 56 and 59 years during a face-to-face interview.
Different analyses of these samples were compared to analyses of speech samples
collected in an age and education matched control group, comprising 15 healthy
participants: five males in each of the three age groups (55, 56 and 59).
Analyses
To identify deficits in connected speech three analyses were performed. First, a
Moving Average Type-Token Ratio (MATTR) was used to determine lexical diversity
in the speech samples of the patient and the control group. Secondly, the number of
repetitions and perseverations in connected speech was counted. Thirdly, speech
rate, defined as the total number of words per minute and total number of syllables
per second was calculated.
32 DE SOMER ET AL.
Results
In the analyses the data were collapsed across the three control groups, as one-
way Analysis of Variance (ANOVA) showed no significant difference between these
groups on any measure (p > 0.10). In comparison with the control group, MATTR
results confirmed poor lexical diversity in the patient’s speech. Lexical diversity was
significantly different with the control group at the age of 55 (= 0.26; Mean = 0.39,
SD = 0.02) and 56 (= 0.28; Mean = 0.39, SD = 0.02) and extremely low at age 59
in the context of a restricted verbal output. These findings reflect the progressive
decay of semantic memory in SD. Analysis of repetitions and perseverations in
the three speech samples revealed a unique pattern at different linguistic levels
(words, constituents, parts of utterances and utterances). Self repetitions were
found in more than 50% of the patient’s utterances. At age 55 years the patient
produced 75 self repetitions out of 145 utterances (Score = 51.72; Mean = 19.76;
SD = 15.92). At the age of 56 years, 49 self repetitions out of 96 utterances were
found (Score = 51.04). The progression to verbal mutism was reflected at age 59
years by a total number of only 4 self repetitions out of 13 utterances. The number
of perseverations was abnormal as well (at age 55 years = 7.59; at 56 years = 4.17;
at 59 years = 0; Mean = 0.09; SD = 0.33) and the number of echo answers was
abnormal at the age of 56 and 59 years (at 56 years = 2.08; at 59 years = 53.85;
Mean = 0.43; SD = 0.59). Speech rate was consistent with a diagnosis of logorrhea
at 55 (209.27 words/min.) and 56 years of age (215.75 words/min.; Mean = 158.21
words/min.; SD = 24.44). At the age of 59 years speech rate became extremely low
(24 words/min.) and entirely consisted of echolalic responses. Within the next year
oral-verbal output evolved to mutism.
Discussion
In this study, a linguistic analysis of spontaneous speech production of a patient
with SD confirms some typical linguistic features of the disorder. Indeed, MATTR
results and speech rate data support the view that patients with SD produce fluent,
but empty speech with an evolution to verbal mutism. However, this study also
provides novel findings. To the best of our knowledge, perseverative linguistic
behavior, as reflected by a high proportion of repetitions and perseverations in
connected speech, has not been reported in SD before. A possible explanation
might be that these repetitions reflect a compensatory strategy. The patient repeats
himself and the interlocutor in an attempt to improve oral-verbal comprehension.
Alternatively, the verbal repetitions may result from frontal lobe dysfunction. A
more general frontal desinhibition disorder might explain why the patient is not
able to suppress his own speech and speech stimuli produced by the interlocutor.
These findings may be useful when refining the diagnostic criteria for linguistic
behavior in SD.
REPETITIONS IN THE CONNECTED SPEECH IN SEMANTIC DEMENTIA 33
References
Gorno-Tempini, M.L., Hillis, A.E., Weintraub, S., Kertesz, A., Mendez, M., Cappa, S.F., Ogar,
J.M., Rohrer, J.D., Black, S., Boeve, B.F., Manes, F., Dronkers, N.F., Vandenberghe,
R., Rascovsky, K., Patterson, K., Miller, B.L., Knopman, D.S., Hodges, J.R., Mesulam,
M.M. & Grossman, M. (2011). Classification of primary progressive aphasia and its
variants. Neurology, 76, 1006-1014.
Neary, D., Snowden, J.S., Gustafson, L., Passant, U., Stuss, D., Black, S., Freedman, M.,
Kertesz, A., Robert, P.H., Albert, M., Boone, K., Miller, B.L., Cummings, J., & Benson,
D.F. (1998). Frontotemporal lobar degeneration. A consensus on clinical diagnostic
criteria. Neurology, 51, 1546-1554.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 34-38
Atypical language dominance in a right-handed

patient: An anatomoclinical study with Direct
Electrical Stimulation (DES) and functional
Magnetic Resonance Imaging (fMRI)
Elke De Witte1 , Wim Van Hecke2 , Guido Dua3 , Maarten Moens4 ,
Didier De Surgeloose5 & Peter Mariën1,6
1 Center for Linguistics, Clinical and Experimental Neurolinguistics, Vrije

Universiteit Brussel, Belgium
2 Department of Radiology, Antwerp University Hospital, Belgium & icoMetrix,
Leuven, Belgium
3 Department of Neurosurgery, ZNA Middelheim General Hospital, Antwerp,
Belgium
4 Department of Neurosurgery, Brussels University Hospital, Belgium
5 Department of Radiology, ZNA Middelheim General Hospital, Antwerp, Belgium
6 Department of Neurology, ZNA Middelheim General Hospital, Antwerp, Belgium
Introduction
Surgical resection of brain tumours near or within the dominant Broca area
is still a controversial issue because of the high risk of postoperative linguistic
disturbances. Consequently, when resecting tumour tissue in the dominant
Broca area, the localisation and preservation of language function is of crucial
importance. Intraoperative DES during surgery allows to identify the essential
language areas and pathways in a more accurate way than non-invasive mapping
techniques such as preoperative fMRI, PET (Positron Emission Tomography) and
DTI (Diffusion Tensor Imaging) (Giussani et al., 2010; De Witte & Mariën, 2013).
As such, the use of DES has extended the indications for surgery of gliomas
located in eloquent areas that were previously considered ‘inoperable’ (Duffau,
2007). A number of recent studies showed that surgical resection around so-called
eloquent Broca’s area can be performed with good functional outcome if DES is
used (Benzagmout et al., 2007; Duffau et al., 2008; Lubrano et al., 2010; Sanai et
al., 2008). In this study unexpected neurolinguistic findings are reported in a right-
handed patient with a left prefrontal-temporal brain tumour.
Methods
Subject
The patient is a 39-year-old right-handed woman with a history of gliomas in the
left prefrontal lobe. She worked as a hairdresser for 15 years and had an educational
ATYPICAL LANGUAGE DOMINANCE IN A RIGHT-HANDED PATIENT 35
level of 12 years. A strong right hand preference was confirmed by means of the
Edinburgh Handedness Inventory (EHI) (Oldfield, 1971), which yielded a laterality
quotient (LQ) of +100. In 1999, a left frontal tumour was surgically resected under
general anaesthesia. Anatomopathological analysis of a tumour specimen showed
an astrocytoma grade II and radiotherapy was administered. In April 2007, the
patient again underwent surgery under general anaesthesia because of regrowth
of the tumoural mass anterior to the resection cavity. Chemotherapy was started.
Three years later, tumour regrowth with evolution to an anaplastic astrocytoma
grade III-IV was found affecting the lower part of the resection cavity in the
posterior frontal-temporal conjunction near the left precentral gyrus. Between
1999 and 2011 formal neurolinguistic investigations did not reveal any linguistic
abnormalities. However, given the fact that in this right-handed patient the tumour
was located in the left prefrontal region near Broca’s area it was decided resection
the tumour under awake conditions with DES.
Neuropsychological assessments
Extensive neuropsychological and neurolinguistic investigations were carried

out before and after surgery on the basis of standardised test batteries. The
neuropsychological assessments included the Mini Mental State Examination
(MMSE) (Folstein et al., 1975), the Stroop Colour-Word Test (Lezak, 1983), the
Wisconsin Card Sorting Test (WCST) (Grant & Berg, 1993), the Trail Making test
(TMT) (Lezak, 1983), the Nederlandse Leestest voor Volwassenen (NLV) (Schmand
et al., 1992), and the Weschler Memory Scale - Revised (WMS-R) (Weschler, 1984).
Language was formally investigated by means of the Akense Afasie Test (Graetz
et al., 1992), the Comprehensive Aphasia Test – NL (CAT-NL) (Visch-Brink et al.,
in press) and the Boston Naming Test-NL (Mariën et al., 1998). Intraoperatively,
naming, reading, repetition and verb generation tasks from the Dutch Linguistic
Intraoperative Protocol (DuLIP) (De Witte et al., 2013) were assessed.
Neuroimaging
Exact localisation of the tumour was determined by means of 3D T1-weighted

Magnetic Resonance Imaging (MRI) in the pre- and postoperative phase.
Functional MRI (fMRI) was performed postoperatively on a 1.5T scanner (Sonata,
Siemens, Erlangen, Germany), equipped with 40 mT/m gradients and a standard
CP head coil. The fMRI experiment was performed using a block-designed
paradigm consisting of two conditions, each lasting for 30 seconds: a resting period
and the noun-verb association task (as in Baillieux et al., 2009). Diffusion Tensor
Imaging tractography (DTI) was also performed in the postoperative phase. Axial
diffusion tensor images were obtained on a 3 Tesla (T) MR scanner (Siemens,
Erlangen, Germany) using a single-shot SE-EPI sequence.
36 DE WITTE ET AL.
Surgical procedure
An asleep-awake-asleep procedure was used in which the patient is asleep during
craniotomy and during closing of the skull but awake during language mapping
with DES (Klimek & Vincent, 2011). Cortical and subcortical stimulation were
performed.
Results
In the preoperative phase, in-depth cognitive testing revealed an executive
syndrome as reflected by defective results on the Stroop, the TMT and the WCST.
Except for defective results on the verbal fluency task of the CAT-NL extensive
examination of linguistic functions was entirely normal. No articulation problems
were noted. During the awake procedure, repeat cortical stimulation of Broca’s
area and the fronto-temporal junction did not induce any speech and language
deficits.. Subcortical stimulation was used and the tumour was removed until the
corticospinal tract was reached.
Postoperative MRI showed that a large part anterior to Broca’s area was surgically
removed. Formal linguistic assessments, however, did not reveal any language
or speech problems. Repeat neuropsychological investigations showed persistent
executive problems. Given the absence of linguistic symptoms, a postoperative
fMRI and DTI were conducted. fMRI analyses showed activations in the right
precentral gyrus, the right anterior insular gyrus at the supratentorial level. In
addition, significant activations were found in the contralateral left cerebellum. On
the DTI images the arcuate fasciculus (AF) was significantly more pronounced in
the right cerebral hemisphere than in the left cerebral hemisphere.
Discussion
We hypothesize that a favourable linguistic outcome in this patient is likely due
to atypical language lateralisation in the right cerebral hemisphere. Crossed
dominance in this case might be due to brain plasticity mechanisms or
maturational variation. A number of arguments favour the latter explanation:
1) no language problems (even transient language problems) were ever observed,
2) no positive stimulation sites were found during DES. 3) the Arcuate Fasciculus
was clearly more pronounced in the right hemisphere than in the left hemisphere
(DTI), 4) a crossed route between the right cerebrum and the left cerebellum was
detected (fMRI).
To conclude, this case shows that surgery in and around Broca’s area can be safely
conducted using awake craniotomy and intraoperative mapping. In addition, the
findings indicate that although fMRI is currently considered not to be sensitive
enough to identify the eloquent brain areas it can be an important adjunct in the
pre- and postoperative phase of patiens with atypical language lateralisation.
ATYPICAL LANGUAGE DOMINANCE IN A RIGHT-HANDED PATIENT 37
References
Baillieux, H., Vandervliet, E. J. M., Manto, M., Parizel, P. M., Deyn, P. P. D., & Mariën, P.
(2009). Developmental dyslexia and widespread activation across the cerebellar
hemispheres. Brain and Language, 108(2), 122–132. doi:10.1016/j.bandl.2008.10.001
Bastiaanse, R., Bosje, M., & Visch-Brink, E. G. (1995). PALPA: De Nederlandse versie. (PALPA:
The Dutch version). Hove: Lawrence Erlbaum.
Benzagmout, M., Gatignol, P., & Duffau, H. (2007). Resection of world health
organization grade II gliomas involving Broca’s area: Methodological and functional
considerations. Neurosurgery, 61(4), 741–752.
De Witte, E., & Mariën, P. (2013). The neurolinguistic approach to awake surgery reviewed.
Clinical Neurology and Neurosurgery, 115(2); 127-145.
De Witte, E., Satoer, D., Robert, E., Colle, H., Visch-Brink, E., Mariën, P. (2013). Essentiële
taalzones detecteren tijdens wakkere neurochirurgie. In E. Robert, E. Visch-Brink, &
A.-S. Beeckman (Eds.), Het (voor)beeldig Brein. Taal & Interventionele Geneeskunde.
Garant.
Duffau, H. (2007). Contribution of cortical and subcortical electrostimulation in brain
glioma surgery: Methodological and functional considerations. Neurophysiologie
Clinique/ Clinical Neurophysiology, 37, 373–382.
Duffau, H., Gatignol, P., Mandonnet, E., Capelle, L., & Taillandier, L. (2008). Intraoperative
subcortical stimulation mapping of language pathways in a consecutive series of
115 patients with Grade II glioma in the left dominant hemisphere. Journal of
Neurosurgery, 109(3), 461–471.
Duffau, H., Leroy, M., & Gatignol, P. (2008). Cortico-subcortical organization of language
networks in the right hemisphere: An electrostimulation study in left-handers.
Neuropsychologia, 46(14), 3197–3209.
Folstein, M., Folstein, S., & McHugh, P. (1975). Mini-Mental State: A practical method
for grading the cognitive state of patients for the clinician. Journal of Psychiatric
Research, 12, 189–198.
Giussani, C., Roux, F.-E., Ojemann, J., Sganzerla, E. P., Pirillo, D., & Papagno, C.
(2010). Is Preoperative Functional Magnetic Resonance Imaging Reliable for
Language Areas Mapping in Brain Tumor Surgery? Review of Language Functional
Magnetic Resonance Imaging and Direct Cortical Stimulation Correlation Studies.
Neurosurgery, 66(1), 113–120.
Graetz, P., De Bleser, R., & Willmes, K. (1992). De Akense Afasie Test. Lisse, The Nederlands:
Swets and Zeitlinger.
Grant, D. A., & Berg, E. A. (1993). Wisconsin Card Sorting Test. Odessa, FL, USA: Psychological
Assessment Resources.
Klimek, M., & Vincent, A. J. (2011). Awake craniotomy for brain tumor resection -
what does the anaesthesist do? Anästhesiologie, Intensivmedizin, Notfallmedizin,
Schmerztherapie: AINS, 46(6), 386–391.
Lezak, M. D. (1983). Neuropsychological Assessment. New York: Oxford University Press.
Lubrano, V., Draper, L., & Roux, F.-E. (2010). What Makes Surgical Tumor Resection Feasible
in Broca’s Area? Insights Into Intraoperative Brain Mapping. Neurosurgery, 66(5),
868–875.
Lubrano, V., Draper, L., & Roux, F.-E. (2010). What Makes Surgical Tumor Resection Feasible
38 DE WITTE ET AL.
in Broca’s Area? Insights Into Intraoperative Brain Mapping. Neurosurgery, 66(5),

868–875.
Mariën, P., Mampaey, E., Vervaet, A., Saerens, J., & De Deyn, P. P. (1998). Normative data
for the Boston Naming Test in native Dutch-speaking Belgian elderly. Brain and
Language, 65(3), 447–467. doi:10.1006/brln.1998.2000
Oldfield, R. (1971). The Assessment and Analysis of Handedness: The Edinburgh Inventory.
Neuropsychologia, 9(1), 97–113.
Sanai, N., Mirzadeh, Z., & Berger, M. S. (2008). Functional Outcome after Language Mapping
for Glioma Resection. New England Journal of Medicine, 358(1), 18–27.
Schmand, B., Lindeboom, J., & Van Harskamp, F. (1992). Nederlandse Leestest Voor
Volwassenen. Amsterdam, The Nederlands: Pearson.
Visch-Brink, E., De Smet, H. J., & Mariën, P. (in press). Comprehensive Aphasia Test-NL.
Amsterdam, The Netherlands: Pearson.
Weschler, D. (1984). Weschler Memory Scale - Revised. New York: Psychological Corporation.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 39-43
Subcortical lesions and agrammatic aphasia: A

case study in a highly inflected language
Valantis Fyndanis
University of Athens & Technological Educational Institute of Patras, Greece
Introduction
It has been argued that agrammatic aphasia derives not only from cortical but
also from subcortical lesions (Damasio et al., 1982; Ullman, 2001) involving the
basal ganglia that are connected to Broca’s area and other cortical language areas
(Caplan et al., 2007; Ullman, 2006). The basal ganglia have been suggested to be
involved in aspects of semantic integration, such as assignment of thematic roles
(Kotz et al., 2003), in linguistic/syntactic sequencing (Chan et al., in press), and
in combinatorial processes (occurring, for example, in regular verbs or nouns)
(Ullman, 2001), among other things. Lesions in subcortical structures other
than the basal ganglia may also cause language problems. Thalamic lesions, for
example, can cause sentence comprehension problems (De Witte et al., 2011),
while it has been found that the thalamus is also involved in sentence production
(Brown et al., 2006).
Against this background, I will examine the linguistic profile of a Greek-speaking
individual with a lesion predominantly in the basal ganglia. The relevant data that
will be discussed here have already been reported elsewhere (Fyndanis et al., 2006,
2010, 2012, 2013).
Methods
Participants
A Greek-speaking brain-damaged individual, GT, and a healthy control participant,
AK, participated in this study. An initial CT scan revealed a left intra-ventricular
haemorrhage (most probably from the anterior cerebral artery) for GT, which
did not affect the cortex but rather the basal ganglia (in particular, the putamen
and caudate nucleus), the thalamus and the adjacent white matter of the left
hemisphere (GT’s dominant hemisphere). He was diagnosed with agrammatic
aphasia. Information on diagnosis procedure and language testing data are
reported in Fyndanis et al. (2013).
Experiments
Participants were tested on five constrained tasks tapping (morpho)syntactic
abilities: a wh-question elicitation task, two anagram tasks (one with pictures and
one without pictures, both based on Rispens et al., 2001), a sentence completion
task, and a grammaticality judgment task. The last two tasks investigated
40 FYNDANIS
participants’ ability to produce and comprehend/judge subject-verb agreement,

tense, and aspect, while the anagram tasks explored their ability to construct
negative and affirmative sentences. The sentence completion task included both
regular and irregular verbs. Information about the design of these tasks is provided
in Fyndanis et al. (2006, 2010, 2012, 2013).
Results
GT performed significantly worse than AK in all cases except for affirmative
sentences in the anagram task I and for agreement in the sentence completion
task (see Table 1). GT was found severely impaired in the production of wh-
questions, and significantly more impaired in constructing negative sentences,
compared to affirmatives. He performed significantly worse on aspect than
Table 1: Participants’ raw and percent accuracy scores, and statistical comparisons (by
Fisher’s exact test for count data).
GT AK GT vs. AK
Wh-question elicitation task 14/36 (39%) 36/36 (100%) p = .000
Anagram task I
Negative sentences 2/16 (13%) 16/16 (100%) p = .000
Affirmative sentences 13/13 (100%) 13/13 (100%) p=1
Anagram task II
Negative sentences 1/18 (6%) 18/18 (100%) p = .000
Affirmative sentences 13/18 (72%) 18/18 (100%) p = .045
Sentence completion task
Agreement 53/56 (95%) 56/56 (100%) p = .243
Tense 24/56 (43%) 52/56 (93%) p = .000
Aspect 10/56 (18%) 49/56 (88%) p = .000
Grammaticality judgment task
Agreement 39/56 (70%) 55/56 (98%) p = .000
Tense 35/56 (63%) 50/56 (89%) p = .002
Aspect 23/56 (41%) 43/56 (77%) p = .000
Anagram I: neg. vs. affirm. p = .000 p=1 n.a.
Anagram II: neg. vs. affirm. p = .000 p=1 n.a.
Compl.: Agreement vs. Tense p = .000 p = .118 n.a.
Compl.: Agreement vs. Aspect p = .000 p = .013 n.a.
Compl.: Tense vs. Aspect p = .007 p = .527 n.a.
Judgm.: Agreement vs. Tense p = .550 p = .113 n.a.
Judgm.: Agreement vs. Aspect p = .004 p = .001 n.a.
Judgm: Tense vs. Aspect p = .037 p = .130 n.a.
SUBCORTICAL LESIONS AND AGRAMMATIC APHASIA 41
he produced agreement significantly better than tense. No significant dissociation

emerged between regular and irregular verbs in any of the three conditions of the
sentence completion task. (For a detailed presentation of the results, see Fyndanis
et al., 2006, 2010, 2012, 2013.)
Discussion
The poor performance of GT on the five tasks shows that he has severe
difficulties with (morpho)syntax. That his deficit in processing verb related
functional categories is morphosyntactic (and not morphophonological) in nature
is evidenced by the lack of dissociation between regular and irregular verbs in the
sentence completion task (see Badecker, 1997). GT’s results, thus, confirm his
previous diagnosis as agrammatic speaker, which is consistent with the view that
agrammatism may arise not only from lesions in cortical structures, but also from
subcortical lesions affecting the basal ganglia (Ullman, 2006).
GT’s data on the anagram tasks are consistent with Chan et al.’s (in press) finding
that linguistic/syntactic sequencing is subserved by the basal ganglia (among other
brain structures). (Note that in Greek, a language with a relatively flexible word
order, sequencing skills are critically involved in constructing negative sentences
rather than affirmatives.)
GT’s selective morphosyntactic deficit in comprehension/judgement could be
attributed to his difficulty judging the compatibility of adverbials (particularly,
aspectual ones) with verb forms, which seems to be a type of semantic integration
deficit. These data, therefore, support the proposal that the basal ganglia are partly
responsible for aspects of semantic integration (Kotz et al., 2003).
More general integration problems (i.e., not strictly semantic integration problems)
are possibly the source of GT’s selective difficulties in the sentence completion
task, where he performed on aspect and tense significantly worse than
on agreement. Unlike agreement, which only involves implementation of
morphosyntactic knowledge, aspect and tense require processing and integration
of information from two distinct levels of representation, morphosyntactic and
extralinguistic/conceptual (see Fyndanis et al., 2012). It seems, thus, that the basal
ganglia are generally involved in integration processes and not only in semantic
integration. Given past evidence (Brown et al., 2006; De Witte et al., 2011), a
contribution of GT’s thalamic lesion to his poor performance on the sentence
completion and grammaticality judgement tasks could not be ruled out.
Finally, one would tend to interpret the lack of dissociation between regular and
irregular verbs, in any of the three conditions of the sentence completion task,
as challenging the hypothesis that the basal ganglia are critically involved in
combinatorial processes (Ullman, 2001). In particular, on Ullman’s hypothesis, one
would likely expect GT to perform better on irregular rather than on regular verbs.
The above finding, however, does not contradict Ullman’s hypothesis, because,
although regular and irregular verbs in Greek differ in terms of ±involvement of an
42 FYNDANIS
allomorph’s retrieval, both verb types involve combinatorial/affixation processes.

This is so, because verb stems cannot stand alone in Greek. Therefore, given
affixation processes are required in both regular and irregular verbs, and following
Ullman’s assumption that such processes are subserved by the basal ganglia
system, it is expected that a lesion in this system can cause equal impairments
in producing regular and irregular verbs. Affixation processes being constant
in both verb types, we would expect only individuals with lesions confined to
temporal cortex—assumed to subserve retrieval processes (Ullman, 2001)—to
exhibit dissociation between regular and irregular verbs. Specifically, we would
expect them to perform better with regular verbs, since these do not involve
retrieval of allomorphic stems.
References
Badecker, W. (1997). Levels of morphological deficit: Indications from inflectional regularity.
Brown, S., Martinez, M.J., & Parsons, L.M. (2006). Music and language side by side in the
brain: A PET study of the generation of melodies and sentences. European Journal of
Neuroscience, 23, 2791-2803.
Caplan, D., Waters, G., Kennedy, D., Alpert, N., Makris, N., Dede, G., Michaud, J., & Reddy, A.
(2007). A study of syntactic processing in aphasia II: Neurological aspects. Brain and
Language, 101, 151-177.
Chan, S. H., Ryan, L., & Bever, T. G. (in press). Role of the striatum in language: Syntactic and
conceptual sequencing. Brain and Language.
Damasio, A., Damasio, H., Rizzo, M., Varney, N., & Gersch, F. (1982). Aphasia with
nonhemorrhagic lesions in the basal ganglia and internal capsule. Archives of
Neurology, 39, 15-20.
De Witte, L., Brouns, R., Kavadias, D., Engelborghs, S., De Deyn, P.P., & Mariën, P. (2011).
Cognitive, affective and behavioural disturbances following vascular thalamic
lesions: A review. Cortex, 47, 273-319.
Fyndanis, V., Tsapkini, K., Varlokosta, S., Petropoulou, K., & Papathanasiou, I. (2006).
Negation in agrammatism: Evidence from Greek. Brain and Language, 99, 162-163.
Fyndanis, V., Varlokosta, S., & Tsapkini, K. (2010). Exploring wh-questions in agrammatism:
Evidence from Greek. Journal of Neurolinguistics, 23, 644-662.
Fyndanis, V., Varlokosta, S., & Tsapkini, K. (2012). Agrammatic production: Interpretable
features and selective impairment in verb inflection. Lingua, 122, 1134-1147.
Fyndanis, V., Varlokosta, S., & Tsapkini, K. (2013). (Morpho)syntactic comprehension in
agrammatic aphasia: Evidence from Greek. Aphasiology, 27, 398-419.
Kotz, S. A., Frisch, S., Von Cramon, D. Y., & Friederici, A. D. (2003). Syntactic language
processing: ERP lesion data on the role of the basal ganglia. Journal of the
International Neuropsychological Society, 9, 1053-1060.
SUBCORTICAL LESIONS AND AGRAMMATIC APHASIA 43
Rispens, J., Bastiaanse, R., & van Zonneveld, R. (2001). Negation in agrammatism: A cross-
linguistic comparison. Journal of Neurolinguistics, 14, 59-83.
Ullman, M. T. (2001). A neurocognitive perspective on language: The declarative/procedural
model. Nature Reviews Neuroscience, 2, 717-726.
Ullman, M. T. (2006). Is Broca’s area part of a basal ganglia thalamocortical circuit? Cortex,
42, 480-485.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 44-47
Facilitating word retrieval in people with

aphasia: an exploration of the relationship
between language and wider
neuropsychological processing
Jennie Grassly1,2 , Caroline Newton1 & Wendy Best1
1 University College London, UK
2 Buckinghamshire Hospitals NHS Trust, UK
Introduction
The challenge of understanding word retrieval is one that has long been the
subject of investigation in aphasia therapy research, with Nickels (2002, page 4)
noting that ‘part of the approach to understanding the impact of aphasia on
an individual’s life is understanding word retrieval’. Progress has been made
in predicting an individual’s response to naming therapy but there are still
inconsistencies across studies in the relationship between language profile and
therapy outcome. Consideration has now turned to additional factors when
planning intervention, with clinicians and researchers recognizing the need to
address the relationship between language and wider cognitive processes in order
to advance ‘our understanding of the dynamic nature of language impairment in
aphasia and also for directly informing its treatment’ (Martin and Reilly, 2012,
page 254). For example, Lambon Ralph, Snell, Fillingham, Conroy and Sage
(2010) found that both cognitive factors (reasoning, problem-solving, attention,
and visual recall) and language factors (naming severity as measured by the Boston
Naming Test (Kaplan, Goodglass and Weintraub, 1983)) were the best predictors of
therapy outcome in an analysis of the assessment and therapy results from thirty-
three people with aphasia .
In addition the cognitive skill of self-monitoring, the on-line tracking of ability
(Toglia and Kirk, 2000), has come under scrutiny. For example, self-monitoring
ability has been found to be negatively related to the amount of jargon produced
by people with aphasia (Sampson and Faroqi-Shah, 2011) but when targeted in
therapy, can lead to generalized improvement in naming performance (Franklin,
Buerk and Howard, 2002).
The aims of the current study were:
• to identify a group of neuropsychological assessments, from a larger

battery, that can be administered to people with aphasia, without language
impairment confounding the interpretation of results.
• to investigate the relationship between underlying neuropsychological

processing and response to intervention as measured by facilitation.
FACILITATING WORD RETRIEVAL IN PEOPLE WITH APHASIA 45
Methods
Eight adults with aphasia, aged between 25 and 81, participated in a case series
design. A novel battery of language and neuropsychological assessments was
administered. Language assessments were primarily focused at the single word
level, examining input and output processes. Neuropsychological assessments
were standardised assessments across the domains of attention, memory and
executive function, and a non-standardised assessment of self-monitoring ability.
Facilitation studies were carried out, in which the effect on word retrieval at a later
point in time was investigated for 6 different linguistic cues.
For each of the domains, attention, memory and executive function, a number
of different assessments was selected and the relationships within and between
these assessments were analysed. It was hypothesised that assessments within
specific areas of each cognitive domain would show strong, positive correlations
if measuring the same, or overlapping, processes.
The relationship between neuropsychological assessment scores and response to
facilitation was investigated using non-parametric correlations, all 2-tailed.
Results
There were large, significant correlations identified between the
neuropsychological assessments of recognition memory rho = 0.77, n = 8, p
=0.03 and attention rho = 0.76, n = 8, p =0.03. The relationship between
assessments of executive function was non-significant, rho = -0.27, n = 8, p = 0.52.
With regards to the relationship between neuropsychological assessments and
response to facilitation (combined score across facilitation studies), there was
no significant relationship with attention (as measured by the total score on
the subtests of the Test of Everyday Attention (Robertson, Ward, Ridgeway and
Nimmo-Smith (1994)), rho = 0.22, n = 8, p = 0.61, or with executive function
(as measured by the time taken to complete Trail Making part B), rho = -0.29,
n = 8, p = 0.49.However, there was a significant correlation between response
to facilitation and both recognition memory (as measured by Wechsler Memory
Scale-III (Wechsler, 1997) subtests: Faces I, Faces II and design visual recognition),
rho = 0.76, n = 8, p = 0.03, and self-monitoring ability (as measured by self-
judgment score on a naming task), rho = 0.84, n = 8, p = 0.009.
Discussion
Neuropsychological assessments were sensitive to specific domains of cognition in
people with aphasia, with participants able to carry out assessments without being
impeded by their language impairment. This is especially true for assessments
in the domain of memory and attention. Measures of executive function were
less straightforward, due in part to its multifaceted nature; assessments of this
46 GRASSLY ET AL.
domain should therefore be considered on an individual basis with regards to the

underlying processes measured.
Recognition memory and self-monitoring ability were found to strongly correlate
with naming improvements following facilitation. The indication is that ability to
self-monitor may be an important prerequisite in order to benefit optimally from
language intervention.
The findings from the current study support the use of neuropsychological
assessments by speech and language therapy clinicians as an essential component
of a screening battery, helping inform understanding of cognitive processing and
guide selection of appropriate therapy methods. Specifically, assessments of
recognition memory and self-monitoring were found to show potential for use
within a wider battery administered by clinicians to help predict response to
therapy.
These extra-linguistic cognitive processes have great potential for further
investigation, especially as presently research has not found language skills alone
to be predictive of response to therapy or functional communication skills.
References
Best, W., Herbert, R., Hickin, J., Osborne, F., & Howard, D. (2002). Phonological and
orthographic facilitation of word-retrieval in aphasia: Immediate and delayed
effects. Aphasiology, 16(1-2), 151-168.
Franklin, S., Buerk, F., & Howard, D. (2002). Generalised improvement in speech production
for a subject with reproduction conduction aphasia. Aphasiology, 16(10/11), 1087-
1114.
Helm-Estabrooks, N. (2002). Cognition and aphasia: A discussion and a study. Journal of
Communication Disorders, 35(2), 171-186.
Howard, D., Patterson, K., Franklin, S., Orchard-Lisle, V., & Morton, J. (1985a). The facilitation
of picture naming in aphasia. Cognitive Neuropsychology, 2(1), 49-80.
Kaplan, E., Goodglass, H., & Weintraub, S. (1983). The Boston Naming Test. Philadelphia: Lee
and Febiger.
Lambon Ralph, M. A., Snell, C., Fillingham, J. K., Conroy, P., & Sage, K. (2010). Predicting the
outcome of anomia therapy for people with aphasia post CVA: Both language and
cognitive status are key predictors. Neuropsychological Rehabilitation, 20(2), 289-
305.
Luria, A. R. (1973). The Working Brain: An introduction to neuropsychology. London, UK:
Allen Lane The Penguin Press.
Martin, N., & Reilly, J. (2012). Short-term/working memory impairments in aphasia:
Data, models, and their application to aphasia rehabilitation INTRODUCTION.
Aphasiology, 26(3-4), 253-257.
Nickels, L. (2002). Therapy for naming disorders: Revisiting, revising, and reviewing.
Aphasiology, 16(10-11), 935-979.
Robertson, I. H., Ward, T., Ridgeway, V., & Nimmo-Smith, I. (1994). The Test of Everyday
Attention (TEA). Bury St Edmunds, UK: Thames Valley Test Company.
FACILITATING WORD RETRIEVAL IN PEOPLE WITH APHASIA 47
Toglia, J., & Kirk, U. (2000). Understanding awareness deficits following brain injury. Neuro
Rehabilitation, 15(1), 57-70.
Wechsler, D. (1997). Wechsler memory scale - III. San Antonio, TX: The Psychological
Corporation.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 48-51
Phonological similarity between target and

semantic errors in picture naming: Are aphasic
patients a homogeneous group? A study of 31
cases
Erminio Capitani1 , Marcella Laiacona2 , Chiara Rosci1 , Maddalena Costanzo3 ,
Rita Capasso3 , Nadia Allamano2 , Lorena Lorenzi2 & Gabriele Miceli3
1 Neurology Unit, Health Sciences Department, Milan University-S.Paolo Hospital,
Italy
2 Fondazione S.Maugeri, IRCCS, Neurology Unit, Veruno Scientific Institute, (NO),
Italy
3 Centro di Riabilitazione Neurocognitiva, CIMeC-Università di Trento, Italy
Introduction
In the aphasia literature many studies on picture naming have reported that
the phonological similarity between targets and semantic errors is higher than
expected by chance (Martin et al., 1996; Rapp et al., 2000). However, some single
case studies are not in line with this claim (Best, 1996). This invites to investigate
if patients are indeed homogeneous, and to identify clinical or linguistic variables
that influence the phonological similarity rate of semantic errors.
Methods
Participants
We report a series of 31 aphasic patients who completed a picture naming test
based on 80 stimuli from the Snodgrass and Vanderwart (1960) set that are
exemplars of 10 semantic categories (Battig and Montague, 1969). Twenty-eight
patients suffered from cerebro-vascular damage, 2 from traumatic brain injury and
1 from brain abscess. They were examined 3 times within a 1-2 day period, during
which their clinical status was stable.
Response Classification
We defined the chance level of the target/response phonological similarity for the
single stimulus. We considered incorrect responses corresponding to a real word
that was semantically related to the stimulus. We analysed only errors that were
semantic coordinates, and excluded responses consisting of the superordinate
category name (for example, we retained the response “orange”, but not the
response “fruit” to the stimulus “apple”). Operatively, we considered as a reference
the Phonological Overlap Index (POI), obtained by considering each target name
and the name of the 15 most typical exemplars of the same category reported
PHONOLOGICAL SIMILARITY BETWEEN TARGET AND SEMANTIC ERRORS 49
in Battig and Montague (1969). For each pairing between the stimulus and a
semantically-related response, the POI was the ratio between (a) the number of
phonemes shared by the two words and (b) the total number of phonemes that
appeared in the two words. A POI of zero means that the two words have no
phonological overlap at all.
Statistical Methodology
The statistical distribution of POI is typical of a given target word. On the 15
ranked POI values related to each target, we estimated the 80th centile value: the
“high overlap” region is above that value. Consequently, only 20% of the semantic
associates of each target word are expected to fall in the “high overlap” region. A
semantic error was scored as having “high POI” when its phonological overlap with
the target fell in this upper 20% region.
In data analysis, we excluded stimuli with a derivational/compounding
relationship resulting in semantic and phonological similarity, for instance
“lampada" (lamp) and “lampadina" (light bulb). For each patient the rate of
semantic errors in the high-POI region was calculated. A rate significantly higher
than 20% indicates that semantic errors show greater- than-chance phonological
similarity with the target.
Results
The data of the 31 patients were analyzed by means of a logit-linear model with
binomial error. The logistic parameter for the whole set of patients corresponded
to a binomial probability of 0.241. The lower confidence limit of this parameter
(alpha= 0.05) was 0.210: as this limit does not include 0.200, we can reject the null
hypothesis that the observed rate of high-POI semantic errors merely conforms
to the basic phonological similarity between each target word and its semantic
matches.
Figure 1 shows the distribution of high-POI semantic errors in our sample.
On logistic regression, the deviance of a model that only includes the mean was
48.318. A chi-square of 48.318 with df=30 corresponds to p=0.018, and this
indicates that a model based purely on the mean does not completely explain the
data. This could be accounted for in 3 ways: (i) the group might include a few
outliers; (ii) some variables not introduced in the model might influence results (eg,
naming deficit severity, or relative severity of phonological impairment); (iii) the
tendency to produce semantic errors with increased phonological similarity might
be in itself a randomly distributed and scarcely predictable variable.
The POI-rate distribution of our patients (Figure 1) seems smooth. The last
hypothesis (random POI-rate variation between subjects) should be considered
only after studying the logistic regression models aimed to check the relevance of
the available predictors for each patient: (a) the overall accuracy of each patient, (b)
the number of phonological errors corresponding to a real word; (c) the number
50 CAPITANI ET AL.
7
number of subjects
0
0-0,08 0,08-0,16 0,16-0,24 0,24-0,32 0,32-0,4 0,4-0,48 0,48-0,56 0,56-0,64 0,64-0,72
rate of high POI semantic errors

Figure 1: Distribution of the rate of high-POI semantic errors in our sample. In the
histogram, the height of each column represents the number of patients with a given value
(in abscissa) of the rate of high-POI semantic errors.
of phonological errors resulting in non-existing words, (d) the overall number of

phonological errors, and (e) the ratio between the overall number of phonological
errors and the overall number of naming errors. None of the considered predictors
influenced the tendency of each patient to present with a given rate of high-
POI responses, and the corresponding chi-square values always fell far short of
significance (chi-square <1, df=1).
Discussion
Our findings confirm that the increased phonological similarity of semantic errors
is not uniform in an unselected group of aphasic patients. More interestingly,
the increase of phonological similarity of semantic errors seems to depend
neither on overall severity, nor on the degree of phonological impairment. The
increased phonological similarity of semantic errors has been taken as a support
to interactive models of the functional interplay between lexical representations
and phonology (Martin et al., 1996; Rapp et al., 2000). The independence of this
effect from the level of phonological impairment seems an interesting constraint
for a more precise definition of the interaction between lexicon and phonology.
PHONOLOGICAL SIMILARITY BETWEEN TARGET AND SEMANTIC ERRORS 51
References
Best, W. (1996) When racquets are baskets but baskets are biscuits, where do the words
come from? A single case study of formal paraphasic errors in aphasia. Cognitive
Neuropsychology, 13, 443-480.
Battig, W. F., Montague, W. E. (1969). Category norms of verbal items in 56 categories:
A replication and extension of the Connecticut category norms. Journal of
Experimental Psychology, 80, 1-46.
Martin, N., Gagnon, D.A., Schwartz, F., Dell, G.S., Saffran, E. (1996) Phonological facilitation
of semantic errors in normal and aphasic speakers. Language and Cognitive
Processes, 11 , 257-282.
Rapp, B., Goldrick, M. (2000) Discreteness and interactivity in spoken word production.
Psychological Review, 107, 460-499.
Snodgrass, J.G., Vanderwart, M. (1980) A standardized set of 260 pictures: norms for
name agreement, image agreement, familiarity, and visual complexity. Journal of
Experimental Psychology: Human Learning and Memory; 6,174–215.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 52-54
Eye movements tell us more about the

underlying reading strategy in lexical readers
Irene Ablinger1 , Katja Halm1 , Walter Huber2 & Ralph Radach3
1 Neuropsychology, Department of Neurology, RWTH Aachen University Hospital,
Germany
2 Clinical and Cognitive Neurosciences, Department of Neurology, RWTH Aachen
University Hospital, Germany

3 General and Biological Psychology, University of Wuppertal, Germany
Introduction
Patients with acquired dyslexia do not show the well coordinated interaction of
lexical and segmental reading procedures known from normal skilled readers
(Radach & Kennedy, 2013). Instead, they pathologically rely predominantly on one
of the two reading routes. A psycholinguistic error analysis of dyslexic responses
in various reading tasks provides a basis for clinically discriminating subtypes
of pathological reading and for evaluating the respective recovery processes
(Coltheart, 1980; Greenwald, 2000). Phonology-related errors are indicative of
a sequential word processing strategy, whereas lexical and semantic errors are
associated with a lexical processing strategy. Despite the large number of published
intervention studies, there are only a few studies on changes in error distributions
during recovery in dyslexic patients. More specifically, to our knowledge only two
cases of a change from lexical to segmental reading have been reported in the
literature (DePartz, 1986; Nolan, Volpe, & Burton, 1997).
In the present study we assessed the impact of a specific therapy intervention on
the underlying reading strategy over the time course of recovery in patients with
acquired dyslexia. An innovative aspect of our work is not only to focus on reading
responses to evaluate the recovery process, but also to include online eye tracking
methodology (Ablinger et al., in preparation; Ablinger et al., 2012; Schattka, Radach,
& Huber, 2010) to get detailed information about the process of word identification
in real time.
Methods
Participants
We report data of five aphasic patients with acquired dyslexia. The etiology was
vascular in four patients, one suffered from a craniocerebral injury. On the basis
of oculomotor behaviour all subjects had been classified as lexical readers (see
Schattka et al, 2010, for details). In all participants oral word and pseudoword
reading was poor before therapy intervention. Errors were classified as omissions,
neologisms, phonological, lexical and semantic errors.
THE UNDERLYING READING STRATEGY IN LEXICAL READERS 53
Materials and procedure

All five subjects were included in an eye movement based reading intervention,
where lexical and segmental reading was facilitated (Ablinger et al., in preparation).
As a result of the individual baseline performance 150 individual items were
selected for therapy and diagnostic assessment. All 150 target nouns varied in
word length from 6 to 9 letters and were balanced for frequency and semantics.
Reading performance was assessed before (T1) and after (T2) therapy intervention
via recording of eye movements (EyeLink 1000, SR Research, see Inhoff & Radach,
1998, for general methodology). We developed a novel way to examine the
spatio-temporal dynamics of word processing, based on dividing the total number
of fixations per target word into three equal bins and computing the resulting
distributions of fixation positions. Comparing these sub-distributions reveals the
gradual shifting of fixation positions over the time course of word processing,
providing an adequate metric for objective classification of reading strategies.
Verbal expressions were digitally recorded for subsequent linguistic error analysis.
Results
Overt reading behaviour
While word reading accuracy improved significantly in all participants, there were
no changes in pseudoword reading. At T2 lexical and phonological errors were
dominant in two participants (DH, KM), while the other patients predominantly
exhibited one error type, lexical errors in SHJ, phonological errors in WS, and no
responses in ST.
Word fixation patterns

Therapy intervention led to a significant decrease of total reading time and total
number of fixations per word in three of five subjects (DH, ST, WS). Interestingly, in
two subjects (SHJ, KM) total reading time and total number of fixations increased
significantly from T1 to T2. Both subjects showed a relatively high proportion of no
responses before the beginning of reading intervention
Evaluation of the reading strategy

At T2 in three of five subjects (DH, SHJ, ST) the spatio-temporal analysis of fixation
positions revealed a restructuring in the underlying reading mechanisms from
predominantly lexical to more segmental word processing. More specifically,
subdistributions of fixation positions for early, medium and late time bins clearly
exhibited three spatially distant peaks indicating a transition towards sequential
word processing. In contrast, two subjects (KM, WS) maintained their lexical
reading procedures, as indicated by spatially overlapping subdistributions with
maxima near the target word centre.
54 ABLINGER ET AL.
Discussion
Therapy intervention led to improved reading accuracy in all subjects. However,
neither changes in error distribution nor standard eye movement parameters
were indicative of a possible change in the underlying reading strategy. The
assumption that predominant phonological errors may refer to a segmental word
processing routine, while the production of lexical errors is indicative of a holistic
strategy could not be verified. Instead, detailed analyses of fixation position
distributions provided detailed information about the time course of underlying
word identification processes. To conclude, our results indicate that despite
general improvements in reading performance, only some patients reorganized
their word identification process after therapy intervention. These findings raise
doubts on the validity of psycholinguistic error analysis as an indicator of changes
in reading strategy.
References
Ablinger, I., Huber, W., Schattka, K., & Radach, R. (2012). Recovery in a letter-by-letter
reader: More efficiency at the expense of normal reading strategy. Neurocase. iFirst;
(doi10.1080/13554794.2012.667119).
Ablinger, I., Weisse, K. , Vorstius, C., Halm, K., Huber, W., & Radach, R. (in preparation). Eye
movement guided reading intervention in lexical and segmental readers.
Inhoff, A. W., & Radach, R. (1998). Definition and computation of oculomotor measures in
the study of cognitive processes. In G. Underwood (Ed.), Eye guidance in reading and
scene perception (pp. 29-54). Oxford: Elsevier.
De Partz, M. (1986). Re-education of a deep dyslectic patient: Rationale of the methods and
results. Cognitive Neuropsychology, 3, 149-177.
Coltheart, M. (1980). Deep dyslexia: A review of the syndrome. In M. Coltheart, K. Patterson,
& J. Marshall, Deep dyslexia, 22-48. London: Routledge & Kegan Paul.
Greenwald, M. L. (2000). The acquired dyslexias. In S., E., Nadeau, B. A., Crosson, L., &
Gonzalez-Rothi G. (Eds.), Aphasia and language: Theory to Practice. (pp. 159-183).
NJ: The Guilford Press.
Nolan, K. A., Volpe, B. T., & Burton, L., A. (1997). The continuum of deep/surface dyslexia.
Journal of Psycholinguistic Research, 26,(4), 413-424.
Radach, R. & Kennedy, A. (2013). Eye movements in reading: Some theoretical context.
Quarterly Journal of Experimental Psychology, 66, 429-452.
Schattka, K. I., Radach, R., & Huber, W. (2010). Eye movement correlates of acquired central
dyslexia. Neuropsychologia, 48, 2959 - 2973.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 55-58
The subject-object asymmetry in aphasic

argument question comprehension:
Eye-tracking reveals the role of morphology
Sandra Hanne, Frank Burchert, Shravan Vasishth & Ria De Bleser
Linguistics Department, Centre of Excellence for Cognitive Sciences,
University of Potsdam, Germany
Introduction
Patients with agrammatic Broca’s aphasia show difficulties in producing and
comprehending wh-question structures (e.g., Grodzinsky, 2000). It is still a matter
of debate, to which extent different types of questions (e.g., referential vs. non-
referential, subject-extracted vs. object-extracted wh-questions) can be selectively
impaired in aphasia. Specifically, it is an open question whether for argument
wh-questions the same subject-object extraction asymmetry exists as has been
reported for declarative sentences. While some studies failed to show diverging
effects for these structures (Fyndanis et al., 2008; Hickok & Avrutin, 1996), others
have reported dissociations between subject-extracted wh-questions as compared
to object-extracted wh-questions (Neuhaus & Penke, 2008; Thompson et al., 1999).
The picture is even complicated by the fact that some patients in these studies
showed a reversed asymmetry with better comprehension of object-extracted as
compared to subject-extracted wh-questions (e.g., Kljajevic & Murasugi, 2010).
This pattern would not be expected by representational accounts on sentence
comprehension deficits in aphasia (e.g., the Trace-Deletion-Hypothesis / the Trace
Based Account by Grodzinsky, 1995) as these predict deficits in processing non-
canonical (i.e. object-extracted) compared to canonical (i.e. subject-extracted)
syntactic structures.
So far, studies investigating argument question comprehension have focused on
offline tasks and little is known about online processing of question structures in
aphasia.
Aim
The goal of our study is to systematically investigate offline and online processing
of subject-and object-extracted argument wh-questions in German patients with
aphasia. We ask whether different subgroups of patients exist who show
individual asymmetric or non-asymmetric question-comprehension patterns.
Most importantly, we aim to explore how eye-tracking can help to identify these
subgroups and, moreover, to identify the source of their individual impairment.
Furthermore, as studies have shown that deficits in syntactic comprehension are
not restricted to one single aphasia syndrome (Dronkers et al., 2004), we include
Broca’s as well as anomic patients in our study.
56 HANNE ET AL.
Methods
Design and Procedure
We conducted a visual-world study in which participants performed an auditory
sentence-picture matching task while their eye-movements were being measured.
The target structures comprised subject-extracted and object-extracted who-
questions (n=20 each), derived from simple semantically reversible declarative
sentences. All noun phrases (NPs) in the target sentences were masculine and,
hence, the wh-element as well as the determiner in the post-verbal NP were
unambiguously and overtly marked for case (nominative/accusative). In addition,
20 semantically irreversible what-questions served as fillers. Examples are given in
(1)-(3).
(1) Subject-extracted who-question: WerNOM kneift denACC Vater? (Who is pinching
the father?)
(2) Object-extracted who-question: WenACC kneift derNOM Vater? (Who is the father
pinching?)
(3) Filler: Irreversible what-question: Was küsst der Mann? (What is the man
kissing?)
For each trial, participants saw two pictures on a screen (a target picture and
a foil displaying a theta-role reversal) for a fixed preview time (15.000 ms) and
simultaneously the persons depicted in the pictures as well as the action were
being mentioned auditorily. This preview was followed by an asterisk shown for
600 ms to centre participants’ eye fixations. Afterwards, both pictures re-appeared
on the screen and the question was played. The task was to identify the picture
that correctly answers the question. Participants responded by button press. Each
participant received one of four different pseudo-randomized presentation lists.
Picture position as well as action direction within the pictures were balanced across
trials.
We measured accuracy and reaction times (RTs) in the sentence-picture matching
task as well as eye-movements in terms of proportions of fixations on the target and
the foil picture.
Participants
We collected data from 14 control subjects (5 male, 9 female; mean age: 62 years,
age range: 38-77 years) and, so far, 3 patients with aphasia (2 male, 1 female;
1 Broca’s, 2 amnesic; age range: 41-58 years; 9-12 years post-onset) have been
investigated. All participants were native speakers of German and (pre-morbidly)
right-handed. Patients with aphasia suffered from a unilateral lesion in their
dominant hemisphere and presented with preserved auditory analysis as well as
retained auditory single-word comprehension. In addition, comprehension of
semantically irreversible sentences and simple canonical sentences (subject-verb-
object structures) was relatively unimpaired.
SUBJECT-OBJECT ASYMMETRY IN APHASIC QUESTION COMPREHENSION 57
The experiments are still ongoing and we intend to test at least five more patients
with aphasia.
Results
Controls’ accuracy was at ceiling in both conditions (98% correct), and,
interestingly, RTs were higher for subject-extracted who-questions than for object-
extracted who-questions. This advantage of the non-canonical question type was
also evident in controls’ eye-movements, as there were significantly more fixations
to the target picture during the post-verbal noun phrase for object-questions as
compared to subject-questions.
The preliminary results from patients revealed three different patterns with
respect to accuracy: P01 performed above chance in both conditions, P02 was at
chance with object-extracted questions and above chance with subject-extracted
questions, and P03 performed at chance in both conditions. As in controls,
patients’ RTs were higher for subject-extracted than for object-extracted questions.
However, a distinct analysis of patients’ correct and incorrect responses revealed
that this was only the case for their incorrect responses, whereas for correct
responses the pattern was opposite to controls with higher RTs for object- as
compared to subject-questions.
Patients’ eye-movement data so far reveal that, contrary to controls, processing
of the wh-element and the verb alone does not provide them with sufficient
information in order to show a fixation advantage for the target picture. Instead,
patients heavily rely on the case-marking cue of the post-verbal NP and, thus, show
a delay in fixation patterns. Moreover, errors in comprehension of object-extracted
who-questions are not associated with a breakdown in non-canonical word order
processing, but rather seem to arise from deficient processing of the morphological
cue provided at the post-verbal NP.
Discussion
Overall, the results point to the existence of heterogeneous, individually distinctive
patterns in argument wh-question comprehension in aphasia. Moreover, the eye-
tracking data reveal deficits in morphological cue processing in the light of at least
partially retained syntactic processing of non-canonical word order.
References
Dronkers, N. F., Wilkins, D. P., Van Valin, R. D., Redfern, B. B., & Jaeger, J. J. (2004). Lesion
analysis of the brain areas involved in language comprehension. Cognition, 92(1-2),
145-77.
Fyndanis, V., Varlokosta, S., & Tsapkini, K. (2010). Exploring wh-questions in agrammatism:
Evidence from Greek. Journal of Neurolinguistics, 23(6), 644-662.
Grodzinsky, Y. (2000). The neurology of syntax: language use without Broca’s area.
Behavioral and Brain Sciences, 23(1), 1-71.
58 HANNE ET AL.
Grodzinsky, Y. (1995). A restrictive theory of agrammatic comprehension. Brain and

Language, 50(1), 27-51.
Hickok, G., & Avrutin, S. (1996). Comprehension of Wh-Questions in Two Broca ’ s Aphasics.
Kljajevic, V., & Murasugi, K. (2010). The role of morphology in the comprehension of wh
-dependencies in Croatian aphasic speakers. Aphasiology, 24(11), 1354-1376.
Neuhaus, E., & Penke, M. (2008). Production and comprehension of wh-questions in
German Broca’s aphasia. Journal of Neurolinguistics, 21(2), 150-176.
Thompson, C. K., Tait, M. E., Ballard, K. J., & Fix, S. C. (1999). Agrammatic aphasic subjects’
comprehension of subject and object extracted Wh questions. Brain Lang, 67(3),
169-187.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 59-63
Object naming may overestimate patient’s

language performance after neuro-oncological
surgery: A case study
Adrià Rofes1,2 , Elke De Witte3 , Peter Mariën1,3,4,5 & Roelien Bastiaanse1,6,7
1 International Doctorate in Experimental Approaches to Language And the Brain
(IDEALAB), University of Potsdam, University of Groningen, University of Trento,
Macquarie University and Newcastle University
2 Center for Mind/Brain Sciences (CIMeC) and Center for Neurocognitive
Rehabilitation (CeRIN), University of Trento, Rovereto, Italy

3 Department of Clinical and Experimental Linguistics, Vrije Universiteit Brussel,
Brussels, Belgium
4 Department of Neurology, ZNA Middelheim, Antwerp, Belgium
5 Vlaams Academich Centrum (VLAC), Advanced Studies Institute of the Royal
Flemish Academy of Belgium for Sciences and the Arts, Brussels, Belgium
Groningen, The Netherlands

7 University Medical Center Groningen (UMCG), Groningen, The Netherlands
Introduction
Before and after neuro-oncological surgery, patients are starting to be assessed with
more complex language batteries (Miceli et al., 2012; Papagno et al., 2012). Despite
that, during surgery, object naming is the only task typically used to map overall
language capacities (De Witte & Mariën, 2012). Object naming is a relevant task, as
damage in the semantic and lexical levels is known to affect language performance
and quality of life (Goodglass & Wingfield, 1997). However, object naming alone
may not be enough to assess overall language capacities as the lexical-grammatical
or syntactic level and many other semantic and lexical properties of language that
can be assessed with verbs but not with nouns are left aside (Vigliocco et al., 2011).
Given that patients that undergo neuro-oncological surgery present mild language
deficits that allow them to perform demanding tasks (Bizzi et al., 2012; Santini et al.,
2012; Satoer et al., 2011; Satoer et al., 2012; Talacchi et al., 2012), we may consider
using relevant language tools that allow mapping as much language processes as
possible. We compared the type of object naming task typically used in neuro-
oncological surgery and a new task that uses finite verbs in sentence context before
and after surgery (Rofes et al., 2012). Our predictions favor the use of tasks with
verbs.
60 ROFES ET AL.
Methods
Participant
ADB is a 41-year old, right-handed, Dutch-speaking male from West Flanders,
Belgium. He graduated in informatics and has worked as an IT project manager. In
2008, ADB underwent surgery for a glioblastoma multiforme (WHO IV) in the right
occipital lobe. In 2012, he underwent a second intervention in the right temporal
lobe, due to a recurrence of the tumor. ADB had an extensive vocabulary and
often used low-frequency words. He had some difficulties staying focused and
talked excessively. However, he was understanding and cooperative during the
assessments.
Materials
We assessed ABD using two tasks: an object naming task and a task that uses
finite verbs. The object naming task assesses the semantic and the lexical level of
language. Participants read an introductory sentence and produce the name of the
object that is in the drawing (e.g., Dit is een... paard ‘This is a... horse’). The task
with finite verbs in sentence context assesses the semantic and the lexical level as
well as the lexical-grammatical or syntactic level of language. Participants read the
subject of the sentence and produce the name of the action depicted, using the
correct inflected form (e.g., De man... fietst ’The man... bikes’).
The stimuli of both tasks consist of black-and-white line drawings with more than
80% of picture agreement, as previously measured. Each drawing is shown in a
computer screen for 4000ms with a specialized software and it is preceded by a
beep 500ms before stimulus presentation. The items are controlled and pseudo-
randomized for relevant semantic and lexical variables (e.g., frequency, age of
acquisition, imageability, length in phonemes, transitivity, number of internal
arguments).
Procedure
We administered the two tasks three times: one week before surgery (before), one
week after surgery (early after), and one month and three weeks after surgery (late
after). All responses were recorded and transcribed. Accuracy and oral reaction
times were collected for each of the items. Responses were marked as correct when
the target lexical item was produced. Responses out of time (>4000ms) and, in the
test with verbs, non-inflected forms and responses in tenses other than the third
person present were marked as incorrect.
LANGUAGE PERFORMANCE AFTER NEURO-ONCOLOGICAL SURGERY 61
Results
ADB responded to the object naming task 90% correct before surgery, 80% early
after and 97% correct after surgery. To the task with finite verbs, he responded 87%
correct before surgery, 68% correct early after and 89% correct after surgery (Figure
1).
Figure 1: Percentage correct over time (Finite verbs and Object naming)
We compared the performance in each test across the three testing moments
(‘before’ vs. ‘early after’ vs. ‘late after’). A Cochran’s Q test revealed significant
differences in object naming (Q(2)=29.200;df=2; p=0.000) and in the task with
finite verbs (Q(2)=26.000;df=2; p=0.000). For the object naming task, post-
hoc McNemar tests with Bonferroni corrections (α < 0.0167) revealed significant
differences between the three assessments (‘before’ vs. ‘early after’: Q(1)= 0.008;
‘early after’ vs. ‘late after’: Q(1)= 0.000; ‘before’ vs. ‘late after’: Q(1)= 0.003). For the
task with finite verbs post-hoc analyses revealed significant differences between
the assessments ‘before’ and ‘early after’ surgery (Q(1)= 0.000) and between ‘early
after’ and ‘late after’ surgery (Q(1)= 0.000). However, for the test with finite verbs,
we did not find significant differences between the assessments ‘before’ and ‘late
after’ surgery (Q(1)= 0.754).
We compared the two tests in each assessment moment (object naming ‘before ‘
surgery versus task with verbs ‘before ‘ surgery; object naming ‘early after ‘ surgery
vs. task with verbs ‘early after ‘ surgery; and object naming ‘late after ‘ surgery
vs. task with verbs ‘late after ‘ surgery). Mann Whitney U tests with Bonferroni
corrections (α < 0.0167) did not reveal significant differences ‘before’ surgery (Z=
-0.735; p= 0.462) and ‘early after’ (Z= -2.148; p= 0.032). However, they revealed
significant differences between tests ‘late after’ surgery (Z= -2.713; p= 0.007).
62 ROFES ET AL.
Discussion
Among other things, ADB was overall better in object naming as compared with
the task with finite verbs. This is expected given that the task with verbs not
only assesses the semantic and the lexical level but also the lexical-grammatical
or syntactic level of language. The differences may also accrue due to the
natural variance in the semantic features of nouns and verbs. Interestingly, ADB
significantly improved in object naming after surgery although he remained the
same in the test with finite verbs.
This raises the point that if we were to take object naming as the only outcome
of ADB’s language capacities, we would be overestimating his performance. We
interpret these results as an indication that the language assessment of patients
in neuro-oncological surgery should be revisited and that relevant language tasks
such as the one we used with finite verbs may be a relevant complement.
References
Bizzi, A., Nava, S., Ferrè, F., Castelli, G., Aquino, D., Ciaraffa, F., ... & Piacentini, S. (2012).
Aphasia induced by gliomas growing in the ventrolateral frontal region: assessment
with diffusion MR tractography, functional MR imaging and neuropsychology.
Cortex, 48, 255-272. doi:10.1016/j.cortex.2011.11.015
De Witte, E., & Mariën, P. (2012). The neurolinguistic approach to awake
surgery reviewed. Clinical Neurology and Neurosurgery, 2, 127-45. doi:
10.1016/j.clineuro.2012.09.015115
Goodglass, H., & Wingfield, A. (1997). Word-finding deficits in aphasia: brain-behavior
relations and clinical symptomatology. In H. Goodglass, & A. Wingfield (Eds.),
Anomia: Neuroanatomical and Cognitive Correlates (pp. 3-27). San Diego, CA:
Academic Press.
Miceli, G., Capasso, R., Monti, A., Santini, B., & Talacchi, A. (2012). Language testing in brain
tumor patients. Journal of Neurooncology, 108, 247-252. doi: 10.1007/s11060-012-
0810-y
Papagno, C., Casarotti, A., Comi, A., Galluci, M., Riva, M., & Bello, L. (2012). Measuring
clinical outcomes in neuro-oncology. A battery to evaluate low-grade gliomas (LGG).
Journal of Neurooncology, 108, 269-275. doi: 10.1007/s11060-012-0824-5
Rofes, A., De Witte, E., Mariën, P., & Bastiaanse, R. (2012). The Verb in Sentence Context Test
(VISC). Rijksuniversiteit Groningen, Groningen, The Netherlands.
Santini, B., Talacchi, A., Squintani, G., Casagrande, F., Capasso, R., Miceli, G. (2012).
Cognitive outcome after awake surgery for tumors in language areas. Journal of
Neurooncology, 108(2), 319-326.doi:10.1007/s11060-012-0817-4
Satoer, D, Vincent A., Dirven, C., Smits, M., & Visch-Brink, E. (2011). Spontaneous speech in
patients with presumed low-grade gliomas in eloquent areas before and after awake
craniotomy. Neuro-oncology, 13, 121-126. doi:10.1093/neuonc/nor159
Satoer, D,Work.J, Visch-Brink, E., Smits, M., Dirven, C., & Vincent, A. (2012).Cognitive
functioning early after surgery of gliomas in eloquent areas. Journal of Neurosurgery,
117(5), 831-838. doi: 10.3171/2012.7.JNS12263
LANGUAGE PERFORMANCE AFTER NEURO-ONCOLOGICAL SURGERY 63
Talacchi, A., Avella, D., Denaro, L., Santini, B., Meneghelli, P., Savazzi, S., & Gerosa, M. (2012).
Cognitive outcome as part and parcel of clinical outcome in brain tumor surgery.
Journal of Neurooncology, 108(2), 327-332.doi:10.1007/s11060-012-0818-3
Vigliocco, G., Vinson, D., Druks, J., Barber, H., & Cappa, S. (2011). Nouns and verbs
in the brain: A review of behavioural, electrophysiological, neuropsychological
and imaging studies. Neuroscience and Biobehavioral Reviews, 35, 407-426. doi:
10.1016/j.neurobiorev.2010-04.007
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 64-66
Proper and common noun learning: Same or

different?
Anastasiia Romanova1,2,3 , Lyndsey Nickels1,2 ,
Kati Renvall1,2 & David Howard3
1 ARC Centre of Excellence in Cognition and its Disorders, Department of Cognitive
Science, Macquarie University, Sydney, Australia

2 NHMRC Centre for Clinical Research Excellence (CCRE) in Aphasia Rehabilitation
3 School of Education, Communication and Language Sciences, Newcastle
University, Newcastle-Upon-Tyne, UK
Introduction
Proper (e.g. John) and common (e.g. person) names form two different categories
of nouns with distinct features that characterise them. The data from studies
on normal healthy population, elderly people and speakers with brain damage
show that retrieval and learning of proper and common nouns might not evolve
in similar ways. Convergent data from diary and laboratory studies on word
retrieval as well as learning for young adults, elderly speakers and people with
aphasia demonstrate that proper nouns are more vulnerable to tip-of-the tongue
phenomena (Young, Hay, & Ellis, 1985; Seamon & Travis, 1993) and impairment in
brain-damaged speakers (Semenza & Zettin, 1988; Harris & Kay, 1995). In learning
(for both young and elderly healthy participants), subsequent recall of a learnt
proper name(e.g. a personal name) is often found to be poorer than recall of a
learnt common name (e.g. occupation) when associative learning techniques are
used (Stanhope & Cohen, 1993).
There are two hypotheses regarding the underlying cause of differences between
proper and common noun learning and processing. On the one hand, such
divergence might be due to different logical properties, i.e. nature of proper and
common names as word classes. Common nouns refer to a category of beings
or objects that share certain semantic properties, while proper names designate
specific individual beings or objects with their unique features. Proper nouns
are also sometimes called ‘pure referring expressions’ (e.g. Kripke, 1980) as they
refer us to an individual entity but do not convey any semantic features, any
meaning. On the other hand, proper and common nouns can be considered to
be on a ’continuum of word retrieval difficulty’ as suggested by Kay, Hanley, and
Miles (2001) on the basis of a number of statistical properties such as frequency,
familiarity, and age of acquisition.
We had two aims:
1. Theoretical : By using novel stimuli the statistical properties of the to-be-
learnt proper and common nouns are equal. We aimed to test the hypothesis
that logical properties underpin differences in proper and common noun
retrieval and learning.
PROPER AND COMMON NOUN LEARNING 65
2. Clinical: To determine whether people with aphasia show disproportionate

impairments in learning proper nouns compared to common nouns relative
to age-matched unimpaired speakers.
Methods
Participants
Three experimental groups are to participate in the study: young subjects with no
history of brain damage (n=16), elderly subjects with no history of brain damage
(n=10) and elderly subjects with mild-moderate aphasia as a result of a stroke
(n=5). Thus far we have tested the young adult group.
Stimuli and Design

Twenty bisyllabic phonologically plausible non-words, 5-8 letters in length (e.g.
cheskel) were created and matched to pictures of imaginary creatures. All the
items were divided into two sets of 10 items each. Across the two sets verbal
items were matched for length, initial phonemes/cluster of phonemes, written
bigram frequency, phonological and orthographic neighbourhood and phonotactic
probability. Each set was presented in either the common or proper noun
condition. Items were shown as representatives of a number of species (10 in
total) in the common noun condition and as individual creatures (10 in total) in
the proper noun condition.
Following the procedure used in Tuomiranta et al. (2011) our experiment was
conducted within eight sessions: four learning sessions and four follow-up sessions
one day, one, four and eight weeks after the last learning session. During learning
sessions participants were presented with the full set of items twice. In the common
noun condition participants were given three representatives of each species,
whereas in the proper noun condition they saw three depictions of each individual.
Training was followed by naming and word-picture verification tasks (feedback was
given to enhance learning).
During the naming task participants were asked to recall the name of a
species/individual name of a creature shown. Irrespective of whether naming was
successful or not, participants were given a phonological cue (the first phoneme
and schwa). Participants saw and heard the correct response after presentation of
each item.
During the word-picture verification task participants were asked to say if a
presented word was the right label for a presented picture. This task was designed
among all to verify if participants learnt items as members of respective noun
classes rather than pure labels.
Follow-up sessions started with word (written and auditory) and picture
recognition tasks which were followed by naming and verification tasks (no
feedback was given).
66 ROMANOVA ET AL.
Results
The preliminary analysis of the naming task performed by young adults showed no
significant difference in the rate and extension of learning of common and proper
names (F (1)=2.65, p >.05). Full data on the naming and word-picture verification
tasks for all three participant groups will be reported at the conference. Further
analysis is to demonstrate whether elderly participants show similar learning
patterns to young adults: if learning and retention of common and proper nouns
are impaired to the same extent, this will serve as another evidence for them
being processed similarly. Data from people with aphasia is to reveal whether
any disproportionate impairments in learning and retrieval of proper and common
nouns are to be found provided statistical properties of the items from both word
classes are equal.
Discussion
According to the preliminary analysis of the young adult data, there is no difference
in the rate of learning and retention between common and proper nouns. Such
results are in favour of the account that statistical properties have a dominating role
in learning and retrieval of common and proper names. Thus, the fact that proper
names are more prone to failure or delay in retrieval and learning are most likely
due to the fact that common nouns are more frequent and familiar, and acquired
earlier in life in comparison to proper nouns. The data from people with aphasia
and their age-matched controls will be discussed at the conference.
References
Harris, D., & Kay, J. (1995). I know your face, but I can’t remember your name. Is it because
names are unique? British Journal of Psychology, 86, 345-358.
Kay J, Hanley, J.R., & Miles, R. (2001). Exploring the relationship between proper name
anomia and spoken word retrieval: A single case study. Cortex, 37, 501-517.
Kripke, S. (1980). Naming and Necessity. Oxford: Basil Blackwell.
Seamon, J.G. & Travis, Q.B. (1993). An ecological study of professors’ memory for student
names and faces: a replication and extension. Memory, 1, 191-202.
Semenza, C., & Zettin, M. (1988). Generating proper names: a case of selective inability.
Cognitive Neuropsychology, 5, 711-721.
Stanhope, N., & Cohen, G. (1993). Retrieval of proper names: Testing the models. British
Journal of Psychology, 84, 51-65.
Tuomiranta, L., Groenholm-Nyman, P., Kohen, F., Rautakoski, P., Laine, M., & Martin, N.
(2011). Learning and maintaining new vocabulary in persons with aphasia: Two
controlled case studies. Aphasiology, 25 (9), 1030-1052.
Young, A.W., Hay, D.C., & Ellis, A.W. (1985). The faces that launched a thousand slips:
Everyday difficulties and errors in recognizing people. British Journal of Psychology,
76, 495-523.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 67-71
Broca meets Wernicke in one single case

Dorien Vandenborre1,2 & Peter Mariën1,3
1 Department of Clinical and Experimental Neurolinguistics,
Vrije Universiteit Brussel, Belgium
2 Cepos, Rehabilitation Centre, Duffel, Belgium
3 Department of Neurology, ZNA Middelheim Hospital, Antwerp, Belgium
Introduction
Detailed clinical descriptions of language disorders date back to the mid 1800s
- early 1900s, when clinicians started to identify lesion-aphasia correlates. Early
investigators also established that written language symptoms in aphasia typically
mimic the type of distortions found at the oral-verbal level. Indeed, as a rule the
nonfluent aphasias are characterized by expressive language disturbances at both
the oral and written language level. Fluent aphasia symptoms in expressive and
comprehensive speech are present in reading and writing as well.
Methods
We present the neurolinguistic follow-up findings in an exceptional, right-
handed patient who following traumatic brain injury to the left temporal and
left frontal brain regions presented with a marked dissociation between oral and
written language. In-depth neurolinguistic and neurocognitive assessments were
performed at 5, 15 and 27 weeks posttrauma.
Results
Five weeks after the trauma in-depth neurolinguistic investigations identified
a marked discrepancy between oral and written language. A neologistic
paragrammatic output at the oral level sharply contrasted with a superior and
relatively spared agrammatic written output. As demonstrated by the test results
(table 1) oral and written language comprehension as well as speech production
significantly improved after 15 weeks and the dissociation resolved. However, the
correct use of function words remained problematic and difficulties with tense
agreement persisted. Phonological jargon resolved but phonological errors still
intruded speech and writing. Repeat investigations at 27 weeks posttrauma showed
a further improvement of auditory comprehension skills.
Neuropsychological investigations performed five weeks after the trauma showed
attention disturbances, memory problems and executive dysfunctions. Repeat
examinations disclosed a significant improvement at 15 and 27 weeks postonset
but attentional deficits persisted: tempo remained reduced and attention span
was variable. Due to aphasia verbal memory problems were found. Executive
68 VANDENBORRE & MARIËN
functioning improved but planning and organisation skills remained distorted. The
patient solved problems in a highly atypical, inefficient and often complex way.
Table 1: Neurolinguistic Data (AAT and CAT-NL) 5, 15 and 27 weeks posttrauma
Results Percentile
Max Mean ± 1SD
5w/15w/27w (Cut-off )
AACHEN APHASIA TEST
Comprehension 87/104/115 55/86/99 120 108.5 10.24
auditory: words 25/29/30 77/95/98 30 26.49 3.30
auditory: sentences 15/24/28 20/69/91 30 26.79 3.41
Total 40/53/58 60 53.28 6.08
written: words 26/26/29 67/67/90 30 28.30 2.29
written: sentences 21/25/28 63/82/95 30 26.91 3.39
Total 47/51/57 60 55.21 4.90
Token Test (errors) 50/16/13 6/75/81 50 2.28 2.75
Spontaneous Speech
Communicative behaviour 2/3/4 - 5 4.63 0.54
Articulation and prosody 2/3/4 - 5 4.63 0.67
Automatisms 2/3/5 - 5 4.59 0.65
Semantic structure 2/3/5 - 5 4.59 0.53
Phonematic structure 2/2/3 - 5 4.54 0.56
Syntactic structure 3/3/4 - 5 4.41 0.55
Imposed Speech
Total repetition 2/112/128 2/55/73 150 144.1 8.07
Phonemes 0/30/30 1/88/88 30 28.91 2.09
Monosyllabic words 2/28/29 5/68/80 30 29.22 1.32
Loan- &foreign words 0/24/27 3/53/71 30 28.94 2.31
Compounds 0/12/20 5/34/57 30 28.45 2.22
Sentences 0/18/22 5/59/70 30 28.55 1.90
Total naming 12/109/113 15/94/98 120 109.3 8.42
Simple nouns 1/27/30 13/74/97 30 27.92 2.90
Colour names 6/29/30 18/93/98 30 27.69 1.99
Composed nouns 0/28/28 8/91/91 30 28.04 2.61
Sentences 5/25/27 27/91/97 30 25.69 3.72
Written Language 23/82/87 26/87/95 90 85.52 7.63
Reading aloud 10/28/30 22/79/96 30 28.95 1.93
Composing 9/28/29 32/89/93 30 28.57 2.75
Dictational writing 4/26/28 31/83/91 30 28 3.67
COMPREHENISIVE APHASIA TEST-NL

Comprehension 100/122/124 128
Auditory: words 28/ 30/ 30 30
Auditory: sentences 11/27/30 32
Auditory: paragraphs 2/4/4 4
Total 41/61/64 66
Written: words 28/30/30 30
Written: sentences 31/32/30 32
Total 59/62/60 62
Spontaneous Speech
Spoken picture description
Written picture description
BROCA MEETS WERNICKE IN ONE SINGLE CASE 69
Table 1: continued
Results Percentile
Max Mean ± 1SD
5w/15w/27w (Cut-off )
Imposed Speech
Total repetition 0/34/56 74
Words 0/16/26 32
Complex words 0/0/6 6
Nonwords 0/8/8 10
Digit strings 0/4/8 14
Sentences 0/6/8 12
Total naming 0/39/54 58
Objects 0/30/44 48
Actions 0/9/10 10
Written Language
Total reading 2/56/62 70
Words 0/34/40 48
Complex words 0/6/6 6
Function words 2/6/6 6
Nonwords 0/10/10 10
Total writing 49/78/78 82
Copying 31/31/31 31
Picture names 17/23/23 23
Dictational writing 1/24/24 28
Legend: w = weeks; - = data not available; * = significant (z = >2)
Discussion
To the best of our knowledge only a handful of cases have been reported in which
the aphasia profile is characterized by a marked typological contrast between oral
and written language (Lhermitte & Derousné, 1974; Hier & Mohr, 1977; Basso,
Taborelli & Vignolo, 1978; Assal, Buttet & Jolivet, 1981; Ellis, Miller & Sin, 1983;
Saffran, Coslet & DeSalme, 2000; Kemmerer, Tranel & Manzel, 2005). In this
patient oral output contained all typical characteristics of Wernicke aphasia, i.e.
neologistic jargon and severely impaired auditory comprehension. Written output
on the other hand consisted of all the typical features of Broca’s aphasia, i.e.
impaired, though comprehensible written language production with relative intact
written comprehension. Several interesting dissociations characterise the patient’s
profile. The most striking dissociation relates to non-fluent and relatively correct
written output and severely disrupted fluent jargonic speech. In addition writing
was characterised by agrammatic symptoms while paragrammatic disturbances
affected speech. Phonological jargon disrupted speech but at the written level
phonological processing was correct. This pattern of linguistic deficits did not allow
classification in terms of a neurological aphasia taxonomy.
A significant improvement occurred during follow-up at 15 weeks posttrauma and
the dissociations resolved: the patient’s oral output became intelligible. In writing
as well as in speech more morphological variability occurred and syntactically
70 VANDENBORRE & MARIËN
more complex structures were used (with many clauses but with an overuse of
auxiliary verbs). Problems with the correct choice of function words and tense
agreement persisted. Phonological jargon resolved but phonological paraphasias
remained in speech. Phonological paragraphias were observed in writing.
Following remission of the oral-written language dissociations, the patient’s
aphasic symptom complex corresponded to a diagnosis of conduction aphasia
(CA): a fluent aphasia syndrome, characterized by a deficit at the phonological
processing level. Spontaneous speech was marked by frequent phonemic
paraphasias, conduite d’approche, word-finding difficulties, and paraphrasing.
Spontaneous writing was characterized by more complex grammatical structures
and phonological paragraphias were observed as well.
Anatomoclinical studies have causally related CA to damage of the left superior
temporal gyrus (STG) and the left supramarginal gyrus, which play a crucial role
in phonological processing (Graves et al., 2008). The left STG is crucially implicated
in the modulation of phonological complexity such as word length and frequency
rather than regularity (Tomasina et al., 2011). Phonemic paraphasias are the
result of a disruption of the sensory-motor integration circuit which leads to an
impairment in the capacity for auditory representations of speech to constrain and
guide the corresponding articulatory representations thought to be stored in the
inferior frontal gyrus (IFG) and ventral premotor cortex (Hickok & Poeppel, 2004).
The morphosyntactical errors of this patient may be linked to the small traumatic
lesion in the left frontal lobe. The IFC plays a role in grammatical processing,
moreover in allowing morpho-phonological properties of the verb (Tomasina et al.,
2011).
The exceptional typological contrast between nonfluent written and fluent
oral language in this patient implicates the use of different anatomical areas
for different aspects of language processing. For example his phonological
intact writing might be able to a preserved medial occipital gyrus whereas his
phonological jargon might be due to a structural damage to the left IFG, STG
and the primary auditory cortex (Booth et al., 2002). Significant improvement
in language processing is modified by functional repair of the right temporal
area that is responsible for sustained attention, motor planning and response
inhibition (Hagoort, 2005). The contrast underlies the over-simplification of
the Geschwind-Wernicke model where CA was interpreted as the expression of
the inability of temporal regions to monitor Broca’s area speech output through
subinsular connections resulting in identical aphasia (Geschwind, 1965). Moreover
the contrast founds the hypotheses of Martino et al. (2013) and Catani & Mesulam
(2008) who describe the tripartite role of the subcortical superior longitudinal
fasciculus, which in addition to the deep and long direct segment, namely the
arcuate fasciculus (FA), consists of two superficial short tracts, an anterior segment
linking Broca’s area with the inferior parietal lobe and a posterior segment linking
the inferior parietal lobe with Wernicke’s area.
BROCA MEETS WERNICKE IN ONE SINGLE CASE 71
References
Assal, G., Buttet, J., & Jolivet, R. (1981). Dissociation in aphasia: a case report. Brain and
Language, 13 (2), 223-240.
Basso, A., Taborelli, A., & Vignolo, L.A. (1978). Dissociated disorders of speaking and writing
in aphasia. Journal of Neurology, Neurosurgery and Psychiatry, 41, 556-563.
Booth, J.R., Burman, D.D., Meyer, J.R., Gitelman, D.R., Parrish, T.B., & Mesulam M.M. (2002).
Modality independence of word comprehension. Human Brain Mapping, 16, 251-
261.
Catani, M., & Mesulam, M. (2008). The arcuate fasciculus and the disconnection theme in
language and aphasia: history and current state. Cortex, 44 (8), 1105-1132.
Ellis, A.W., Miller, D., & Sin, G. (1983). Wernicke’s aphasia and normal language processing: a
case study in cognitive neuropsychology. Cognition, 15, 111-144.
Geschwind, N. (1965). Disconnexion syndromes in animals and man. Brain, 88, 237-394.
Graves, W.W., Grabowski, T.J., Mehta, S., & Gupta, P. (2008). The left posterior superior
temporal gyrus participates specifically in accessing lexical phonology. Journal of
Cognitive Neuroscience, 20, 1698-1710.
Hagoort, P. (2005). On Broca, brain, and binding: a new framework. Trends in Cognitive
Science, 9, 416-423.
Hickok G., & Poeppel, D. (2004). Dorsal and ventral streams: a framework for understanding
aspects of the functional anatomy of language. Cognition, 92, 67-99.
Hier, D.B. & Mohr, J.P. (1977). Incongruous oral and written naming. Brain and Language, 3,
115-126.
Kemmerer, D., Tranel, D., & Manzel, K. (2005). An exaggerated effect for proper nouns in a
case of superior written over spoken word production. Cognitive Neurology, 22, 2-27.
Lhermitte, F., & Derousesné, J. (1974). Paraphrasies et jargonaphasie dans le langage oral
avec conservation du langage écrit. Revue Neurologique, 130, 21-38.
Martino, J. De Witt Hamer, P., Berger, M., Lawton, M., Arnold, C., de Lucas, E., Duffau, H.
(2013). Analysis of the subcomponents and cortical terminations of the perisylvian
superior longitudinal fasciculus: a fiber dissection and DTI tractography study. Brain
Structure and Function, 218, 105-121.
Saffran, M., Coslet, H.B., & DeSalme, E. (2000). How is your B-A-B-Y? Dissociated Oral and
Written production. Neurocase, 6, 193-204.
Tomasina, M.O., Liming, K.T., Ng, P., Berne, Y.I., & Graham, S. (2011). The past tense debate:
Is phonological complexity the key to the puzzle, NeuroImage, 57, 271-280.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 72-75
The WORD project: a case series study on

intervention for developmental word-finding
difficulties
Lucy Hughes1 , Jackie Masterson2 , Michael Thomas3 , Anna Fedor3 ,
Silvia Roncoli2 , Anna Kapikian2 , Liory Fern-Pollack2 & Wendy Best1
1 University College London, UK
2 Institute of Education, UK
3 Birkbeck College, UK
Introduction
Children with developmental word-finding difficulties (WFD) can experience
problems in retrieving words, which can usefully be compared and contrasted to
those of adults with anomia as part of acquired aphasia. WFD can influence a
young person’s relationships, self-esteem and education.
There is controversy in the literature on both the source of WFD (e.g. Constable, et
al., 1997; Messer & Dockrell, 2006) and the most effective intervention (e.g. Ebbels
et al., 2012). The WORD (WOrd Retrieval and Development) project investigates
the development of lexical retrieval in children and explores the outcomes of
contrasting treatment strategies for those who have WFD.
The ongoing study has three strands. The first involves collecting data from
children with typically developing (TD) language and a clinical WFD group.
Naming speed and accuracy are compared for children at different stages of
development, alongside their performance on related tasks tapping semantic and
phonological processing.
The second strand entails computational modelling of the processes involved in
word retrieval. Constraints within the model can be varied to reflect the patterns
shown by children with TD language and with WFD.
The final strand is an experimentally controlled intervention in which a sub-set of
the children with WFD take part in one form of therapy most appropriate to their
difficulty and one building on their relative strengths to determine which approach
is more effective. Therapy protocols were devised taking account of techniques
used successfully with adults with anomia as part of their aphasia (semantic and
phonological feature analysis, e.g. Coehlo et al, 2000; Leonard et al., 2008; Boyle
& Coelho, 1995). The results of each intervention are explored in relation to the
primary outcome measure: picture naming. We also consider participants’ own
views of change.
THE WORD PROJECT 73
Methods
Primary data collection
In the first stage of our project, we aim to better understand word-finding
difficulties by placing them in the context of typical development. Thus far we have
assessed 66 children with TD language (4;0 - 8;4 years) and 9 children with WFD (6;0
- 8;7 years).
Assessments include:
• Naming (accuracy, RT, errors)
• Word-picture verification
• Semantics (picture judgement)
• Non-word repetition (CN Rep)
Additional assessments for the clinical population include:
• Test of Word Finding in Discourse
• Clinical Evaluation of Language Fundamentals (R) - 4
Data collected from TD children are used to construct a trajectory of normal

development for naming (Thomas et al., 2009). We compare these with the profiles
of WFD participants to establish individual differences and try to identify clinical
sub-types, which can be compared with those of adults with aphasia, e.g. children
who are primarily impaired on measures of semantic processing versus those with
primarily phonological difficulties.
Intervention
Children from the WFD group are invited to take part in the intervention stage,
in which we contrast different approaches using a tightly-controlled case series
design. Each child participates in 2 treatment conditions and acts as his/her own
control. They receive a ‘model appropriate’ therapy, e.g. semantic elaboration
for a child whose difficulty appears to be rooted in word meanings, and a ‘model
inappropriate’ intervention, e.g. the same child also receives phonological therapy,
targeting an area of relative strength.
Results
The results from the first strand suggest the following:
a) Children, like adults with aphasia, can exhibit specific difficulties in retrieving
words that are in their vocabulary as assessed by word-picture verification. This is
illustrated in figure 1, below.
74 HUGHES ET AL.
b) Different profiles are emerging that can be related to those found in adults with
anomia as part of their aphasia
Intervention:
Children’s naming can be improved by the intervention. The effects are largely
limited to treated items. Importantly, the first children through the study show
differential effects for the semantic and phonological intervention.
Figure 1: A comparison between word comprehension (as measured by word-picture

verification) and word finding (as measured by picture naming) for children with TD
language and with WFD.
Discussion
It has been established that both phonological components analysis (PCA) and
semantic features analysis (SFA) can improve adults’ naming and generalize (Van
Hees et al. 2013). However, the relationship between the level of deficit and
outcomes of intervention is not straightforward (Lorenz & Ziegler, 2009).
It is important for research on adults with aphasia to inform studies of children with
THE WORD PROJECT 75
language needs and vice versa. Initial findings from our ongoing WORD project
help deepen our understanding of word-finding difficulties by placing them in the
context of typical development.
Results from our assessment phase suggest word-picture verification is a sensitive
measure for word comprehension, which may be considered as an alternative to
word-picture matching tasks, traditionally used for adults with aphasia.
Meanwhile, preliminary outcomes from the intervention strand suggest that
targeting areas of relative weakness may be the optimal approach for improving
word-finding.
WFD - whether developmental or acquired - are not a unitary phenomenon
and can have different sources. This highlights the importance of comparing
interventions within individuals, as well as across groups.
References
Boyle, M., Coehlo, C. (1995). Application of semantic feature analysis as a treatment for
aphasic dysnomia. American Journal of Speech-Language Pathology, 4, 94-98.
Coehlo, C., McHugh, R. E. & Boyle, M. (2000). Semantic feature analysis as a treatment for
aphasic dysnomia: A replication. Aphasiology, 14 (2), 133-142.
Constable, A., Stackhouse, J., & Wells, B. (1997). Developmental word finding difficulties
and phonological processing: the case of the missing handcuffs. Applied
Psycholinguistics, 18, 507-536.
Ebbels, S. H., Nicoll, H., Clark, B., Eachus, B., Gallagher, A. L., Horniman, K. , Jennings,
M., McEvoy, K., Nimmo, L. & Turner, G. (2012). Effectiveness of semantic therapy
for word-finding difficulties in pupils with persistent language impairments: a
randomized control trial. International Journal of Language & Communication
Disorders, 47 (1), 35-51.
Leonard, C., Rochon, E. & Laird, L. (2008). Treating naming impairments in aphasia:
Findings from a phonological components analysis treatment. Aphasiology, 22 (9),
923-947.
Lorenz, A., Ziegler, W. (2009). Semantics vs. word-form specific techniques in anomia
treatment: A multiple single-case study. Journal of Neurolinguistics, 22 (6), 515-537.
Messer, D. & Dockrell, J. (2006) Children’s word-finding difficulties: Descriptions, and
explanations. Journal of Speech, Language and Hearing Research, 49, 309-324.
Thomas, M. S. C., Annaz, D., Ansari, D., Serif, G., Jarrold, C., & Karmiloff-Smith, A. (2009).
Using developmental trajectories to understand developmental disorders. Journal
of Speech, Language and Hearing Research, 52, 336-358.
Van Hees, S., Angwin, A., McMachon, K. & Copland, D. (2013). A comparison of semantic
feature analysis and phonological components analysis for the treatment of naming
impairments in aphasia. Neuropsychological Rehabilitation, 23 (1), 102-132.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 76-78
First data from constraint induced aphasia

therapy for Hungarian patients
Lívia Ivaskó1 , Alinka Tóth2 , Katalin Jakab2 & László Vécsei3,4
1 Developmental and Neuropragmatic Research Group, University of Szeged,
Hungary
2 Center for Neurorehabilitation, Neurology Department, Faculty of Medicine,
University of Szeged, Hungary

3 Neurology Department, Faculty of Medicine, University of Szeged, Hungary
3 Neuroscience Research Group, Hungarian Academy of Sciences,
University of Szeged, Hungary
Introduction
Our aim with this presentation is to show how could we make a special adaptation
of constraint induced aphasia therapy (Pulvermüller et al., 2001; Pulvermüller &
Berthier, 2008) for Hungarian aphasic patients. It is a very important and current
field in Hungary, because of the medical, social and therapeutical attention of
this population. Focusing on the theoretical basis of the original conception
of constraint induced procedure, we also claim that intensity, abstraction and
generalization of linguistic rules are able to improve patients’ language skills even
if they had a stroke some years before. On the basis of neuroplasticity and language
processing it seems to be a more successful way in reorganization, as it was
emphasized by Leon, Maher & Gonzalez Rothi (2011, 222).
Methods
Following the theoretical principals of constraint induced aphasia therapy
(Pulvermüller et al., 2001; Pulvermüller & Berthier, 2008) we have made the first
Hungarian version in 2012. It means 540 pairs of cards including nouns, verbs,
numbers, adjectives (some predicative adjectives), sentences. Noticing language
specific rules of Hungarian grammar we have made a special adaptation of the
model. Regarding on semantic, syntactic and phonological features of Hungarian
it was obvious to create a complex matrix for drawing patients’ attention to these
generative rules.
Left hemisphere damaged patients with agrammatic aphasia were patients of the
Center for Neurorehabilitation of the Department of Neurology at the University of
Szeged.
We worked with groups including three patients. The first group arrived in October
2012. Now we have data from 6 groups, from 18 agrammatic aphasic patients who
had suffered from stroke. Most of them had chronic aphasia, but none of them
were diagnosted with dementia or depression. All groups had been practiced 3 and
4 hours (depending on their physical capacity) per ten consecutive days. Before
CONSTRAINT INDUCED APHASIA THERAPY FOR HUNGARIAN PATIENTS 77
and after the intensive therapy all of them were tested by Boston Naming Test,
Hungarian adaptation of Western Aphasia Battery, Token Test, PragmaComp and
a special picture test (specially for frontal lobe executive functions), CIALT test
(using selected cards from the package for the therapy), and their relatives were
interviewed by the questionary CAL. All of the patients and their relatives were
informed of the clinical trial of the new adaptation.
Results
Practicing language abality with intensive grammatic and pragmatic
communicative stimuli has developed patients’ grammatic competence. Their
success was honoured by the other members of the group, as well as by their
caregivers at the Department. After the statistical analysis of data we should claim
that patients had very high score in picture stories beause of their intact executive
functions. Tests of capacity for verbal communication from Boston Naming Test,
Token Test and the WAB were analyzed in a very detailed way. Concentrating
on the patients intact cognitive functions we differentiated between semantic
and phonological properties of the verbal forms, so data were subcategorized
depending on the types of false or missed answers. Comparing data from
pretherapeutic time and the post we can conclude that most of the patients
have been more sensitive to the grammatically relevant semantic, phonological
features of verbal stimuli after the therapy (Tabor Connor, 2012, 227-8). Their
pragmatic competence was intact, they understood jokes well, they interpreted
conversational norms and were able to differentiate between the proper forms
of politeness but they were not able to use these forms as well. Playing with the
cards of the therapy they were asked to use some pragmatically relevant utterances
for asking people something, or suggesting, requesting or refusing something.
Based on the questionary CAL they became motivated to use these kinds of
communicative verbal forms in their everyday language use after they left the
Department.
Discussion
Therapeutic language games could regain some of patients’ lost linguistic functions
in a very short time with a quite intensive guided practice not to avoid the
problematic forms of verbal stimuli but with the help of corporation and
motivation. The indicators of this activity and intensity based therapy focus
on both the production and the comprehension of speech in an appropriate
communicative context. This experience makes them able to use this severally
confirmed form of generating linguistic abstract rules to use their reinforced
grammar productively in an adequate situation after the therapeutical work.
78 IVASKÓ ET AL.
References
Leon, S. A., Maher, L. M. & Gonzalez Rothi, L. J. (2011) Laguage Therapy In: S. A. Raskin ed.
Neuroplacticity and Rehabilitation. The Guilford Press, New York, London p. 209-
233.
Pulvermüller, F., Neininger, B., Elbert, T., Mohr, B., Rockstroh, B., Koebbel, P. et al. (2001)
Constraint induced therapy of chronic aphasia after stroke. Stroke 32. 1621-1626.
Pulvermüller, F. & Berthier, M. L. (2008) Aphasia therapy on a neuroscience basis.
Aphasiology. 22 (6) 563-599.
Tabor Connor, L. (2012) Language. In: L. M. Carey ed. Stroke Rehabilitation. Insights from
Neuroscience and Imaging. Oxford University Press. p. 222-230.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 79-82
The influence of working memory on the

inflection of verbs in Alzheimer’s and
Parkinson’s disease: a case study
Fedor Christiaan Jalvingh1,2 & Roelien Bastiaanse3
1 St.Marienhospital-Vechta, KKOM, Geriatric Clinic of Vechta, Germany
2 Graduate School for Humanities, University Groningen, The Netherlands
3 Center for Language and Cognition Groningen (CLCG), University Groningen, The
Netherlands
Introduction
Memory loss is the characteristic feature of dementia of the Alzheimer type (DAT)
and can be considered as the beginning of a complex and degenerative process
of cognitive functions. Language problems, a result of deteriorating cognitive
functions, also occur in the early stages of Alzheimer’s disease and have received
a relatively little attention despite its clinical importance.
Several authors argue that all DAT-patients have language problems (e.g. Obler &
Albert, 1981). Some assume that the existence of language problems can play a
central role in diagnosing DAT and the severity of the disease (e.g. Cummings &
Benson, 1986). The presence of language problems has also been described as a
preclinical marker and main feature of the Alzheimer’s disease (Cuetos et al., 2007).
According to Prins and colleagues (2002), language comprehension, grammatical
and phonological processing are relatively unimpaired in the early stages of DAT.
However, they mention that it is difficult to assess these linguistic functions,
because it is hard to judge sentence comprehension in discourse and because the
influence of probable cognitive deficits, such as impaired working memory during
language processing, is difficult to identify.
Several studies mention that grammatical processing is intact in DAT (e.g. Rochon
et al., 1994; Waters et al., 1995). Others, however, suggest that it is already impaired
in early stages of the Alzheimer’s disease (e.g. Bschor et al., 2001). This contrast may
be the result of the heterogeneity of the mental disorder, being caused by individual
degenerative neuroanatomical spreading patterns, leading to the impairment at
distinct linguistic levels. Another explanation for these divergent results is the
fact that grammatical impairments are hidden in deeper underlying structures of
language processing and that these deficits can only be explored by using particular
assessment conditions.
Although poor performance on language tests has been observed in Parkinson’s
Disease (PD) as well, it has been argued that these deficits are not cause by
a language problem per se, but rather by impaired executive functioning (see
Bastiaanse & Leenders, 2009, for an overview).
80 JALVINGH & BASTIAANSE
Methods
Participants
One female patient with probable DAT, 75 years old, MMSE-scores 21 (2011) / 16
(2013); One female patient with Parkinson Disease (PD), without depression and
dementia, 86 years old, MMSE-scores 30 (2011) / 29 (2013); 5 non-brain-damaged
speakers (NBDs), 4 female persons, 1 male person, mean age 84.
Materials
The German Version of the Test for Assessing Reference of Time (TART ; Bastiaanse,
Jonkers, & Thompson, 2008; German version Jalvingh & Bastiaanse, 2011) was
used. This test was developed for a cross-linguistic project to examine production
and comprehension of grammatical morphology associated with time reference
(Bastiaanse et al, 2011). The following verb forms were elicited: past imperfect
(e.g. schälte ‘peeled’), present perfect (geschält hat: ‘has peeled), present imperfect
(schält: ‘peels’) and future imperfect (schälen wird: ‘will peel’).
Responses were scored both quantitatively and qualitatively. A response was
considered correct when both the required verb form and the object were correctly
produced.
Procedure
The female patient with DAT and the female patient with PD were tested with the
digital version of the German TART twice, in Spring 2011 and approximately two
years later. Here the results of the production test are presented.
Results
The results are given in Table 1. The non-brain-damaged-participants all scored at
ceiling.
Table 1: Number of correctly produced items on the German TART of the individuals with
DAT and PD.
Past imperf. Pres. imperf. Pres. perf. Fut. imperf. total

DAT 2011 4 8 19 20 51/80
DAT 2013 0 0 15 19 34/80
PD 2011 19 20 19 19 77/80
PD 2013 18 20 20 19 77/80
The individual with PD scored outside the normal range, but only makes a few
errors on both occasions.
The individual with DAT has severe problems producing finite forms of lexical verbs
(past and present imperfect), whereas the periphrastic verb forms (present perfect
VERB INFLECTION IN ALZHEIMER’S AND PARKINSON’S DISEASE 81
and future imperfect) are relatively intact. This difference is significant (Fischer’s
exact p<0.0001). There is a significant decline over the years (Fisher’s exact
p<0.0110), which, interestingly also affects the periphrastic categories, although
not significantly.
Discussion
The person with DAT hardly produced any errors and there is no perceivable
difference between the categories. However, the errors that she produced are
interesting: she changed the required base word order (object-verb in German) to
derived word order (verb -object).
Cue: Hier könnte man sagen: Das ist der Mann der gerade... die Milch trank
object verb
Answer: Here one could say: That’s the man who just... drank the milk
verb object
However, her deficit is very mild and has not declined during the past two years.
Finite lexical verbs are difficult to retrieve and inflect for the individual with
DAT. We suggest that this is the consequence of problems integrating lexical
and grammatical information. A similar phenomenon with finite lexical verbs
compared to periphrastic forms was observed in individuals with fluent aphasia.
Bastiaanse (2011) found that the finite lexical verbs produced by these speakers
have less variation and are of higher frequency than normal, whereas the participles
of the periphrastic verb form have normal variation and frequency. Here we see the
same in the person with DAT. Interestingly, when she produces a finite verb, i.e. in
the past and present imperfect conditions, she often omits the object. Apparently,
producing both a finite verb and an object is too complex. Further research is
needed to show whether these parallels in the production of verbs between fluent
aphasia and DAT can be explained by similar pathological linguistic processes or
should be considered as having a different cause, leading to the same surface
problems in retrieving and inflecting finite verbs.
References
Bastiaanse, R. (2011). The retrieval and inflection of verbs in the spontaneous speech of
fluent aphasic speakers. Journal of Neurolinguistics, 24, 163-172.
Bastiaanse, R., Bamyaci, E., Chien, J. H, Lee, J., Thompson, C.K. & Yarbay Duman, T. (2011).
Time reference in agrammatic aphasia: a cross-linguistic study.
Bastiaanse, R., Jonkers, R. & Thompson, C. K. (2008). Test for Assessing Reference of Time
(TART). Groningen: University of Groningen.
Bastiaanse, R. & Leenders, K.L. (2009) Language and Parkinson’s Disease. Cortex, 45, 912-914.
Bschor, T., Kühl, K-P. & Reischies, F.M. (2001): Spontaneous Speech of Patients with
Dementia of the Alzheimer Type and Mild Cognitive Impairment. International
Psychogeriatrics, 13, 289-298.
Cummings, J.L. & Benson, D.F. (1986). Dementia of the Alzheimer type: An Inventory of
diagnostic clinical features. Journal of the American Geriatric Society, 34: 12-19.
82 JALVINGH & BASTIAANSE
Cuetos, F., Arango-Lasprilla, J.C., Uribe, C., Valencia, C., Lopera, F. (2007). Linguistic changes
in verbal expression: A preclinical marker of Alzheimer’s disease. Journal of the
International Neuropsyhological Society, 13, 433-439.
Obler, L.K., & Albert, M.L. (1986). The action naming test. Boston: VA Medical Centre.
Prins, R.S., Prins, N.D., & Visch-Brink, E.G. (2002). Taalstoornissen bij dementie. In:
Handboek Stem-, Spraak-, Taalpathologie, 17. Houten: Bohn Stafleu Van Loghum.
Rochon, E., Waters, G.S., & Caplan, D. (1994). Sentence comprehension in patients with
Alzheimer’s disease. Brain and Language, 46, 329-349.
Waters, G., Caplan, D., & Rochon, E. (1995). Processing capacity and sentence
Comprehension in patients with Alzheimer’s disease. Cognitive Neuropsychology, 12,
1-30.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 83-85
Verb comprehension in aphasic speakers of

Standard Indonesian
Bernard A. Jap1∗ , Ratna D. Soebadi-Haryadi2 & Roelien Bastiaanse3
1 European Master’s of Clinical Linguistics (EMCL); University of Indonesia
2 Airlangga University / Soetomo Hospital, East Java, Indonesia
3 Center for Language and Cognition (CLCG) Groningen, The Netherlands
∗ e-mail: b.a.jaya.jap@student.rug.nl, bernardjaya@gmail.com
Introduction
Studies on comprehension of single verbs are scarce, and there was only a case
study (Postman, 2003) looking at verb comprehension at the sentence level in
Standard Indonesian (SI). Kim and Thompson (2000) showed that comprehension
of verbs of agrammatic speakers is relatively spared and Jonkers and Bastiaanse
(2006) showed that instrumentality and name relation with a noun may influence
verb comprehension in non-fluent and fluent aphasia in a different degree.
The aim of the current study is to provide additional crosslinguistic data on verb
comprehension in aphasic speakers. We collected data from Standard Indonesian
(SI) aphasic speakers with a test adapted from Verb and Sentence Test (VAST; see
Bastiaanse, Edwards, Maas, and Rispens, 2003). In this test, verbs are controlled for
several factors that may influence verb production and comprehension in aphasia:
frequency; age of acquisition; imageability; transitivity; argument structure,
instrumentality; name relation to a noun; length.
SI verbs are morphologically simple: there is no inflection for Tense, Aspect or
Agreement. The only possible inflection is an (optional) accusative marker on the
verb. However, many verbs are compounds. Verbs, particularly instrumental verbs
can be noun-based or verb-based (see 1).
(1a) a noun based verb

sapu → menyapu
broomN → to sweepV
(1b) a verb based verb

hapus → menghapus
to eraseV → to eraseV
Noun- and verb-based verbs differ in the sense that the word class of the base is
different. The prefix of a verb-based verb does not have any function apart from
enforcing well-formedness (Sneddon, 1996). In the noun-based verbs, the prefix
interacts with the noun-base to form new meaning. In the example of ‘to sweep’,
meN- means to apply or use the base-word.
The aim of the current study is to find out whether instrumentality, name relation
84 JAP ET AL.
and the nature of compound verbs influence comprehension of aphasic individuals

in SI.
Methods
Participants
Six participants (3 transcortical motor, 1 Broca’s, 1 Anomic, and 1 Wernicke’s
aphasia) were included in the study. Their aphasia type was determined by Tes
Afasia untuk Diagnosis, Informasi, dan Rehabilitasi (TADIR, Dharmaperwira-Prins,
1996), except for the Wernicke’s aphasic speakers who only completed the TADIR
partially and was evaluated by their speech therapist to belong to this aphasia type.
Test Construction
The verbs included in the test were controlled for verb morphology, transitivity,
instrumentality, name relation with a noun, imageability, age of acquisition,
name agreement, length, thematic roles, frequency, and visual complexity. This
Indonesian adaptation of the VAST verb comprehension task uses a set of four
pictures per item. The picture of the target verb is paired with a semantically
related verb as a distractor. Then the target and the distractor are each paired
with a semantically related noun as additional distractors. For instrumental verbs,
the paired noun distractor is its instrument (e.g. target ‘hammering’, distractors
‘sawing’, ‘a hammer’ and ‘a saw’). The test consists of 48 items, 17 non-instrumental
verbs, 31 instrumental verbs of which 16 were name-related and 15 were not name-
related. Within the category of name-related instrumental verbs, there are 8 noun-
based and 8 verb-based target verbs.
Participants were shown the four pictures and were instructed to point at the
picture that best matches the target (spoken) verb. The answers were recorded
manually on the scoreform.
Results and Discussion

Twelve non-brain-damaged individuals (NBDs) were initially tested. All NBDs
performed at ceiling (mean correct percentage of group=99%, range=47-48 out of
48). The results of the aphasic individuals are given in Table 1.
The group is yet small and, therefore, it is hard to do any reliable statistical testing.
What can be seen from this table is that the largest individual discrepancy is in
the scores of the individual with Broca’s aphasia: the difference between scores
on instrumental and non-instrumental verbs is close to significant (Fisher’s exact:
p=0.07). This is in line with the findings of Bastiaanse and Jonkers (2006). They
argued that the poor performance of non-fluent aphasic individuals is due to the
fact that when the instrumental verb is activated, the name of the instrument is
co-activated. On a test where this instrument is one of the alternatives, this is
confusing and this results in poor performance.
VERB COMPREHENSION IN APHASIC SPEAKERS OF STANDARD INDONESIAN
85
Table 1: Proportions correct for the aphasic individuals (A1-A6). Mpo: months post onset;
tcm: transcortical motor aphasia;non-instr: non instrumental verbs; instr: instrumental
verbs; instr + name: instrumental verbs are related in name with the instrument; instr-name:
instrumental verbs are not related in name with the argument.
total non- instr instr verb- noun-

age mpo type instr
correct instr +name -name based based
A1 29 3 tcm 0.90 1.00 0.84 1.00 0.67 0.00 0.00

A2 52 3 tcm 0.90 1.00 0.84 0.88 0.80 0.13 0.13
A3 72 5 tcm 0.63 0.59 0.65 0.56 0.73 0.25 0.13
A4 48 4 Broca 0.58 0.77 0.48 0.50 0.47 0.38 0.63
A5 56 56 Wernicke 0.85 0.94 0.81 0.88 0.73 0.13 0.13
A6 55 4 Anomic 0.92 1.00 0.87 0.88 0.87 0.13 0.13
mean 52 12.5 0.80 0.88 0.75 0.78 0.71 0.17 0.19
With respect to verb- and noun-based verbs there is no clear difference in this small
group. Apparently, the problems of these patients are not cause by how the verb is
composed, but by the fact that it is a verb.
References
Bastiaanse, R., Edwards, S., Maas, E., & Rispens, J. (2003). Assessing comprehension and
production of verbs and sentences: The Verb and Sentence Test (VAST), Aphasiology,
17, 49-73.
Dharmaperwira-Prins, R.I.I. (1996) Tes Afasia untuk Diagnosis, Informasi, dan Rehabilitasi.
Jakarta: Balai Penerbit Fakultas Kedokteran, Universitas Indonesia.
Jonkers, R., & Bastiaanse, R. (2006). The influence of instrumentality and name-relation to a
noun on verb comprehension in Dutch aphasic speakers. Aphasiology, 20, 3-16.
Kim, M., & Thompson, C. K. (2000). Patterns of comprehension and production of nouns and
verbs in agrammatism: Implications for lexical organisation. Brain and Language,
74, 1-25.
Postman, W. A. (2003). What the Indonesian morphological causative can tell us about
aphasic comprehension. Cornell Working Papers in Linguistics, 19, 188-202.
Sneddon, J.N., (1996). Indonesian: A comprehensive grammar. London: Routledge.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 86-89
VAST-App - testing verbs and sentences with

the iPad
Dörte de Kok1 , Nienke Wolthuis2 & Roelien Bastiaanse1,2
1 Groningen Center of Expertise for Language and Communication Disorders,
University of Groningen, The Netherlands
2 Neurolinguistics Department, University of Groningen, The Netherlands
Introduction
In order to treat aphasia efficiently, a thorough diagnosis is needed. The results
of a general aphasia test, such as the Aachen Aphasia Test for Dutch (AAT ; Graetz
et al., 1993) can only provide an overall idea of the syndrome and the particular
problems of an aphasic speaker but does not inform about the underlying disorder.
Therefore, model-oriented diagnostics are needed to provide information about
the underlying disorder. While the Dutch version of the PALPA (Bastiaanse et
al., 1995) can provide this for single noun processing, many aphasic individuals
have (specific or additional) problems with verbs and sentences (Bastiaanse
& Bol, 2001; Jonkers & Bastiaanse, 2006). These can be diagnosed with the
“Werkwoorden en Zinnen Test (WEZT)” (“Verb and Sentence Test”) by Bastiaanse
and colleagues (2000). The WEZT is controlled for various linguistic variables and
allows for a diagnosis based on a psycholinguistic model of sentence processing
(Bastiaanse, 2010). The analysis and interpretation of this test, however, are rather
complicated and require a thorough understanding of the underlying linguistic
theory. Furthermore the administration is time-consuming and involves several
testing booklets. Therefore the WEZT has not been broadly used in the clinical field.
While therapists agree that determining the underlying deficit is an essential step
to beneficial treatment, they cannot find the necessary time and do not have the
necessary skills to administer the test and interpret the results as required.
The use of modern technology, such as a tablet computer, can help to overcome
the issues raised above. Administration of the test is more straight-forward and the
scoring can be done automatically. Even the interpretation of the results can be
aided by the application, by making use of machine learning.
The aim of the current project is therefore to develop a new aphasia test for verbs
and sentences, which will be administered, scored and analyzed with a tablet
computer. Initially the focus is on the iPad, but later on it is planned to make this
tool available for all kind of tablets. The project consists of three phases: (1) the
development of the linguistic materials and pictures, (2) the programming of the
iPad application and (3) the standardization and validation of the test. The data
collected in this last phase will be used to gain insight in the interaction between
all factors that may influence performance of language-impaired individuals.
VAST-APP - TESTING VERBS AND SENTENCES WITH THE IPAD 87
Methods
Test construction
The first test bundle consists of four subtests, “verb comprehension”, “action
naming”, “filling in infinitives in sentences”, and “filling in finite verbs in sentence”.
These tests are carefully constructed taking into account various relevant linguistic
variables, such as lexical frequency, age of acquisition, imageability, name relation
(to a noun), and number of arguments. For each of the used words black and white
line drawings are created by an artist. The complexity of these drawings is also
one of the variables taken into account. For variables that cannot be looked up
in existing corpora, ratings are obtained. This is for example the case for “visual
complexity” and “imageability”. Based on the ratings it is decided which items
are included in the final task. Furthermore the values of ratings and obtained
from corpora are used in the diagnosis in order to determine whether a particular
variable influences the performance of an aphasic individual.
Application development
The development of the application consists of programming on the one hand,
but on the other hand also procedural choices have to be made. For instance, it
needs to be technically implemented how often participants can listen to a pre-
recorded word or sentence and how self-corrections are handled. Next to the
testing interface, there are two additional interfaces: one for scoring the spoken
responses in the naming subtests and one for viewing the results.
Validation
For the first phase of validation, 100 adults (aged 20-80), male and female, from
different social backgrounds and different parts of the country will be tested in
order to acquire norms. Later on a group of 50 aphasic individuals (Dutch and
Flemish) will be tested. The Token Test will be used to measure the construct
validity of the test. Item response theory will be used to analyze the data in order
to determine the reliability and internal validity of the test. A selection of the
language impaired adults will be tested again after 4 weeks, to measure test-retest
reliability. For inter-rater reliability of the production tests (that will not be scored
automatically), two independent professional aphasiologists will rate the answers
and the correlations will be calculated.
Results
Currently the first two phases of the test construction have been carried out.
The items for the different subtasks have been developed and were controlled
for various linguistic features. These items have been incorporated in a first
version of the application. A preliminary version of the application has been
developed (see Figure 1). It provides facilities for displaying the material (visually
88 DE KOK ET AL.
and auditorily) and recording of the results. This application can already be used in
the validation phase, which will start soon. The results of the non-brain-damaged
control participants will be presented at the conference.
Figure 1: Screenshot from the task ‘Insertion of finite verb forms’.
Discussion
Language testing used to be a laborious procedure involving a lot of testing booklets
and forms to be carried along. The analysis was often even more time consuming
and the interpretation of the data difficult. With the tablet application we develop,
these troubles can be overcome. This results in a user-friendly but nonetheless
sophisticated test that controls for various variables and takes these into account
in the analysis of the results. This way, a diagnosis of the underlying disorder
is made possible for therapists lacking a deep linguistic background. Therefore
more aphasic individuals can be properly diagnosed - a prerequisite for effective
treatment.
References
Bastiaanse, R. (2010). Afasie. Houten: Bohn Stafleu van Loghum.
VAST-APP - TESTING VERBS AND SENTENCES WITH THE IPAD 89
Bastiaanse, R. & Bol, G. (2001) Verb Inflection and Verb Diversity in Three Populations:
Agrammatic Speakers, Normally Developing Children, and Children with Specific
Language Impairment (SLI). Brain and Language, 77, 274-282.
Bastiaanse, R., Bosje, M. & Visch-Brink, E.G. (1995). PALPA: De Nederlandse versie. (PALPA:
The Dutch version). Hove: Lawrence Erlbaum.
Bastiaanse, R., Maas, E. & Rispens, J. (2000) Werkwoorden- en Zinnentest (WEZT). Lisse,
Swets Test Publishers.
Graetz, P., De Bleser, R. & Willmes, K. (1992). Akense Afasietest (AAT). Amsterdam: Hogrefe.
Jonkers R., & Bastiaanse, R. (2006) The influence in instrumentality and name-relation to a
noun on verb comprehension in Dutch aphasic speakers. Aphasiology, 20, 3-16.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 90-94
Past tense in children with focal brain lesions

Polyxeni Konstantinopoulou1,2 , Stavroula Stavrakaki1 ,
Christina Manouilidou3 & Demetrios Zafeiriou1
1 Aristotle University of Thessaloniki, Greece
2 Birkbeck University of London, UK
3 University of Patras, Greece
Introduction
Recent studies on children with focal brain lesions (FL) have compared their
performance to that of typically developing (TD) children and examined potentials
for rehabilitation (Balantyne et al. 2008; Marchman et al. 2004).
The present study aims at examining the production of past tense for existing
and novel verbs by Greek FL children in comparison TD children matched on
chronological age (CA). By so doing, we address the question of recoverability in
linguistic abilities of children with FL.
Past tense in Greek is a morphologically complex form, as tense interacts with
aspect (perfective & imprefective). Consequently, Greek distinguishes between a
perfective and an imperfective past tense (Holton et al., 1997). The imperfective
past is formed with the present tense stem (e.g. graf -o, I write; egraf -a, I was
writing) and thus it is morphologically simpler than the perfective one. The
perfective past tense can be either sigmatic (regular), which involves a segmentable
affix (-s-), or non sigmatic (irregular).
Our still on-going study aims at investigating whether there are differences in the
performance of children with FL compared to that of TD children with respect to
the production of
(i) perfective vs. imperfective forms for existing and novel verbs
(ii) sigmatic vs. non-sigmatic past tense forms for existing and novel verbs
Methods
Participants
So far four children with FL have been tested. They all suffered a post-natal stroke
resulting in a FL as shown by MRI.
In more details:
Participant FL1: girl, aged 8;5, with a history of neurofibromatosis. At the age of 5
she suffered ischemic stroke with symptoms of right hemiparesis due to a middle
cerebral artery infraction. The stroke resulted in a left FL (lobes involved: temporal,
PAST TENSE IN CHILDREN WITH FOCAL BRAIN LESIONS 91
parietal). She is currently motor impaired; no history of seizures.
Participant FL2: boy, aged 8;5. At the age of 17 months, he experienced a left
ischemic stroke due to a watershed infarction. The stroke resulted in a left FL
(lobes involved: parietal). He currently suffers from right hemiplegia; no history of
seizures.
Participant FL3: boy, aged 13;2, diagnosed with a congenital heart disease. At the
age of 3, he experienced a left ischemic stroke while he was undergoing heart
operation. He had an infraction of the middle cerebral artery which resulted in a
left FL (lobes involved: temporal, paretial). Currently he does not show any motor
impairment; no history of seizures.
Participant FL4: boy, aged 6. Immediately after birth, he was diagnosed with aortic
stenosis and suffered an infraction of the middle cerebral artery resulted in right FL
(lobes involved: temporal, paretial). Currently he has left hemiparesis and shows
motor impairment; no history of seizures.
The performance of those individuals on past tense is compared to that of TD

children and 10 adults, as it is reported by Stavrakaki & Clahsen (2009) who used
exactly the same task. Specifically, the 8-year-old children with FL have been
matched on CA with 12 TD children aged 8-9 (CH-VIII) and the 6-year-old boy with
16 TD aged 6-7 (CH-VI). The 13-year-old boy’s performance was compared to that
of adults as adult level of performance is shown on this task after a certain CA.

We used the Perfective Past Tense Test (PPTT) designed to elicit perfective past
tense forms (Stavrakaki & Clahsen, 2009). The test materials included:
• Existing sigmatic and non-sigmatic verbs (N=20)
• Novel sigmatic and non-sigmatic rhymes (N=20)
• Novel non-rhymes (N=10)
The experimental procedure was as follows:

The participant was presented with pairs of two pictures (60 picture pairs, 50 for the
experimental items and ten fillers). The first picture presented an ongoing action
(e.g. a boy writing a letter), while the second picture showed the corresponding
completed action, e.g. a written letter. The experimenter pointed to the first picture
saying for example, ‘Here the boy is writing a letter’, and then to the second picture
saying ‘and what did the boy do here?’. Importantly, this set-up requires a perfective
past tense form to describe the second picture (e.g. the boy wrote the letter).
92 KONSTANTINOPOULOU ET AL.
Results
The children’s responses were classified as (i) sigmatic, (ii) non-sigmatic and (iii)
other. Other responses included imperfective past tense forms (exclusively for
participants with FL and mostly for TD population). See Tables 1 - 3.
Existing verbs
All children with FL showed ceiling performance on sigmatic forms except for the
6-year-old child with FS while the three younger children showed delay in the
acquisition of non-sigmatic forms.
FL participants showed more reliance on other forms
(=imperfective past tense forms) compared to TD children (aged 6 and 8) and
adults.
Table 1: Mean percentages of the production of correct and incorrect (sigmatic/non-
sigmatic or other) forms of existing verbs in the sigmatic and the non-sigmatic condition
SIGMATIC CONDITION NON-SIGMATIC CONDITION

Correct Non-sig Other Correct Sigmatic Other
FL 1 100 0 0 60 0 40
FL 2 88.88 0 11.11 66.66 0 33.3
FL 3 100 0 0 80 0 20
FL 4 60 0 40 37.5 12.5 50
ADULTS 100 0 0 97 0 3
CH-VIII (8-9-y.o.) 99.17 0.83 0 90 7.50 2.50
CH-VI (6-7-y.o.) 93.12 1.25 5.62 73.75 18.13 8.12
CH-III (3-4-y.o.) 69.99 0.71 29.3 35.53 11.43 53.04
Novel rhymes
Individual variation but trends for reliance on other
(=imperfective past tense forms) for all FL children.
Table 2: Mean percentages of the production of sigmatic, non-sigmatic or other forms for
novel verbs rhyming with existing sigmatic or non-sigmatic verbs
SIGMATIC CONDITION NON-SIGMATIC CONDITION

Correct Non-sig Other Correct Sigmatic Other
FL 1 20 20 60 0 14.28 85.71
FL 2 55.55 11.11 33.33 22.22 22.22 55.55
FL 3 100 0 0 0 90 10
FL 4 50 12.5 37.5 33.33 33.33 33.33
ADULTS 92 1 7 20 73 7
CH-VIII (8-9-y.o.) 87.50 4.17 8.33 11.02 70.65 18.33
CH-VI (6-7-y.o.) 80.32 3.39 16.29 4.62 80.28 15.10
CH-III (3-4-y.o.) 43.97 0 56.03 2.78 52.63 44.59
PAST TENSE IN CHILDREN WITH FOCAL BRAIN LESIONS 93
Novel non-rhymes
Remarkably high number of other (=imperfective past tense forms) by the 8 and
the 6 year old children with FL in contrast to TD children who produced sigmatic
(regular) verb forms. This performance resembles that of younger TD children
(aged 3) who produced a large number of imperfective past tense forms in this
condition.
Table 3: Mean percentages of the production of sigmatic, non-sigmatic or other forms for
non-rhyming verbs
Sigmatic Non-sig Other

FL 1 0 12.50 87.50
FL 2 70 0 30
FL 3 100 0 0
FL 4 25 0 75
ADULTS 91 5 4
CH-VIII (8-9-y.o.) 80.83 10 9.17
CH-VI (6-7-y.o.) 77.10 5.27 17.63
CH-III (3-4-y.o.) 39.48 9.24 51.28
Discussion
The 13 year-old child with FL showed performance within typical range in most of
the cases. FL and TD participants performed better on existing sigmatics than non-
sigmatics. The findings also revealed a strong tendency for reliance on imperfective
past tense forms and thus show that the FL children avoided the morphological
complexity of the perfective past tense by employing a simpler verb form. That
was evident especially in the novel non-rhymes condition. Remarkably, strong
reliance on imperfective forms was also attested in very young TD children (CH-
III). We suggest that the three younger FL participants follow the typical path for
perfective past tense acquisition with some delay. We point out that reliance
on imperfective past tense was also attested in adult Greek aphasia indicating
preference for morphological simplicity in these patients (Stavrakaki & Kouvava,
2003). With respect to the present findings, we interpret reliance on imperfective
forms as indication of grammatical immaturity and, consequently, delay in the
acquisition of the perfective past tense form by FL children.
References
Ballantyne, A. O., Spilkin, A. M., Hesselink, J. & Trauner, D. A. (2008). Plasticity in the
developing brain: intellectual, language and academic functions in children with
ischaemic perinatal stroke. Brain, 131, 2975-2985.
94 KONSTANTINOPOULOU ET AL.
Holton, D., Mackridge, P., & Phillipaki-Warburton, I. (1997). Greek: A comprehensive

grammar of the modern language. London: Routlege.
Marchman, V. A., Saccuman, C. & Wulfeck, B. (2004). Productive use of the English past
tense in children with focal brain injury and specific language impairment. Brain
and Language, 88, 202-214.
Stavrakaki, S. & Clahsen, H. (2009). The Perfective Past Tense in Greek Child Language.
Journal of Child Language, 36, 113-142.
Stavrakaki, S. & Kouvava, S. (2003). Functional categories in agrammatism: Evidence from
Greek. Brain and Language, 86 (1), 129-141.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 95-97
Dichotic listening in professional simultaneous

interpreters
Mary H. Kosmidis1 , Kalliopi Megari1 & Stavroula Stavrakaki2
1 School of Psychology, Aristotle University of Thessaloniki, Greece
2 School of Italian Language and Linguistics, Aristotle University of Thessaloniki,
Greece
Introduction
Simultaneous interpretation is a highly demanding cognitive process. It requires
not only a high level of proficiency in two languages, but also necessitates complex
linguistic perception and expression: simultaneously decoding the incoming
speech in the source language and expressing it in the target language. As such, it
has been associated with increased verbal memory (Cristoffels et al., 2006), verbal
fluency (Moser-Mercer, 2000), and semantic processing and working memory
capacities relative to those based on mere proficiency in a second language (i.e.,
teachers of a foreign language) (Stavrakaki et al., 2012).
Considering the association of bilingualism with unique cognitive mechanisms,
we wondered whether professional experience in simultaneous interpretation
might be associated with altered cerebral pathways for the processing of linguistic
information. In particular we expected that it would be related to strong
interhemispheric processing, rather than the typical left hemisphere (right ear)
dominance for language. Indeed, while the left hemisphere is dominant for
language in most right-handed monolinguals, there is some evidence suggesting
different cerebral organization for language in bilingual and polyglot individuals
(Fabbro et al., 1991). Additionally, an investigation of Chinese-English bilingual
adults showed interhemispheric processing on a dichotic word listening task,
unlike the right ear advantage (REA) found in their comparison group of
monolingual speakers of English (Ke, 1992). Finally, Fabbro and colleagues (1990)
investigated hemispheric specialization for language in female students training
in interpretation and in monolingual controls. They employed an automatic
speech production task to investigate two paradigmatic modes of interpreting,
namely the word-for-word technique and the meaning-based technique in L1 (first
language/mother tongue), L2 (second language) and L3 (third language). The
results showed no significant cerebral lateralization for the mother tongue (Ll) in
both groups, but weak left hemispheric lateralization for L2 in the interpreting
students. The investigators considered a number of factors, which might account
for their findings, including the gender of their participants. Clearly, more research
on this topic is warranted.
In the present study, we re-addressed the question of hemispheric specialization
for language in professional simultaneous interpreters (SI), including a group of
96 KOSMIDIS ET AL.
experienced SI and another group of professionals with demonstrated proficiency

in a second language to control for bilingualism. Specifically, we compared
the SI to experienced teachers of a foreign language (FL), and to a group of
non-bilingual individuals (NB), using a dichotic listening paradigm. Considering
not only the structural brain areas typically involved in linguistic processing
(i.e., left hemisphere), but also the functional mechanisms (related to bilaterally
represented attentional resource activity; Reinvang et al., 1994), and our own
findings of increased attentional and working memory capacity in SI (Stavrakaki et
al., 2012), we hypothesized that the SI will show less lateralized language processing
than both other groups, reflecting the additional skills mastered beyond mere
proficiency in a second language.
Method
Participants & Procedure
We administered a dichotic syllable listening task to SI (n=15; male:female=1:14),
FL (n=15; male:female=1:14) and NB individuals (n=35; male:female=3:32). (The
latter group presumably had some knowledge of a second language, as it is required
in school curricula and university programs, but had no professional experience
in language teaching or interpreting). All three groups were matched on age
[F(2,64)=1.61, p=.208] and level of education [F (2,64)=1.52, p=.227]; the SI and
FL groups were matched on number of years of professional experience [t(28)=.85,
p=.404]. The task included consonant-vowel syllables (ba, ga, da, pa, ka, ta),
wherein two stimuli were presented simultaneously, one in each ear. Examinees
were instructed to report the first syllable they heard for each pair.
Results
The variable of interest was right ear advantage (REA), calculated as follows:
REA=(right ear syllables-left ear syllables)/(right ear syllables+left ear syllables).
A positive score reflects more correct answers from the right ear than the left, while
a negative score indicates that the individual reported more correct answers from
the left ear than the right. We found no group effect for REA [F(2,62)=.163, p=0.850,
η2 =.005]; the SI group REA [mean=.08 (SD=.40)] was slightly smaller than that
of the FL [mean=.13 (SD=.21) and the NB [mean=.14 (SD=.33)] groups, but this
difference did not meet the criterion of statistical significance.
Discussion
In contrast to our predictions, we found no significant difference in language
processing laterality in our group of SI, relative to that of FL and NB control
adults. In fact, in all three groups, we found a low REA. This is consistent with
previous findings from our lab of a weak REA (=.03) when given instructions
to state all syllables heard vs. a stronger REA when the same individuals were
DICHOTIC LISTENING IN PROFESSIONAL SIMULTANEOUS INTERPRETERS97
instructed to report the syllables heard only from the right ear (=.71) (unpublished
data). Additional data from our group reported similarly low REAs among elderly
women on a word-based dichotic listening task (ranging from .03-.16, depending
on literacy and level of education) (Kosmidis et al., 2004).
The present findings appear to contradict the conclusions reported by Ke (1992).
In that study, however, no REA was calculated per se. In that study, the investigator
compared the bilingual and monolingual groups on the difference in the number of
stimuli reported for each ear. When we, too, compared our groups on each ear we
found no interaction of these factors. Thus, methodological factors may account
for inconsistencies in the relevant literature.
Several other factors may explain our findings as well. One possibility is that
the task is not sufficiently sensitive to detect subtle differences in REA. Another
may be related to the gender ratio, as all groups consisted predominantly of
women. Finally, it is also possible that differences in cognitive skills related
to language may not necessarily reflect differences in cerebral pathways with
respect to language processing laterality. Both the methodological and the cerebral
pathway hypotheses warrant further exploration.
References
Fabbro, D., Gran, B., & Gran, L. (1991). Hemispheric specialization for semantic and syntactic
components of language in Simultaneous Interpreters. Brain & Language, 41, 1-42.
Fabbro, D., Gran, L., Basso, G., & Bava, A (1990). Cerebral lateralization in simultaneous
interpretation. Brain & Language, 39, 69-89.
Ke, C. (1992). Dichotic listening with Chinese and English tasks. Journal of Psycholinguistic
Research, 21, 463-471.
Kosmidis, M. H., Tsapkini, K., Folia, V., Vlahou, C. H., & Kiosseoglou, G. (2004). Semantic and
phonological processing in illiteracy. Journal of the International Neuropsychological
Society, 10, 818-827.
Reinvang, I., Bakke, S.J., Hugdahl, K., Karlsen, N.R., & Sundet, K. (1994). Dichotic listening
performance in relation to callosal area on the MRI scan. Neuropsychology, 8, 445–
450.
Stavrakaki, S., Megari, K., Kosmidis, M. H., Apostolidou, M., & Takou, E. (2012). Verbal
fluency and working memory in simultaneous interpretation. Journal of Clinical and
Experimental Neuropsychology, 34, 624-633.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 98-100
Naming actions in non-fluent aphasia: an fMRI

study of compensatory reorganization of brain
activity1
Elena G. Kozintseva1,2 , Olga V. Dragoy1,2 , Svetlana A. Malyutina3 ,
Maria V. Ivanova2 , Daniil A. Sevan1 , Svetlana V. Kuptsova1,2 ,
Alexey G. Petrushevsky1 , Oksana N. Fedina1 & Evgenij F. Gutyrchik4
1 Center of Speech Pathology and Neurorehabilitation, Moscow, Russia
2 National Research University Higher School of Economics, Moscow, Russia
3 Moscow Lomonosov State University, Moscow, Russia
4 Ludwig Maximilians University, Munich, Germany
Introduction
The key characteristics of motor (Broca) aphasia are verb finding difficulties
and effortful speech production in general (among other symptoms). These
characteristics may be related to different loci of linguistic deficit: lexical retrieval
and motor execution. The aim of the study was to identify the normative brain
activation associated with verbs generation in healthy subjects and patterns of its
reorganization depending on the locus of linguistic deficits in patients with motor
aphasia.
Method
The study involved 18 healthy individuals (mean age 44) and 4 patients with
aphasia due to a lesion in the left hemisphere (mean age 49). All patients were
diagnosed with efferent motor aphasia, according to Luria’s neuropsychological
classification, of mild to moderate severity. Disorders of expressive speech were
observed in all patients. Lesions varied and included the inferior frontal gyrus only
in P2 (pars opercularis). All participants were native speakers of Russian and were
premorbidly right-handed.
Participants were presented with pictures of actions and abstract images. Verbs
were balanced on frequency, imageability, length and argument structure. As a
control condition, abstract images (digitally distorted real images) with the same
level of objective visual complexity were presented. Participants were asked to say
out loud what the hero was doing on the picture or to pronounce the pseudoverb
"kavaet" in response to abstract images.
Each of the two fMRI experimental sessions consisted of 18 blocks (12 with real
actions, 6 with abstract images). A block consisted of three pictures presented for
5.5 sec each, with 0.5 sec interstimulus interval. Blood oxygen level dependent
1 This study was supported by the Russian Foundation for Basic Research (RFBR), grant 13-06-00651a.
NAMING ACTIONS IN NON-FLUENT APHASIA 99
imaging (BOLD) was performed on a 1.5T Siemens Avanto scanner using gradient-
echo planar sequence (TE= 50 ms, TR= 3000 ms, FOV = 25 x 25 cm, 64 x 64 matrix,
voxel dimension 3 x 3 x 3 mm). A high-resolution anatomical image (T1-weighted,
MPRAGE; 0.98 x 0.98 x 1 mm; TE/TR 3/1900 ms) was also acquired. FMRI data
analysis was performed in SPM8.
Action naming was also tested out of the scanner in participants with aphasia the
following day (a preliminary study confirmed that patients with aphasia show no
learning effect in naming identical action pictures on two consecutive days). The
same action pictures were presented with the same timing parameters, but in a
different order. Responses were quantitatively and qualitatively analyzed.
Results
In healthy Russian speakers, action naming elicited extra brain activation in
occipital regions bilaterally, left inferior temporal gyrus and, critically, the
triangular part of inferior frontal gyrus - relative to the baseline condition (uttering
a pseudo-verb in response to an abstract picture). Individuals with aphasia showed
brain activations in bilateral occipital regions, as well as in inferior frontal gyrus,
similarly to normal individuals.
Extra activation found in patients, but not in healthy individuals, was dependent
on their quantitative scores and type of errors during the naming testing
out of the scanner. P1 and P2 named correctly 68% and 65% of actions,
correspondingly. Majority of the incorrect answers were non-dominant, but
synonymous nominations (‘lifting’ - ‘pulling’). Thus, P1 and P2’s ability to name
actions might be considered relatively spared. In contrast, P3 and P4 were only
32% and 26% correct, and the errors were mostly non-responses and semantic
paraphasias (‘searching’ - ‘palpating’). The same patients’ grouping was revealed in
fMRI results. P1 and P2 activated right cerebellum regions for action naming more
than in the baseline condition. For the same contrast, P3 and P4 showed wide-
spread frontal left hemisphere activation (supplementary motor area, precentral
gyrus), as well as additional right hemisphere activation (supplementary motor
area, precentral gyrus, inferior and middle temporal gyrus in P3; middle frontal
gyrus in P4).
Discussion
The activation pattern found in healthy individuals supports critical involvement
of inferior frontal gyrus in verb production. Additional activation in response to
action pictures relative to abstract pictures in bilateral occipital regions and left
inferior temporal gyrus, which are parts of the ventral visual stream, reflects the
more advanced level of complexity of pictures with realistic actions and tools.
The observed two different patterns of brain activation in patients with non-
fluent aphasia suggests that P1 and P2, on one hand, and P3 and P4, on the
other hand, have two different locus of linguistic impairment and use distinct
100 KOZINTSEVA AL.
brain mechanisms to overcome their deficits. Verb retrieval per se was relatively
spared in P1 and P2, as follows from their naming scores. It was motor execution
of the word that caused difficulties in them. The effort to overcome those
difficulties resulted in specific activation in the right cerebellum, which is known
to be a regulator of speech temporal sequencing. In contrast, P3 and P4 had
intrinsic linguistic difficulties with verb finding and used wide-spread bilateral
frontotemporal network to overcome them.
Thus, in addition to the identification of brain substrate involved in normative verb
production, the present study showed how different loci of linguistic deficits within
the same aphasia syndrome are represented in distinct cerebrocerebellar networks.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 101-104
An fMRI study of morphosyntactic processing

in Chinese
Sam-Po Law1 , Xi Yu1 , Yanchao Bi2 & Zaizhu Han2
1 Divisionof Speech and Hearing Sciences, the University of Hong Kong,
Hong Kong SAR
2 National Key Laboratory of Cognitive Neuroscience and Learning,
Beijing Normal University, China
Introduction
Languages vary widely in the complexity of their morphosyntactic system. Chinese,
representing one end of the spectrum, is well-known for its impoverished
inflectional morphology. From the perspective of cross-linguistic study, the relative
simplicity of the Chinese system raises the question of whether the degree of
complexity or richness of a grammatical component would affect its representation
in the brain, analogous to previous reports of null findings for representation of
lexical (or derivational) morphology in English (e.g. Davis, Meunier, & Marslen-
Wilson, 2004; Devlin, Jamieson, Matthews, & Gonnerman, 2004; but see Vannest,
Polk, & Lewis, 2005), but positive findings in Hebrew (e.g. Bick, Goelman, & Frost,
2008), German (e.g. Meinzer, Lahiri, Flaisch, Hannemann, & Eulitz, 2009), and
Italian (e.g. Berlingeri et al., 2008; Marangolo et al., 2006). In addition, contrary
to most European languages in which the verbal paradigm is more complex
than the nominal counterpart, the contrast between the nominal classifier and
verbal aspect marker inventories in Chinese presents the opposite pattern. This
difference renders Chinese a highly interesting case for assessing the view that
neural correlates of morphosyntactic processes, particularly in LIFG, specific to
a grammatical class is driven by computational demands. If correct, one would
expect to find brain areas in the left prefrontal cortex more strongly activated
for nominal than verbal grammatical morphemes in Chinese, and none for the
reverse comparison. On the other hand, if neural representation of grammatical
morpheme processing does not simply reflect processing demand but in fact is
form class specific, it is possible to find separate neural correlates for nominal
classifiers and verbal aspect markers.
Methods
Participants
Forty-seven right-handed native Mandarin speakers with no history of psychiatric
or neurological disorders were recruited from Beijing Normal University to
participate in the current study, with 27 (16 females, Mean age = 20.8, SD = 2.14) in
a grammaticality judgment experiment, and 20 (10 females, Mean age = 21.3, SD =
102 LAW ET AL.
3.00) in a sentence completion experiment.
Tasks and procedures

In the sentence completion task, participants were asked to supply either a
classifier or an aspect marker to complete a sentence. In the grammaticality
judgment task, grammatical violation arose from inappropriate pairing between a
noun and a classifier (CL), or from the incongruity between an aspect marker (ASP)
and the lexical aspect (or semantic structure) of a verb. As in Yu et al. (2011, 2012),
abstract and concrete nouns and verbs were used as stimuli for representativeness.
An event-related design was adopted. Functional MRI scans were collected on a
3.0 Tesla Siemens scanner using a 12-channel transmit/receive gradient head coil
(Beijing Normal University, China). A T2*-weighted gradient-echo planar imaging
(EPI) sequence was applied to acquire the blood oxygen level-dependent (BOLD)
signals (flip angle 90◦ , TE = 30ms, TR = 2000ms, in-plane resolution = 3.125*3.125,
slice thickness = 4mm, slice gap = 0.8mm). Data preprocessing and analysis were
performed using SPM5 (http://www.fil.ion.ucl.ac.uk/spm/software/spm5/).
Conjunction analyses were conducted across concreteness conditions in the
sentence completion experiment to identify brain areas that were more activated
for nominal classifiers than verbal aspect markers as well as those that were more
active for aspect markers than classifiers regardless of concreteness (puncorrected
< 0.001 at voxel level, cluster extent threshold ≥ 60 voxels for each contrast
corresponding to Monte-Carlo corrected clusterwise alpha level of 0.049). These
regions then served as regions-of-interest (ROI) to detect differential activation in
grammaticality judgment to the classifier vs. aspect marker conditions using two-
way ANOVAs. Task-independent regions specifically activated for a grammatical
class were considered for their associated cognitive processes.
Results
Conjunction analyses of CL vs. ASP contrasts between the two concreteness levels
in the sentence completion task revealed that the left posterior middle temporal
gyrus (LpMTG) was activated more strongly for the ASP than CL sentences, whereas
regions showing greater activation for the CL conditions of both concreteness levels
included bilateral calcarine and lingual gyri, BA44 (k =56, corresponding to cluster-
level p = 0.06), bilateral orbital inferior frontal gyri and insula cortices (BA47, right
BA47 (rBA47)), as well as left supplementary motor area and superior medial frontal
gyrus (LSMA&SMedFG).
Results of the ROI analyses using two-way ANOVAs (grammatical morpheme x
grammaticality) of the grammaticality judgment task found (i) significant main
effects of grammatical morpheme and grammaticality in LpMTG with greater
activation for ASP and ungrammatical sentences (p < 0.01), (ii) significant effect
of grammatical morpheme in bilateral posterior cortices with higher activation for
CL sentences but the effect was confounded with sentence length, (iii) significant
grammaticality effects in bilateral BA47 and LSMA&SMedFG, with ungrammatical
AN FMRI STUDY OF MORPHOSYNTACTIC PROCESSING IN CHINESE 103
Figure 1: Task-independent regions of grammatical morpheme processing in Chinese

(nominal classifier regions in white, verbal aspect marker regions in black).
sentences inducing stronger responses (p < 0.05), and (iv) significant interaction
effects between grammatical morpheme and grammaticality in these areas, among
which higher activation for CL sentences than ASP sentences in grammatical trials
was observed in BA47 (t(21) = 2.42, p < 0.05) and LSMA&SMedFG (t(21) = 2.50, p <
0.05). (See Figure 1 for detail).
Discussion
Through contrasting the processing of classifiers and aspect markers representing,
respectively, nominal and verbal grammatical morpheme operations in receptive
and expressive tasks, we have identified task-independent distinct brain regions
differentially responsive to one type of stimuli over the other, and vice versa. We
attributed the activation in the left prefrontal cortex to greater selection demand
during processing of classifiers than aspect markers, which may reflect domain
general computational loads, consistent with views from studies of Indo-European
languages (Grindrod et al., 2008; Righi et al., 2009; Sahin et al., 2006; Siri et al., 2008;
Thompson-Schill et al., 1997), and the LpMTG to more demanding verb semantic
processing stemming from judging congruency between aspect markers and
semantic structure of verbs. The overall findings have significantly demonstrated
the existence of neural correlates of grammatical morpheme processing associated
with nouns and verbs in an analytic and classifier language.
104 LAW ET AL.
References
Berlingeri, M., Crepaldi, D., Roberti, R., Scialfa, G., Luzzatti, C., & Paulesu, E. (2008). Nouns
and verbs in the brain: grammatical class and task specific effects as revealed by
fMRI. Cognitive Neuropsychology, 25, 528-558.
Bick, A., Goelman, G., & Frost, R. (2008). Neural correlates of morphological processes in
Hebrew. Journal of Cognitive Neuroscience, 20, 426-420.
Davis, M. H., Meunier, F., & Marslen-Wilson, W. D. (2004). Neural responses to
morphological, syntactic, and semantic properties of single words: an fMRI study.
Devlin, J. T., Jamieson, H. L., Matthews, P. M., & Gonnerman, L. M. (2004). Morphology and
the internal structure of words. Proceedings of the National Academy of Sciences of
the United States of America, 101, 14984-14988.
Grindrod, C. M., Bilenko, N. Y., Myers, E. B., & Blumstein, S. E. (2008). The role of the
left inferior frontal gyrus in implicit semantic competition and selection: An event-
related fMRI study. Brain Research, 1229, 167-178.
Marangolo, P., Piras, F., Galati, G., & Burani, C. (2006). Functional anatomy of derivational
morphology. Cortex, 42, 1093-1106.
Meinzer, M., Lahiri, A., Flaisch, T., Hannemann, R., & Eulitz, C. (2009). Opaque for the
reader but transparent for the brain: neural signatures of morphological complexity.
Neuropsychologia, 47, 1964-1971.
Righi, G., Blumstein, S. E., Mertus, J., & Worden, M. S. (2009). Neural systems
underlying lexical competition: An eyetracking and fMRI study. Journal of Cognitive
Sahin, N. T., Pinker, S., Cash, S. S., Schomer, D., & Halgren, E. (2009). Sequential processing
of lexical, grammatical, and phonological information within Broca’s area. Science,
326, 445-449.
Siri, S., Tettamanti, M., Cappa, S., Della Rosa, P., Saccuman, C., Scifo, P., & Vigliocco, G. (2008).
The neural substrate of naming events: Effects of processing demands but not of
grammatical class. Cerebral Cortex, 18, 171-177.
Thompson-Schill, S. L., D’Esposito, M., Aguirre, G. K., & Farah, M. J. (1997). Role of the
left inferior prefrontal cortex in retrieval of semantic knowledge: A reevaluation.
Proceedings of the National Academy of Sciences of the United States of America, 94,
14792-14797.
Vannest, J., Polk, T. A., & Lewis, R. L. (2005). Dual-route processing of complex words:
new fMRI evidence from derivational suffixation. Cognitive, Affective, & Behavioral
Yu, X., Law, S-P., Han, Z., Zhu, C., & Bi, Y. (2011). Dissociative neural correlates of semantic
processing of nouns and verbs in Chinese —A language with minimal inflectional
morphology. NeuroImage, 58, 912-922.
Yu, X., Bi, Y., Han, Z., Zhu, C., & Law, S-P. (2012). Neural correlates of comprehension and
production of nouns and verbs in Chinese. Brain and Language, 122, 126-131.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 105-108
The effect of static versus dynamic depictions

of actions in verb and sentence production in
aphasia
Ilona Damen1 , Joyce Blankestijn-Wilmsen1 , Vicky Voorbraak-Timmerman1 ,
Joost Hurkmans2,3 & Roel Jonkers3
1 Rehabilitation Center “Roessingh centrum voor revalidatie”, The Netherlands
2 Rehabilitation Center “Revalidatie Friesland”, The Netherlands
3 University of Groningen, The Netherlands
Introduction
From clinical practice we know that there is a lack of materials for the treatment
of problems with verbs in naming and sentence production. In therapy, we often
use pictures or photographs depicting actions, although the meaning of certain
verbs clearly involves movement. We therefore questioned, if aphasic speakers
might benefit more from speech therapy in working with dynamic materials (video
clips) rather than static materials (pictures) while training action naming and/or
sentence production. During the therapy sessions with static materials speech
therapists often use gestures to cue the patient in naming verbs, which is rather
artificial.
Research to the role of the depiction in verb retrieval is scarce and still
contradictory. On the one hand, Berndt, Mitchum, Haendiges and Sands (1997)
found that five aphasic speakers were equally impaired in naming pictures of
actions and video clips of actions. On the other hand, Druks and Shallice (2000)
reported a patient who had superior verb naming of actions that were acted out
by the experimenter compared to those same actions portrayed in pictures. They
argued that the patient had been helped by the strengthened semantic context
and information given by a more dynamic presentation. D’Honicthun and Pillon
(2008) described a patient with frontotemporal dementia who showed a difference
between action and object naming under picture naming conditions. Verbs were
retrieved significantly worse than nouns. However, when the patient was presented
with the same verbs and nouns in a dynamic condition, specific difficulties with
action naming virtually disappeared.
In the current study, we explore the hypothesis that aphasic speakers are more
accurate in retrieving verbs in isolation and in sentence context in a dynamic
condition rather than in a static condition.
106 DAMEN ET AL.
Methods
Participants
Eleven aphasic speakers were included in this study (9 males, 2 females). The mean
age of the group was 55.6 years (range: 39-79). All participants were native speakers
of Dutch and suffered from a left-hemispheric stroke. Both fluent and non-fluent
aphasic speakers were tested. All participants were diagnosed with the Aachen
Aphasia Test (Graetz, De Bleser & Willmes, 1992). Participants were selected based
on the presence of syntactic disorders and difficulties in verb retrieval. Considering
sentence production, in spontaneous speech their syntactic structures had to be
either agrammatic or paragrammatic (<4 on the syntactic scale).
Materials
A naming task and a sentence production task were administered. Both the naming
task and the sentence production task contained 20 items. Half of the verbs used
were high-frequency verbs, the other half were low-frequency verbs (Celex, 2001).
The verbs were also matched for transitivity, instrumentality and name-relation to
a noun. There were two versions of each task with identical items; a static version
and a dynamic version. Both versions were presented on a computer screen. The
dynamic tasks included 20 video clips each with a duration of 4 seconds. The video
clips were soundless and the actions were filmed in a natural context.
The static tasks included 20 photographs each. The photographs were stills cut
from the video clip used in the dynamic condition. The photographs depicted the
peak moment of the action. The photograph stayed on screen for 4 seconds.
Procedures
There were two sessions per participant, with a lag of approximately a week,
randomly starting with the dynamic or the static condition. Within each session
the participant started with the naming task followed by the sentence production
task. The order of the items in all tasks was identical. Each task started with two
practice items. The tasks were performed on the computer which was controlled
by the researcher. The researcher showed the participant a photograph or video
clip, which automatically disappeared from the computer screen after 4 seconds. A
black screen was presented until a response was given. The participant was asked
to respond as soon as possible. The number of correct responses for both action
naming and sentence production was counted. Slight phonological errors were not
counted as incorrect. In the sentence condition, it was only counted whether the
correct verb was produced, not whether the correct form was mentioned.
Two sided paired sample t-tests (p<0.05) were used to compare the dynamic and
static condition.
STATIC VERSUS DYNAMIC ACTION DEPICTIONS IN PRODUCTION 107
Results
The scores are depicted in Figure 1. The results show that the participants were able
to produce significantly more correct reactions in action naming in the dynamic
condition as compared to the static condition (t(10)=3,105, p=0.011).
At the sentence level there was no significant difference in accuracy considering the
two conditions (t(10)=0.590, p=0.568).
22
Correct items Action naming Correct items Sentence production
20 20
18 18
16 16
14 static 14 static
12 mean: 13.18 (SD 4.6) 12 mean: 12.55 (SD 5.07)
10 dynamic 10 dynamic
8 mean: 14.82 (SD 4.07) 8 mean: 13.82 (SD 4.42)
6 6
4 4
2 2
0 0
1 2 3 4 5 6 7 8 9 10 11 participants 1 2 3 4 5 6 7 8 9 10 11 participants
s
Figure 1: Individual scores for action naming and sentence production for both conditions.
Figure 1: Individual scores for verb retrieval in isolation and in sentence context for both
conditions.
Discussion
The outcomes of this study show that in verb retrieval aphasic speakers profit from
a dynamic display of an action in comparison to a static depiction. However, this
is only seen in action naming and not in sentence construction. The absence of an
effect of depiction in sentence context, might be due to the fact that participants
suffer from a syntactic disorder and that due to this disorder the effect was
undone. Related to this, we have to mention that our aphasic group was rather
heterogeneous. Although participants all had problems with verbs, some produced
agrammatic speech and others paragrammatic speech. We expect a confounding
effect of syntactic problems on verb retrieval in sentence context especially in
aphasic speakers with agrammatic speech and therefore aim to test more aphasic
speakers considering the type of aphasia.
From this study it can be concluded that depicting action verbs in a dynamic
context, helps aphasic speakers in verb retrieval. In treating aphasic speakers with
verb retrieval problems, we therefore recommend to make use of video fragments
of actions.
Acknowledgements
This study was funded by Stichting Hulpfonds Het Roessingh.
108 DAMEN ET AL.
References
Berndt, R.S., Mitchum, C.C., Haendiges, A.N., & Sandson, J. (1997) Verb retrieval in aphasia:
1. Characterizing single word impairments. Brain and Language, 56, 68-106
Druks, J., & Shallice, T. (2000) Selective preservation of naming from description and the
“restricted preverbal message”. Brain and Language, 72, 100-128
Graetz, P., De Bleser, R., & Willmes, K. (1992). Akense Afasie Test. Amsterdam: Hogrefe
d’Honicthun, P. & Pillon, A., (2008) Verb comprehension and naming in frontotemporal
degeneration: the role of the static depiction of actions. Cortex, 44, 834-847
The Celex lexical database. (2001). Max Planck Institute for psycholinguistics. Centre for
lexical information Nijmegen.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 109-111
Early and late semantic processing of action

verbs: evidence from fluent and stuttering
speakers
Sarah Vanhoutte1 , Gregor Strobbe2 , Miet De Letter3,4 , Pieter van Mierlo2 , Marjan
Cosyns4 , Paul Corthals4,5 , John Van Borsel4,6 & Patrick Santens1,3
1 Department of Internal Medicine, Neurology, Ghent University, Belgium
2 MEDISIP, Department of Electronics and Information Systems, Ghent University –
iMINDS, Belgium
4 Department of Speech, Hearing and Language Sciences, Ghent University, Belgium
5 University College Ghent, Belgium
6 Universidade Veiga de Almeida, Rio de Janeiro, Brazil
Introduction
Although action verb processing has been repeatedly shown to elicit motor cortex
activation, no consensus is yet achieved on the function, timing and necessity of
this activation (e.g. Hauk & Pulvermüller, 2004; Tomasino et al., 2008, Hickok,
2010). If motor related areas contribute to word understanding, action verb
processing deficits should occur in patients with diseases affecting the motor
system. Stuttering is known to evoke abnormal activations in cortical and
subcortical motor related brain sites during speech production, speech perception
and silent reading. To our knowledge, action verb processing has never been
explored in stuttering. Furthermore, most studies using patient populations
contrast action verbs with non-action nouns. Since verbs are inherently more
difficult than nouns, an action verb processing deficit might rather be related to
grammatical aspects. Therefore, the present study aimed to evaluate action vs
non-action verb processing in both fluent and stuttering speakers. As a research
tool, electro-encephalography (EEG) was used because of its excellent temporal
resolution which permits a delineation of the time course of word processing.
Methods
Subjects
30 adult, fluent speakers (FS) (male/female ratio:22/8) and 30 adults who stutter
(AWS), matched for age, gender and education, were recruited. All participants
were right-handed. Speech samples were collected to ascertain each participant
was indeed a fluent or a stuttering speaker.
110 VANHOUTTE ET AL.
Stimuli
50 action and 50 non-action verbs were selected. The action verbs referred to
voluntary hand and/or arm movements (e.g. to throw), the non-action verbs were
abstract verbs unrelated to actions or body parts (e.g. to believe). They were
randomly presented in their infinitive form as single words. Participants were
instructed to read each of the words mentally and to avoid overt articulation.
Data acquisition and analysis

EEG data were collected from 24 Ag/AgCl electrodes that were placed on the scalp
according to the international 10/20 system. Off-line EEG data were analysed
using BrainVision Analyzer (Brain Products) and averaged to create event-related
potentials (ERPs). Source reconstruction was applied on the individual ERPs from
60 to 500 ms after word onset to evaluate the temporal flow of activation. 60 ms
was chosen as starting point because visual information needs on average 60 ms
to travel from the retina to higher cortical areas. This 440 ms during time window
was divided in 11 successive time windows of 40 ms. The reconstructed activity was
statistically analyzed with ANOVA.
Results
In FS, action verbs elicited stronger brain responses than non-action verbs at
several time points. From 60-100 ms, a higher activation in bilateral superior
parietal and right inferior frontal cortex was observed. From 100-140 ms, a larger
right dorsolateral prefrontal cortex (DLPFC) involvement was seen. A higher
bilateral temporal and sensorimotor activation was found from 260 to 420 ms and
from 300 to 380 ms respectively.
In AWS, bilateral sensorimotor cortex differentiation arose extremely early (60-
100 ms) and re-occurred in the right hemisphere from 300-340 ms. Remarkably,
this differentiation showed an opposite pattern: non-action verbs elicited more
activation than action verbs. No differentiation in inferior frontal nor in parietal
areas was observed.
Discussion
Fluent speakers
Differences between action and non-action verbs occur both in an early and a late
time window. Because action and non-action verbs were matched for visual and
orthographical features, evoked differences can only be attributed to differences in
lexical access and semantic processing.
Between 60 and 100 ms, a larger brain response to action verbs is observed in
bilateral superior parietal and right inferior frontal cortex, which both contain
mirror neurons. These neurons contribute to action verb understanding by
mentally simulating the act the verb refers to, which makes its meaning accessible
for language. At this stage lexical access is suggested to occur.
EARLY AND LATE SEMANTIC PROCESSING OF ACTION VERBS 111
In the consecutive window (100-140 ms), a larger involvement of DLPFC is

observed. We hypothesize that the mental simulation of the mirror neurons
implicitly triggered mental images and/or motor planning for which DLPFC is
responsible.
Differences in both temporal and sensorimotor areas are seen in a later time
window, between 260 and 380 ms. Temporal areas are involved in amodal semantic
analysis, sensorimotor involvement is related to the action content of the stimuli.
Recent theories suggest that since initial semantic retrieval has already emerged
earlier, post-conceptual processes occur in this stage.
Stuttering speakers
The twofold semantic processing is confirmed by the results of the stuttering
speakers. Sensorimotor differences are found in an early (60-100 ms) and a late
(300-340 ms) time window. However, a reversed pattern is observed with non-
action verbs eliciting more activation than action verbs. As expected, motor
cortex abnormalities cause alterations in action verb processing confirming its
importance for action semantics.
Motor alterations in stuttering are attributed to a reduced white matter integrity of
the left superior fasciculus longitudinalis (SLF). This white matter bundle is part
of the dorsal stream responsible for sublexical reading. However, this pathway is
hypothesized to have a role in action semantics because it would be responsible
for sensorimotor activation. Since no structural anomalies are reported in AWS in
the ventral stream, the present results confirm this hypothesis.
Conclusion
The present study confirms recent theories that posit a dual semantic processing
system. After a very early contribution of the mirror neuron system, later post-
recognition analyses occur in temporal and sensorimotor areas. In addition, results
from the AWS demonstrate the important contribution of the dorsal stream.
References
Hauk, O., & Pulvermüller, F. (2004). Neurophysiological distinction of action words in the
fronto-central cortex. Human Brain Mapping,191-201
Hickok, G. (2010). The role of mirror neurons in speech perception and action word
semantics.Language and cognitive processes,749-776
Tomasino, B., Fink, G., Sparing, R., Dafotakis, M., & Weiss, P. (2008). Action verbs and
the primary motor cortex: a comparative TMS study of silent reading, frequency
judgments and motor imagery. Neuropsychologia, 1916-1926
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 112-115
Arabic-speaking aphasics: Analysis of naming

errors
Reem S. W. Alyahya
King Fahad Medical City, Riyadh, Saudi Arabia
Introduction
Naming deficits are common symptoms in all types of aphasia; the degree of deficit,
however, may differ depending on the word class. Several studies have shown
that action- naming is more impaired than object-naming among individuals with
different types of aphasia [1, 2, 3]. Various linguistic and psycholinguistic studies
have been conducted to explain the greater action-naming deficits; however, little
evidence has been adduced from studies into naming errors, which vary broadly.
A picture-naming study on English- speaking aphasics [3] showed that object
pictures elicited more semantic errors than action pictures; whereas action pictures
elicited more visual misinterpretation errors, circumlocutions and omissions. The
authors argued that the greatest indicator of complexities involved in picture-
naming is visual misinterpretation errors, they also claimed that these errors might
affect action-naming more than object-naming because the relationship between
action pictures and their verbal labels are less direct, as they demand additional
inference-making, and because action pictures are conceptually and visually more
complex than object pictures. The present study aimed to provide further empirical
evidence on the differences between error types elicited by object versus action
pictures. It predicted that action pictures will elicit more visual errors than object
pictures, whereas object pictures will elicit a greater number of semantic errors
compared to action pictures.
Moreover, studies have associated the variety of error types to the clinical
diagnosis of aphasia. Some argued that semantic errors are commonly manifested
in Wernicke’s aphasics, but rarely in Broca’s aphasics [4], and others posited
that semantic errors are not associated with a particular type of aphasia [5].
Additionally, a study revealed that semantic errors are common among all types
of post-stroke aphasia [6]. However, no study was found that related different
error types to fluency in aphasia. Therefore, the current study further aimed to
investigate the distribution of naming error types in relation to two groups of
aphasics (fluent and non-fluent) among Arabic-speaking aphasics.
Methods
Fourteen Arabic-speaking adults clinically diagnosed with aphasia (seven fluent
and seven non-fluent) participated in the study. Their aphasia resulted from
acquired unilateral left- hemisphere-brain-damage at least three months post-
onset; and had no concomitant speech or cognitive impairment that interfered
ARABIC-SPEAKING APHASICS: ANALYSIS OF NAMING ERRORS 113
with their performance.

Fifty object and 50 action pictures from the Object and Action Naming Battery
[7] were presented to each participant separately. Participants named all pictures
using a single word and responses were deemed correct if they named the picture
using the target word.
Naming errors were then classified according to a pre-specified error classification
system [3, 6]. The errors were classified into eight types: (i) omission: total
absence of a verbal response, (ii) semantic-coordinate: response from the
same semantic category (iii) semantic-superordinate: too general response (iv)
semantic-associative: response thematically related to the target with a semantic
association (v) phonological: response with phonological distortion (vi) visual
error: response closely similar to the visual appearance of the picture with no
semantic relation or response that names parts of the picture other than the target
(vii) circumlocution: accurate definition of the target (viii) unrelated: response
with no clear relation to the target.
Results
A two-way repeated-measures mixed ANOVA was carried out on the percentage
of total naming errors; the effect of picture type on naming performance
was significant (F(1,12)=44.12, p<0.0001, partial η2 =0.786), with action pictures
eliciting more errors (52.57%) than object pictures (32.43%). The effect of aphasia
group was also significant (F(1.12)=142.64, p<0.0001, partial η2 =0.922), with non-
fluent aphasics producing more errors (70.43%) than fluent aphasics (14.57%).
However, the interaction effect between picture type and aphasia group was not
significant.
To measure the differences between visual and semantic errors (including
coordinate, superordinate and associative) elicited by object versus action pictures,
a two-way repeated-measures ANOVA was performed; the effect of error type was
significant (F(1,13)=11.96, p=0.004, partial η2 =0.479). The effect of picture type
and the interaction effect were not significant. Due to a significant effect of error
type, further post hoc analyses were carried out. Results were, as predicted; visual
errors were elicited by action pictures (9.57%) significantly more than by object
pictures (1.79%) (t(13)=3.108, p=0.008). However, contrary to the prediction,
action pictures elicited more semantic errors (6.43%) than object pictures (5.43%),
but the difference was not significant.
The distribution of error types differs between fluent versus non-fluent aphasics, as
illustrated in Table 1. The most prominent error type among non-fluent aphasics
was omission comprising 79.7% of the total errors, compared to only 7.2% among
fluent aphasics; the difference was significant (t(12)=6.11, p<0.0001). Among fluent
aphasics, visual errors were the most prominent type, compromising 33% of the
total errors, compared to 9.7% of the total errors among non-fluent aphasics, but
the difference was not significant. Semantic errors were not associated with one
114 ALYAHYA
aphasia group, as the difference between fluent and non-fluent aphasics was not
significant.
Table 1: Number (%) of errors made in object and action pictures naming among fluent and
non-fluent aphasics organized according to error type.
Error type Aphasia group Picture type

Fluent Non-fluent Objects Actions
Omission 7 (7.2) 394 (79.9) 167 (75.2) 234 (63.6)
Semantic-coordinate 15 (15.5) 23 (4.7) 23 (10.4) 15 (4.1)
Semantic-superordinate 13 (13.4) 1 (0.2) 2 (0.9) 12 (3.3)
Semantic-associative 16 (16.5) 10 (2.0) 8 (3.6) 18 (4.9)
Phonological 3 (3.1) 8 (1.6) 9 (4.1) 2 (0.5)
Visual errors 32 (33) 48 (9.7) 11 (5) 69 (18.7)
Circumlocution 9 (9.3) 2 (1.4) 1 (0.5) 10 (2.7)
Un-related 2 (2.1) 7 (1.4) 1 (0.5) 8 (2.2)
Discussion
This study revealed that object pictures did not elicit any error type significantly
more than action pictures, whereas action pictures elicited more visual errors
compared to object pictures. These results are compatible with the suggestion that
action pictures are susceptible to greater complex error types than object pictures
[3].
The findings also showed that the distribution of error types in picture naming
varies among individuals with fluent versus non-fluent aphasia. Fluent aphasics
produced different types of errors with no single prominent type, whereas among
non-fluent aphasics, omission was the only prominent type, and the production
of other errors was very minimal. This has an important clinical implication, as it
advocates the use of different naming approaches including hierarchy and cueing
when treating individuals with fluent aphasia versus those with non-fluent aphasia.
This study also showed that action pictures elicited a greater number of semantic,
phonological, visual, circumlocution and unrelated errors than object pictures,
which could be in line with the postulation that language processing differs
between object-naming and action-naming [3], and this might suggest that action-
naming gives rise to greater processing demands cross-linguistically, including a
highly inflected language for both nominal and verbal systems such as Arabic.
References
1. Bastiaanse, R., & Jonkers, R. (1998). Verb retrieval in action naming and spontaneous
speech in agrammatic and anomic aphasia. Aphasiology,12(11), 951-969.
ARABIC-SPEAKING APHASICS: ANALYSIS OF NAMING ERRORS 115
2. Luzzatti, C., Raggi, R., Zonca, G., Pistarini, C., Contardi, A., & Pinna, G.-D.
(2002). Verb-noun double dissociation in aphasic lexical impairments: The role
of word frequency and imageability. Brain and Language, 81(1-3), 432-444.
doi:10.1006/brln.2001.2536
3. Matzig, S., Druks, J., Masterson, J., & Vigliocco, G. (2009). Noun and verb differences
in picture naming: Past studies and new evidence. Cortex, 45(6), 738-758.
4. Goodglass. H., & Kaplan, E (1972). The Assessment of Aphasia and Related Disorders.
Philadelphia, PA: Lea and Febiger.
5. Butterworth, B., Howard, D., & Mcloughlin, P. (1984). The semantic deficit in aphasia:
The relationship between semantic errors in auditory comprehension and picture
naming. Neuropsychologia, 22(4), 409-426.
6. Budd, M., Kortte, K., Cloutman, L., Newhart, M., Gottesman, R., Davis, C., Hillis, A.
(2010). The nature of naming errors in primary progressive aphasia versus acute post-
stroke aphasia. Neuropsychology, 24(5), 581–589. doi: 10.1037/a0020287
7. Druks, J., & Masterson, J. (2000). An Object and Action Naming Battery. Hove, UK:
Psychology Press.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 116-119
A fragile category: Turkish evidential source

markers in agrammatism and bilingualism
Seçkin Arslan1,4 , Dörte de Kok2 & Roelien Bastiaanse3
1 Graduate School for Behavioral and Cognitive Neuroscience (BCN),
The Netherlands
2 Groningen Center of Expertise for Language and Communication Disorders,
University of Groningen, The Netherlands

3 Center for Language and Cognition Groningen (CLCG); University of Groningen,
The Netherlands
4 International Doctorate for Experimental Approaches to Language and Brain
(IDEALAB),University of Groningen, The Netherlands; University of Potsdam,

Germany; University of Trento, Italy; University of Newcastle upon Tyne, UK;
Macquarie University, Australia
Introduction
Evidentiality is the grammatical marking of reference to the information source,
that is, how and from where an event was known. In reference to past events,
the following evidential usages are marked in Turkish: direct perception–DI (seen
past),reportativetextbf–ImIş (heard past), and inference–mIş (inferred past). Studies
in agrammatic aphasia have shown that reference to the past is vulnerable (i.e.,
Bastiaanse, 2013; Bastiaanse et al., 2011; Yarbay-Duman & Bastiaanse, 2009;
Gavarró & Martínez-Ferreiro, 2007). Some recent studies have demonstrated that
past time reference violations are processed at higher cost than their present
counterparts (Bos et al., 2013; Dragoy et al., 2012). Taken together these data, they
suggest that past time reference is a particular case not only in agrammatic aphasia
but also in normal language processing. An interesting notion of Turkish past time
reference is the encoding of reference to the information source. One has to choose
between one of the evidential usages described above while referring to past events.
Arslan et al., (in progress) studied seven individuals with agrammatic aphasia
(mean age 43, four females) and age-and-gender matched non-brain-damaged
control subjects. They were asked to observe 30 events with three modes of
presentation, given either by observing events as whole, their final states, or
someone else’s report about them. In the production task, the participants were
asked to complete the given sentences with an appropriate evidential marker for
each event. In a subsequent information source recall task, they were asked
to listen to the items from the production task and choose the correct mode
of information acquisition. Results showed that agrammatic speakers had more
difficulties in producing sentences in seen past than inferred past and heard
past. Moreover, they had selective deficiency in recalling the correct source.
The performance in production and source recall was controversial; the easiest
TURKISH EVIDENTIAL MARKERS IN AGRAMMATISM AND BILINGUALISM117
condition in production was the most difficult one in source recall. One interesting
outcome from the source recall task was that the agrammatic speakers pointed to
visual perception as the source when the original source was either inference or
report. In other words, they claimed to have seen events that they did not see
(Arslan et al., in progress).
In the current study, we dealt with how evidentials are processed by Turkish-
Dutch bilinguals and Turkish monolinguals. It is proposed that evidentials form
a very sensitive category that is affected not only in aphasic but also in bilingual
individuals. To reveal how bilingualism may affect evidentiality, we conducted an
auditory sentence processing study. We expect that bilinguals are less sensitive to
violations to reference to information source comparably to monolinguals.
Methods
26 monolingual Turkish speakers (mean age 29, fourteen females) without
proficient level of any second language and 21 early bilinguals of Turkish-Dutch
(mean age 19, four females) participated in the study. The material consisted of
120 experimental sentences, comprised of 60 sentences with seen past (a), and 60
sentences with heard past (b).
(a) Yerken gördüm, az once adam yemegi yedi. Hemde nasil ...
Eat-while see-seen past just before man food-acc eat-seen past also how eat-INF surprise-Past1sg
‘Previously I saw the man eating, he ate the food. I was surprised how he ate.’
(b) Yerken görmusler, az once adam yemegi yemis. hemde tabaklari...
Eat-while see-seen past-3rdPl just before man food-acc eat-heard past also plates-acc wash-Neg-PST
‘Previously they saw the man eating, he ate the food. He didn’t even wash the dishes.’
Half of the sentences in each condition included an inappropriate source marking

resulting in a violation. Seen Past presupposes the observer is “I”, and similarly,
Heard Past tends to go with third person more often; if not, the sentences sound
unusual or unnatural. The experimental sentences were mixed with fillers (with
relative clause violations) and presented auditorily. Participants were instructed to
press the space bar as soon as they noticed a violation. Therefore, accuracy and
reaction time could be recorded.
Results
Anovas were done with the factors of group (bilingual vs. monolingual) and
condition (seen past vs. heard past; violation vs. non-violation). The interactions
between group and condition were significant for reaction times: F(3,966)=5.211
p=.001; and for accuracy of hits: F(3, 2812)=119.118 p=.000. We observed an
overall latency in responses of bilinguals and they were less accurate in noticing
violations compared to monolinguals. In seen and heard past, bilinguals correctly
judged only 32% of the violations in each condition. The reaction times of these
32% hits were little over 3000ms for seen past and 2700ms for heard past. However,
monolinguals were quite robust in detecting violations (seen past: 90%; heard
past: 87%) and they were faster than bilinguals (see Figure 1). Monolinguals
118 ARSLAN ET AL.
showed a difference in judging violations: seen past (1600ms vs. 1810ms). The
slower processing of heard past sentences was confirmed by a t-test: t(683)=-2,355;
p=.019. Such a difference was not significant in bilinguals.

Figure 1: Reaction Times (ms) and Accuracy of Hits (%) in response to source marking
violations by Monolingual and Bilingual speakers
Discussion
The studies we have presented here have several implications for the theory how
evidentials are affected in agrammatism and bilingualism. Firstly, in agrammatic
aphasia, the production of seen past was severely impaired while heard and
inferred pasts were well-retained. Evidential markers were shown to be affected in
bilingualism as well: bilingual speakers were quite inaccurate and reacted slower
than monolinguals. We, therefore, argue that evidential markers are quite sensitive
and vulnerable not only in agrammatic aphasia, but also in bilingualism.
The obligatory marking of evidentials in Turkish seems to be affecting the language
processing in the monolingual speakers of Turkish. Particularly, the monolingual
speakers show slower processing in heard past. Comparably, the bilingual speakers
who are under dominance of a non-evidential language do not show such an
evidential effect.
TURKISH EVIDENTIAL MARKERS IN AGRAMMATISM AND BILINGUALISM119
References
Arslan, S., Aksu-Koç, A., Maviş, I., & Bastiaanse, R. (In Progress). Turkish past time reference:
Verb inflection and source recall errors in agrammatic Broca’s aphasia.
Bastiaanse, R. (2013). Why reference to the past is difficult for agrammatic speakers. Clinical
Linguistics & Phonetics, , 1-20.
Bastiaanse, R., Bamyaci, E., Hsu, C., Lee, J., Duman, T. Y., & Thompson, C. K. (2011).
Time reference in agrammatic aphasia: A cross-linguistic study. Journal of
Neurolinguistics, 24(6), 652-673. doi: 10.1016/j.jneuroling.2011.07.001
Bos, L. S., Dragoy, O., Stowe, L. A., & Bastiaanse, R. (2013). Time reference teased apart from
tense: Thinking beyond the present. Journal of Neurolinguistics, 26(2), 283-297. doi:
10.1016/j.jneuroling.2012.10.001
Dragoy, O., Stowe, L. A., Bos, L. S., & Bastiaanse, R. (2012). From time to time: Processing
time reference violations in Dutch. Journal of Memory and Language, 66(1), 307-325.
doi: 10.1016/j.jml.2011.09.001
Duman, T. Y., & Bastiaanse, R. (2009). Time reference through verb inflection
in Turkish agrammatic aphasia. Brain and Language, 108(1), 30-39. doi:
10.1016/j.bandl.2008.09.009
Gavarró, A., & Martínez-Ferreiro, S. (2007). Tense and agreement impairment in Ibero–
Romance. Journal of Psycholinguistic Research, 36(1), 25-46.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 120-122
The effects of (in)direct speech on aphasic

discourse comprehension
Rimke Groenewold1,2 , Roelien Bastiaanse1 ,
Lyndsey Nickels2 & Mike Huiskes1
The Netherlands
2 ARC Centre of Excellence in Cognition and its Disorders (CCD),
Macquarie University, Australia
Introduction
In conversation, direct reported speech (e.g., ‘John said: “I’m hungry!”’) is assumed
to constitute a demonstration of a reported utterance, whereas its indirect reported
speech counterpart (e.g., ‘John said that he was hungry’) provides a description
of what was said (Clark & Gerrig, 1990). The distinction between direct and
indirect speech exists in many languages and has been a major focus in linguistic
studies. Direct speech constructions are perceived as more vivid and perceptually
engaging than their indirect speech counterparts (Wierzbicka, 1974; Macaulay,
1987). In this study, we examine the effects of direct speech on aphasic discourse
comprehension in Dutch. The additional communicative “layers” (e.g., intonation,
facial expression, and gesture) that go along with direct speech may facilitate
language comprehensibility. In addition, its grammatical characteristics may
contribute to the comprehensibility of speech. Direct speech is distinguished from
indirect speech in that the pronouns, spatial and temporal references, and verb
tenses are appropriate to the reported context rather than the current one (Holt,
1996). In addition, in Dutch, indirect speech requires a subordinate construction,
whereas direct speech does not. Since individuals with aphasia are known to
have difficulties with subordinate constructions (Menn & Obler, 1990; Bastiaanse &
Jonkers, 1998; Bastiaanse, Hugen, Kos & Van Zonneveld, 2002), Dutch direct speech
constructions may be easier to comprehend than indirect speech constructions.
Methods
Participants
The aphasic subgroup consisted of 24 Dutch individuals (19 male) with mild to
moderate aphasia. Criteria for selection of the individuals with aphasia were (1)
medical diagnosis of brain damage, (2) no audiologically or medically documented
hearing impairment, (3) diagnosis of aphasia by a speech pathologist using
standardized tests, and (4) time post-onset ≥3 months. The individuals with
aphasia ranged in age from 41 to 82 years (M=57.4, SD=13.5).
The Non Brain Damaged (NBD) subgroup consisted of 16 individuals (7 male)
(IN)DIRECT SPEECH IN APHASIC DISCOURSE COMPREHENSION 121
who were matched for age, gender, and educational level to the aphasic subgroup.
Criteria for selection of the NBD subjects were (1) no documented history of brain
damage, and (2) no audiologically or medically documented hearing impairment.
NBD subjects ranged in age from 35 to 76 years (M=53.4, SD=12.0).
Materials and procedures

Each subject was tested in a single session of approximately 50 minutes for the
aphasic and 25 minutes for the NBD subjects. The aphasic participants were
administered the Token Test to get an indication of the aphasia severity.
DIRECT SPEECH COMPREHENSION TEST
Both the aphasic and the NBD subjects performed the iPad-based Direct Speech
Comprehension (DISCO) Test. The test, which was developed specifically for this
study, consists of 1 practice and 6 target videos, during which short stories are
told. The target stories can be subdivided into 3 story lines each with two stories,
one using direct speech and the other using indirect speech. All participants were
presented with both conditions of a story line (i.e., direct speech and indirect
speech), without hearing the same story twice. This design allowed us to draw
comparisons both within and between groups.
The stories had an average length of 217 words (SD=22.8), 19 utterances (SD=1.9),
and an average Flesch Reading Ease Score of 80 (SD =2.1), indicating that they were
(very) easy to understand. To rule out the effect of order, 12 different presentation
lists were created.
After each of the stories the participants heard 8 questions, which they could
answer with “yes” or “no” touching a response button that appeared on the screen.
This method ruled out possible confounds from language production difficulties.
The DISCO scores reflect the proportion of correctly answered questions per
condition type (0-1.0).
Results
For all participants the proportion of correctly answered items per story was
calculated, resulting in 6 scores per participant. The aphasic subgroup (n=24)
had an average score of 0.80 (SD=0.10), and the NBD subgroup’s (n=16) mean
score was 0.90 (SD=0.06). In order to examine the effects of group (aphasic, NBD)
and condition type (direct, indirect), we conducted an ANOVA using a repeated
measures design. There was a significant main effect of listener type: the NBD
group performed better than the aphasic group, F(1, 38) = 12.18, p = .001. In
addition, there was a significant main effect of condition, F(1, 38) = 4.22, p
< .05. A paired t-test split for groups showed that in the case of the aphasic
subgroup there was a significant effect of condition type: they scored better on the
direct speech condition (M=0.83, SD=0.13) than on the indirect speech condition
(M=0.77, SD=0.11), t(23) = 2.27, p = .03. No such effect was found for the NBD
subgroup, t(15) = .74, p = .47. A negative correlation between the Token Test
122 GROENEWOLD ET AL.
and DISCO scores was found (r = -.67, n = 24, p = .00), indicating that preserved
comprehension is associated with high DISCO scores.
Discussion
Previous studies have suggested that direct speech constructions may facilitate
language comprehensibility since they are perceived as more vivid than their
indirect speech counterparts. In this study, the effects of direct speech
constructions on aphasic discourse comprehension were examined. The
experimental design allowed us to make direct comparisons between the
comprehensibility of stories told using direct speech and those with indirect
speech. For the aphasic subgroup we found an effect of condition type: the stories
that were told using direct speech proved easier to comprehend than the stories
with indirect speech. A possible explanation for this finding is the occurrence
of additional “layers” of communication that often accompany direct speech
constructions, such as intonation and facial expression. Another possible account
is the difference in grammatical complexity: in Dutch, unlike direct speech, indirect
speech requires subordinate constructions, which are known to be difficult for
particularly agrammatic aphasic individuals. A repetition of this study in English
will provide us with insight into the role of the grammatical characteristics of the
two construction types.
References
Bastiaanse, R. Hugen, J., Kos, M. & Zonneveld, R. van (2002) Lexical, morphological and
syntactic aspects of verb production in Dutch agrammatic aphasics. Brain and
Language, 80, 142-159.
Bastiaanse R, Jonkers R. (1998). Verb retrieval in action naming and spontaneous speech in
agrammatic and anomic aphasia. Aphasiology 12(11): 951-969.
Clark, H. H., & Gerrig, R. J. (1990). Quotations as Demonstrations. Language, 66, 764-805.
Holt, E. (1996). Reporting on Talk: The Use of Direct Reported Speech in Conversation.
Research on Language and Social Interaction, 29 (3): 219-245.
Macaulay, R. (1987). Polyphonic monologues: Quoted Direct Speech in Oral Narratives. IPRA
Papers in Pragmatics, 1, 1-34.
Menn, L., & Obler, L. K. (Eds.) 1990. Cross-language data and theories of agrammatism
in Agrammatic aphasia: A cross-language narrative sourcebook (pp. 1369-1389).
Philadelphia: John Benjamins.
Wierzbicka, A. (1974). The Semantics of Direct and Indirect Discourse. Linguistics, 7, 267-
307.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 123-125
Spreading the word ‘aphasia’. New

international comparisons of the public
awareness of aphasia in Argentina, Canada,
Croatia, Greece, Norway and Slovenia
Chris Code1 , Ilias Papathanasiou2 , Silvia Rubio-Bruno3 ,
María de la Paz Cabana3 , Maria Marta Villanueva3 , Line Haaland-Johansen4 ,
Tatjana Prizl-Jakovac5 , Ana Leko5 , Nada Zemva6 , Ruth Patterson7 ,
Richard Berry7 , Elizabeth Rochon8 , Carol Leonard9 & Amelie Robert9
1 University of Exeter, UK
2 Technological Educational Institute of Patras, Greece
3 Fundacióm Argentina de Afasia, Argentina
4 Bredtvet Resource Centre, Oslo, Norway
5 University of Zagreb, Croatia
6 Rehabilitation Institute, Slovenia
7 March of Dimes York Durham Aphasia Centre Canada
8 University of Toronto, Canada
9 University of Ottawa, Canada
Introduction
The public awareness of aphasia is vital for extending services, research support,
social inclusion (Elman et al., 2000) and, importantly, for targeted awareness
raising. Studies to date (Code et al., 2001; Simmons-Mackie et al., 2002) have
showed that knowledge of aphasia varies across a range of variables, but is low
compared to comparable conditions like Parkinson’s disease.
Methods
Convenience samples were surveyed in face-to-face interviews in shopping
centres, parks, libraries, stations, etc. in cities in Argentina (N=800), Canada (832),
Croatia (400), Greece (N=800), Norway (N=251) and Slovenia (N=400) using the
same (suitably translated) questionnaire used in previous studies (Code et al, 2001;
Simmons-Mackie et al., 2002) requesting information on age, gender, occupation,
whether respondents had heard of aphasia and where they had heard. Those
who selected from features that were (e.g., ‘speech problems’, ‘language problems’,
‘communication problems’) and were not (e.g., ‘impaired intelligence’, ‘mental’
problems) features of aphasia and also noted that aphasia follows ‘brain damage’
were classified as having some basic knowledge of aphasia, and were questioned
further to determine how and where they had learnt about aphasia.
124 CODE ET AL.
Results
Mean age of the entire sample was 43.16 (SD 17.68). Between 57.4% (Norway)
and 20% (Argentina) had heard of aphasia (37.1% overall) but those with basic
knowledge ranged from 13.9% (Norway) to 1% (Argentina). The combined mean
percentage of those with a basic knowledge of what aphasia is, was 9.2%. Those who
had heard of aphasia were significantly younger (p<.0001) and females had higher
levels of awareness (Chi Sq=9.65; df=2; p=.008). Of those with basic knowledge,
only Greeks were significantly older (t=4.868; df=798; p<.0001).
Those who had knowledge of aphasia gained their knowledge from occupational
exposure to aphasia, because a relative or friend had aphasia and through the
media (TV, radio, newspapers, magazines, etc.).
Discussion
We surveyed public places because we wanted to tap the knowledge of those
who come into contact with aphasic people in shops, banks, restaurants, etc.
(Code, 2003). Levels of awareness were predictably low, but there was significant
variability between most countries. Figures contrast in some respects with
Simmons-Mackie et al who found between 9.25% and 18% of their English-
speaking samples had heard of aphasia (13.6% overall), while those with basic
knowledge ranged from 1.54% to 11.53 (combined 5.42%). The earlier study too
found age and gender interacted significantly with knowledge.
Interactions between socio-economic variations and cross-cultural comparisons
will be examined and the significance of the results for campaigns to raise the
public awareness of aphasia will be discussed and compared with earlier results.
We stress that caution must be applied in generalizing these findings to entire
national and international populations and examine ways in which different
countries, and regions or cities within countries, can utilise locally surveyed levels
of understanding of aphasia to plan ways to improve levels of understanding of
aphasia to improve community access and involvement for aphasic people. We also
describe our development of a new dedicated Aphasia Awareness Website which
is due to be launched by the National Aphasia Association (www.aphasia.org) in
February, 2013 and is designed to provide data on international levels of awareness
for professionals engaged with aphasia, the general public, aphasic people and
their families and, importantly, for the media.
References
Code, C. (2003) The quantity of life for people with chronic aphasia. Neuropsychological
Rehabilitation, 13, 365-378.
Code, C. Simmons-Mackie, N., Armstrong, E., Stiegler, L., Armstrong, J., Bushby, E., Carew-
Price, P., Curtis, H., Haynes, P., McLeod, E., Muhleisen, V., Neate, J., Nikolas, A., Rolfe,
D., Rubly, C., Simpson, R. and Webber, A. (2001) The public awareness of aphasia:
INTERNATIONAL COMPARISONS OF THE PUBLIC AWARENESS OF APHASIA
125
an international survey. International Journal of Language and Communication

Disorders 36, 1-6.
Elman, R. Ogar, J. & Elman, S., 2000, Aphasia: Awareness, advocacy, and activism.
Aphasiology 14, 455-459.
Simmons-Mackie, N., Code, C., Armstrong, E., Stiegler, L. & Elman, R. (2002) What is
Aphasia? Results of an international survey. Aphasiology, 16, 837-848.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 126-128
Construction and validation of a

speech-systematic aphasia screening (SAPS)
and its appendent therapy regimen
Stefanie Abel1,2 , Franziska Krzok2 , Verena Chwalek2 , Katharina Niemann2 ,
Ruth Nobis-Bosch3 , Irmgard Radermacher2 , Walter Huber2 & Klaus Willmes1
1 Division Neuropsychology, Department of Neurology, Medical Faculty,
RWTH Aachen University, Germany

2 Division Clinical and Cognitive Neuroscience, Department of Neurology,
Medical Faculty, RWTH Aachen University, Germany

3 Department for Advanced Training,
Deutscher Bundesverband für Logopädie (dbl) e.V., Germany
Introduction
Cognitive-linguistic therapies have been recommended as a Practice Standard
for the remediation of language and communication deficits, with the approach
favorably being accompanied by participation-based approaches and potentially
being assisted by supervised computer-based training (Cicerone et al., 2000).
Thus, a standardized aphasia test that, in one go, covers the multitude of speech-
systematic impairments and directs subsequent therapy would be of high value.
Moreover, the involvement of participation-based procedures and computer-based
training would be desirable.
Thus, we constructed a speech-systematic aphasia screening (SAPS) which
comprises the psycholinguistic language components of phonetics/phonology,
lexicon/semantics, and morphology/syntax at increasing degrees of demands and
in both receptive and expressive modalities, in order to direct and evaluate a
therapy regimen based on resulting performance profiles (see Fig. 1). Moreover,
we created a computer-based home training that builds on the screening tasks and
that is performed in combination with SAPS-based therapy.
In the present study, we aimed to (i) validate an already optimized SAPS
version as well as the home training procedure, (ii) determine therapy effects,
and (iii) consider correlations between patient performance in SAPS and in a
communicative-pragmatic test.
Methods
We included 16 patients with a mean age of 50;5 years (range 25;5 - 67;11; ten
male, six female) of the Aachen Aphasia Ward. 13 patients were in the chronic
(>12 months post-onset) and three patients in the post-acute (1-12 months) phase
of recovery. All four standard syndromes were present, with ten patients having a
non-fluent and six a fluent aphasia.
CONSTRUCTION AND VALIDATION OF SAPS AND ITS THERAPY REGIMEN127
Figure 1: Speech-systematic aphasia screening (SAPS): Overview

∗Two goals of a certain level and modality are chosen based on the patient performance
profile (i.e. selection among 18 specific goals). Each receptive task entails 24, each expressive
task 16 items, half of which are trained in each goal (i.e., trained and untrained item sets).
The computer-based home-training consisted of multi-modal stimulations of half

of the items of a SAPS task (i.e., 12 items in receptive tasks and 8 ones in expressive
tasks); the other half served as untrained control items. For each patient, two
impairment-specific therapy goals were chosen according to the individual SAPS
profile, and the according task and therapy items were part of the therapy sessions
and again trained in home-training sessions, with each session lasting 1 hour per
day. The therapy regimen lasted for a maximum of 16 treatment days. Patients
were administered the SAPS and the Amsterdam-Nijmegen Everyday Language
Test (ANELT) (Blomert et al., 1997) before and after the therapy regimen.
Results
As a result, the properties of our test construction largely could be confirmed:
The screening tasks of each language component and modality were well graded
in difficulty (univariate analysis of variance, factor: degree of demands, all p ≤
.003), and all but two tasks yielded high estimates of reliability (Cronbachs Alpha
> 0.9). The computer-based training also proved its value regarding construction
and feasibility. The therapy regimen was effective for nearly all modality-specific
128 ABEL ET AL.
language components and tasks as featured in SAPS (exact Wilcoxon-test, one-

tailed, p < .05). Given published data on critical differences (Kawalla, 2011), the
understandability (A-) score of the ANELT improved significantly in three patients.
Moreover, ANELT performance correlated with expressive lexical and morpho-
syntactic SAPS tasks (Spearman correlations, range: rS = .698 - .904, one-tailed,
all p < .01).
Discussion
To conclude, the newly developed screening allows the assessment of an individual
performance profile and the derivation of psycholinguistically defined therapy
goals. Thus, the screening can serve as an instrument to measure therapy outcomes
both in clinical settings and in randomized controlled trials (RCTs) for patients
with aphasia. Moreover, SAPS-based therapy and the associated computer-based
home training have shown to be effective, and expressive SAPS performance in part
correlates with communicative-pragmatic performance.
References
Blomert, L., Koster, C., & Kean, M.L. (1997). Amsterdam-Nijmegen-Everyday-Language-Test
(ANELT). Lisse, NL: Swets Test Services.
Cicerone, K. D., Dahlberg, C., Kalmar, K., Langenbahn, D. M., Malec, J. F., Bergquist, T. F. et
al. (2000). Evidence-based cognitive rehabilitation: Recommendations for clinical
practice. Archives of Physical Medicine and Rehabilitation, 81, 1596-1615.
Kawalla, M. (2011). The evaluation of speech-pragmatic abilities using the ANELT -
An empirical study on therapy effectiveness at the Aachen Aphasia Ward (English
translation). Rheinisch-Westfälische Technische Hochschule Aachen, Germany.
Unpublished Master thesis.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 129-131
Language acquisition, learning and dissolution

Dorota Leśniak
The Jan Kochanowski University in Kielce, Poland
Introduction
It was Roman Jakobson (1941/68) who probably first presented a comprehensive
theory related to the acquisition of phonological oppositions in a systematic way.
Many Jakobson’s ideas were challenged by later research but his main ideas are still
valid. Jakobson focused his attention on child language development and language
dissolution in aphasia. He indicated that these two processes are based on the same
rules.
The discovery that language develops in a special, general order and the fact that
language breaks down according to special rules let us develop many ideas in terms
of how to work with babies who acquire language and with aphasic patients in
order to help them to use language again. During the years of studies new ideas
appeared. One of them is that language problem in aphasia can be connected
with the access to the components of language but not only with the components
themselves (Linebarger, Schwartz and Saffron 1983).
Although language is a highly complex system, most children learn their mother
tongue in a natural way. Biological, psychological, and social factors seem to be the
most important in this area. For the author it is important to check if the factors
work in the same way for those who suffer from aphasia. The most important seems
to be the issue of phonological opposition, mainly because as Jakobson stated, it is
not the phoneme itself but their oppositions that bear the meaning (Jakobson 1968,
68).
The first experiment investigated the acquisition of English oppositions by Polish
young learners. The main idea states that acquisition of oppositions in L2 is similar
to L1 but not the same.
Methods
The participants of the experiment attended primary school. Their phonological
system of first language was shaped and they distinguished Polish oppositions
perfectly.
To check the hypothesis Blache’s test (1975) was used. The oppositions that were
estimated by the author as not important were not included in the test.
The participants were presented with a set of minimal pairs that had been read by
a native speaker and recorded. The next step was to play them to the participants
who listened to them through the headphones. The participants decided which
words they heard and repeated them. The opposition pairs that were repeated by
the children were recorded and then it was decided which of them were correct
130 LEŚNIAK
and which were not. It is important to stress that, in this experiment, the most
important was the acquisition of phonological oppositions but not the correct
pronunciation of the segments.
Results
The analysis of the experiment
During the experiment it turned out that the order of acquisition of phonological
oppositions in L2 is not exactly the same as in L1 . It can suggest that the sound
discrimination process that was developed during the acquisition of the first
language sometimes facilitates the development of oppositions of L2 . In many
situations the children did not go through the exact order of the acquisition of the
oppositions suggested by Jakobson. The oppositions that do not exist in Polish like
/θ / and /ð/ were distinguished by the children, although the pronunciation was
not perfect. It was also noticed that the front sounds presuppose the existence of
the back ones. There is also explanation why Polish speakers devoiced voiced stops
in word-final position in English.
Discussion
The author connects her ideas of language restitution in aphasia with the research
she conducted in terms of the acquisition of English phonology by Polish young
learners, mainly phonological oppositions. She wants to check if the results are
similar, the same or different in the field of phonological oppositions in L2 of
the patients who recover from aphasia and compare them with L1 phonological
oppositions. In this project she presents three main questions:
• if language recovery goes parallel in L1 and L2 or
• L2 is recovered before L1 or
• L1 is recovered before L2 .
This attitude seems to be important because nowadays more and more people
communicate in more than two foreign languages and it is essential to know how
to work with these patients. That is why author raises more additional questions
that she is going to answer. Some of the most important are:
• is the recovery from aphasia in terms of phonological acquisition the same in

different languages that were used by one person who is at least bilingual?
• is the recovery in L2 the same or similar to L1 ?
The child is not able to control vocal organs until he/she is about six months
old (Kaplan and Kaplan 1971). The sounds that appear are rather accidental in
character. There some more questions arise:
LANGUAGE ACQUISITION, LEARNING AND DISSOLUTION 131
• is the situation similar to that in aphasia?
• are the sounds that appear at the beginning in the process of language
recovery only accidental in character or are they ordered in a special way?
When we learn language after the period of puberty our right hemisphere is more
engaged in the process of learning. However, L2 is influenced by L1 in both ways:
positively and negatively since transfer works in this way. Because during the
process of L2 learning the right hemisphere is more involved it may suggest that
after stroke, when the left hemisphere is affected, it is possible that L2 can work. It
should be stressed, however, that the brain functions as a whole and we can only
talk about domination of the hemisphere. In this context it would be helpful in
language recovery to check phonological oppositions in L2 of bilingual patients:
• how they work when a patient acquired L2 ,
• how they work when he/she learnt L2 ,
• are in L2 preserved those oppositions that are distinctive also in L1 or the

oppositions that have distinctive features only in L2 .
It is beyond this paper to present these issues in more details. However, the author
is willing to talk with all interested in these problems.
References
Jakobson, R. 1972. Child Language, Aphasia and Phonological Universals. The Hague:
Mouton (originally published, 1941, as Kindersprache, Aphasie und allgemeine
Lautgesetze)
Kaplan, E. & G. Kaplan. 1971. The prelinguistic child. In J. Eliot (ed.). Human Development
and Cognitive Processes. New York: Holt, Rinehart, and Winston. 358-8
Linebarger, M. C., Schwartz, M., and Saffran, E. 1983. Sensitivity to grammatical structure in
so-called agrammatic aphasia. Cognition, 13, 361-92.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 132-135
Processing pseudo-words in mild cognitive

impairment: On-line and off-line evidence
from Slovenian
Christina Manouilidou1 , Barbara Dolenc2 , Tatjana Marvin3 ,
Katarina Marjanovič3 & Zvezdan Pirtošek2
1 University of Patras, Greece
2 University Medical Centre, Ljubljana, Slovenia
3 University of Ljubljana, Slovenia
Introduction
The term Mild Cognitive Impairment (MCI) refers to a condition between normal
aging and dementia which is often seen as prodromal for dementia (Chertkow,
2002). When MCI individuals demonstrate impairments in other-than-memory
domains, including language, they are more likely to develop dementia than are
those with a pure memory impairment (Petersen, 2003). Thus, understanding
the nature of language impairment and possibly identifying sensitive measures of
linguistic impairment constitutes a vital tool in early detection of dementia. While
there exists plentiful evidence of language deficits in MCI mainly from standardized
assessment tools (for a review see Taler & Phillips, 2008), psycholinguistic
studies of language processing are scarce. The few studies that have employed
psycholinguistic methodology have revealed disturbances in performance mainly
at the lexical-semantic level (Olichney et al, 2002; Puregger et al, 2003; Davie et al,
2004; Taler & Jarema, 2004; 2006; Duong et al, 2006) reflecting an impaired semantic
network in this population. At the same time, structural aspects of language,
namely phonological and morphological structure as well as syntax are thought to
show no alterations. One important dimension of previous research is that, with
few exceptions, most studies have employed off-line measurements, thus, only
targeting “controlled” and not automatic processing.
In this context, the present study examines aspects of both controlled and
automatic lexical processing in MCI patients by looking at their performance in
differentiating pseudo-words which appear to have different patterns of violability.
The goal of the study is to present data regarding the boundaries of lexical
representations and their decay in this population, thus contributing to the
establishment of the nature of linguistic deficits seen in MCI. Moreover, by
employing both off-line and on-line chronometrized tasks we attempt to detect
differences between MCI and healthy populations in more subtle aspects of
pseudo-word processing.
PROCESSING PSEUDO-WORDS IN MILD COGNITIVE IMPAIRMENT 133
Methods
Participants
So far six healthy volunteers (all females, aged 60-79) and eight individuals with
MCI (6 females, 2 males, aged 55-82) participated in the study. Testing is still
being carried out targeting a total of 15 participants for each group. Patients
were recruited from the Neurology Clinic of the University Medical Centre in
Ljubljana. All of them were diagnosed by a neurologist or a neuropsychologist
at the Neurology Clinic. In addition, their performance was examined by means
of MMSE and MoCA translated and adapted versions for Slovenian. All patients
performed below the proposed cut-off scores, indicating the presence of cognitive
disturbances.
Materials comprised 3 groups of words violating certain constraints of word
formation in Slovenian (B-D), one group of unattested (due to blocking) possible
words without violations (E), one group of real words (F) and one group of non-
words (A). All were formed with a masculine-gender nominal (‘-er’) suffix. Materials
were based on a normative study for Slovenian (Marjanovic et al, 2013)
A. Non-Words based on non-existing stems and existing suffixes (*lastje,

*dovina, ‘conper-er’) (n=30)
B. PseudoWs violating grammatical category constraints of the base (*črkilec

‘letter-er’) (n=30)
C. PseudoWs violating thematic constraints of the base (*počivalec ‘rest-er’)

(n=22)
D. PseudoWs violating aspectual constraints of the base (*preplavalec (from

preplavati ‘to swim-perfective’)) (n=30)
E. Possible unattested Ws without violations (*kuhalec (possible but blocked

by kuhar ‘cook’)) (n=30)
F. Real Words (plavalec (’swimmer’) (n=30)
All stems of pseudo-words were matched on average for frequency and the actual
pseudo-words were also matched for length and number of syllables.
Study 1 - Off-line acceptability judgment task: For each item, participants had
to say “yes” or “no”, answering the question “In your opinion, is this word part of
Slovenian vocabulary?”. If their answer was “yes”, they were also asked to provide
the meaning of this word.
Study 2 - On-line lexical decision task: The experiment was run on an IBM
computer using E-prime professional. Stimuli were presented at the center of a
134 MANOUILIDOU ET AL.
computer screen in black font on a white background and were randomized for
each participant. Each item was preceded by a row of hashmarks that remained on
the screen for 200 ms, and a pause of 150 ms.
Results
Off-line task: Percentages of correct responses (“Yes” for words, “No” for all
pseudo-words) were calculated for each group of words for patients and controls.
Patients’ data were similar to controls’ with the exception of pseudo-words with
aspectual violations (group D) where there is a tendency for MCI patients to accept
them as words of Slovenian more often than the controls (p=0.11). Percentages are
displayed in Table 1.
Table 1: Off-line and on-line results for both groups of participants
Off-line acceptability task: Mean percentages of correct responses

Non-words Categorical Thematic Aspectual Possible Real Words
MCI 94 95 87 78 62 97
Control 97 97 87 66 61 97
On-line lexical decision task: Mean RTs in ms of correct responses

Non-words Categorial Thematic Aspectual Possible Real Words
MCI 1891 1926 1955 2327 2255 1205
Control 1298 1501 1674 2242 1802 1101
On-line lexical decision task: Accuracy (percentages of correct responses)

Non-words Categorial Thematic Aspectual Possible Real Words
MCI 93 90 68 50 54 99
Control 97 95 77 58 61 98
On-line task: Mean RTs in ms and accuracy results are displayed in Table 1.
In terms of RTs, a 2(group) x 6(word type) mixed model ANOVA has revealed a
main effect of word type while post-hoc tests revealed a significant difference
between MCI & control when it comes to non-words (p=0.02) and to possible
(p=0.05), while the difference between RTs for other types of pseudo-words are
substantial but did not reach significance. An interesting observation derives from
the comparison between the accuracy data obtained from the two tasks for MCI
patients. Namely, MCI patients accepted significantly more pseudo-words with
aspectual violations (p=0.01), thematic violations, (p=0.04) and possible (p<0.001)
in the on-line task than in the off-line, revealing an important task effect.
PROCESSING PSEUDO-WORDS IN MILD COGNITIVE IMPAIRMENT 135
Discussion
Study 1 showed that patients seem to have maintained the ability to detect
violations, which further suggests the preservation of word formation rules.
However, although their performance was strikingly similar to that of healthy
controls, the higher percentages in pseudo-words with aspectual violations (group
D) might suggest that the boundaries of their lexical representations are becoming
loose and their lexicon more flexible. Study 2 confirmed in a more robust way the
differences between the two populations revealing that when automatic processing
is required, MCI patients fail to behave as healthy participants do. In other words,
patients perform within normal range when there is no time pressure, but do
worse under time pressure revealing a reduced speed of processing, which is in
accordance with their deficit. This task effect is suggestive of the potential that on-
line studies provide for the detection of risk groups for dementia.
References
Chertkow, H. (2002). Mild cognitive impairment. Current Opinion in Neurology, 15, 401-407.
Davie, J. E., Azuma, T., Goldinger, S. D., Connor, D. J., Sabbagh, M. N., & Silverberg, N.
B. (2004). Sensitivity to expectancy violations in healthy aging and mild cognitive
impairment. Neuropsychology, 18, 269-275.
Duong, A., Whitehead, V., Hanratty, K., & Chertkow, H. (2006). The nature of lexico-semantic
processing deficits in mild cognitive impairment. Neuropsychologia, 44, 1928-1935.
Marjanovič, K., Manouilidou C., & T. Marvin (2013). Word-Formation Rules in Slovenian
Agentive Deverbal Nominalization: A Psycholinguistic Study Based on Pseudo-
Words. Slovene Linguistic Studies 9: 93-109.
Olichney, J. M., Morris, S. K., Ochoa, C., Salmon, D. P., Thal, L. J., Kutas, M., et al. (2002).
Abnormal verbal event related potentials in mild cognitive impairment and incipient
Alzheimer’s disease. Journal of Neurology, Neurosurgery and Psychiatry, 73, 377-384.
Puregger, E., Walla, P., Deecke, L., & Dal-Bianco, P. (2003). Magnetoencephalographic
features related to mild cognitive impairment. NeuroImage, 20, 2235-2244.
Petersen, R. C. (2003). Mild cognitive impairment. New York: Oxford University Press.
Taler, V., & Jarema, G. (2004). Processing of mass/count information in Alzheimer’s disease
and mild cognitive impairment. Brain and Language, 90, 262-275.
Taler, V., & Jarema, G. (2006). On-line lexical processing in AD and MCI: An early measure of
cognitive impairment? Journal of Neurolinguistics, 19, 38-55.
Taler, V. & Phillips, N.A. (2008). Language performance in Alzheimer’s disease and mild
cognitive impairment: A comparative review. Journal of Clinical and Experimental
Neuropsychology, 30, 501-556.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 136-138
The role of the dorsal pathway in primary

progressive aphasia
Anna Martínez-Álvarez
Universitat de Barcelona, Spain
Introduction
The study of fronto-temporal brain pathways is commonly associated to different
Primary Progressive Aphasia (PPA) subtypes (Gorno-Tempini 2011; Mesulam
et al 2012). The present works aims at exploring how a dissociation of two
different dorsal pathways provides a unique and useful way to dissociate two
different patterns within the heterogeneous group of agrammatic/nonfluent
Primary Progressive Aphasia (G-PPA). To do so, the connection between language
areas in the frontal and temporal lobe are taken into account, and linguistic and
speech processes implemented by distinct neural pathways are considered.
Current neurocognitive models of language processing (Friederici 2002; Hickok
and Poeppel 2007) provide a framework for a functional neuroanatomy of syntactic
and lexical processes. More specifically, neurolinguistic reseach has shown that
Broca’s region consists of three different areas (Friederici et al. 2003) which are
connected with the temporal lobe by a dorsal and a ventral route (Anwander et al.
2007; Catani et al. 2005; Glasser and Rilling 2008; Saur et al. 2008). This work claims
that the three different variants of PPA are associated to linguistic operations rooted
in distinct fronto-temporal networks in the left hemisphere. More concretely, our
main focus is the PPA population and their linguistic and neuroanatomical distinct
profiles.
The present research focuses on the particular case of non-fluent/agrammatic
PPA individuals. Interestingly enough, patients need to show at least two of the
following criteria in order to be included in this subtype: (1) motor speech deficits;
(2) agrammatism in language production; (3) spared single-word comprehension
and impaired comprehension of only the most complex syntactic structures.
Therefore, motor speech deficits and agrammatic production are not necessarily
both found in order to classify patients into this variant.
Methods
The present contribution focus on two different types of agrammatic/nonfluent
PPA deficits, more specifically, AOS and agrammatism, which we claim to relate to a
pattern of damage associated to two different dorsal pathways. To do so, we carried
out a literature review, including pathology driven alterations as well as data from
DTI studies on PPA population, establishing a novel correlation between specific
types of language and speech deficits found in G-PPA individuals and the neural
pathways underlying their different occurrence.
THE ROLE OF THE DORSAL PATHWAY IN PRIMARY PROGRESSIVE APHASIA137
Results
Recent studies using DTI method have shown that patients with PPA differ from
normal controls with respect to brain pathways structure (Agosta et al., 2011;
Galantucci et al. 2011). Comparing the three PPA variants, different connectivity
patterns for each PPA variant have been found (Galantucci et al. 2011). In non-
fluent/agrammatic PPA individuals, the ventral tract is found to be spared while the
dorsal tract is damaged (Galantucci et al. 2011; Wilson et al. 2010). These findings
have been claimed to show that the syntactic processing depends mainly on dorsal
tracts, since there seems to be a correlation between damage to in the superior
longitudinal fasciculus and deficits in syntactic comprehension and production.
In a recent study, Josephs et al. (2012) have studied the occurrence of apraxia
of speech (AOS), characterized by impaired planning or programming of the
movements for speech in degerative diseases. In these diseases, AOP and aphasia
often co-occur, in which cases, patients are usually diagnosed as having PPA. Our
contribution follows the lines of Josephs et al. (2012), suggesting that a separation
of AOS from aphasia is plausible. The clinical presentation they refer to is based
on a progressive neurological disorder dominated by AOS in which there is no
evidence of agrammatism in their spoken language or in their narrative writing
performance. They refer to these individuals as having primary progressive AOS,
or primary progressive apraxia of speech (PPAOS).
Interestingly enough, subjects with PPAOS had focal pattern of grey matter atrophy
affecting the lateral premotor and supplementary motor area. These findings
point towards the premotor cortex as the possible neuroanatomical correlate of
PPAOS (Josephs et al., 2012). The association between premotor cortex and PPAOS
syndrome is plausible due to the fact that the premotor area is said to be involved
in motor programming. These findings suggest that a dissociation of the two
dorsal pathways would be useful for a better understanding of both syndromes and
anatomical correlates of G-PPA.
Discussion
The present work claims that, once AOS and agrammatic deficits are dissociated, a
window to their neural correlates is open. More concretely, we propose that while
damage to the Dorsal Pathway I running from the posterior superior temporal
gyrus to the premotor cortex may be associated to AOS syndrome, damage to
the Dorsal Pathway II connecting the posterior superior temporal gyrus to pars
opercularis (BA44) may give rise to agrammatic deficits. In sum, we claim that the
heterogeneos G-PPA variant can be further subdivided into two different clinico-
pathological variants once a dissociation of the two dorsal pathways is considered.
The present research focuses on the structural connectivities underlying the neural
network of language processing and language breakdown. A dissociation of
language deficits (e.g. PPA) vs. speech disorders (e.g. AOS) and a dissociation of
the two dorsal patways provides a unique and useful way to approach linguistic
138 MARTÍNEZ-ÁLVAREZ
deficits in PPA individuals. This work also contributes to a further understanding of

the Language Faculty implemented in the human brain and sheds some light to the
study of the language processing from which language acquisition and language
breakdown fields are based on.
References
Agosta, F., E. Scola, E. Canu, A. Marcone, G. Magnani, L. Sarro, M. Copetti, F. Caso, C.
Cerami, G. Comi, S. F. Cappa, A. Falini and M. Filippi (2011). White Matter Damage in
Frontotemporal Lobar Degeneration Spectrum. Cerebral Cortex. October 10. 1-10.
Anwander, A., M. Tittgemeyer, D.Y. Von Cramon, A.D. Friederici, T.R. Knosche (2007).
Connectivity-based parcellation of Broca’s area. Cerebral Cortex 17, 816-825.
Catani, M., D.K. Jones & D.H. Ffytche (2005). Perisylvian language networks of the human
brain. Annals of Neurology 57, 8-16.
Friederici, A.D (2002). Towards a neural basis of auditory sentence processing. Trends in
Cognitive Sciences 6, 78-84.
Friederici, A.D., S.A. Rüschemeyer, A.Hahne & C.J. Fiebach (2003). The role of left
inferior frontal and superior temporal cortex in sentence comprehension: localizing
syntactic and semantic processes. Cerebral Cortex 13(2), 170-177.
Galantucci, S., M.C. Tartaglia, S.M. Wilson, M.L. Henry, M. Filippi, F. Agosta, N.F. Dronkers,
R.G. Henry, J.M. Ogar, B.L. Miller, M.L. Gorno-Tempini (2011). White matter damage
in primary progressive aphasias: a diffusion tensor tractography study. Brain, 134,
3011-3029.
Glasser, M.F. & J.K. Rilling (2008). DTI Tractography of the human brain’s language pathways.
Cerebral Cortex 18, 2471-2482.
Gorno-Tempini, M. L., A.E. Hillis, S. Weintraub; A. Kertesz, M. Mendez; S.F. Cappa, J.M.
Ogar, J.D. Roher, S. Black, B.F. Boeve, F. Manes, N.F. Dronkers, R. Vandenbergue,
K. Rascovsky, K. Patterson, B.L. Miller, D.S. Knopman, J.R. Hodges, M.M. Mesulam,
M. Grossman (2011). Classification of primary progressive aphasia and its variants.
Neurology 76, 1-9.
Hickok, G. & D. Poeppel (2007). The cortical organization of speech perception. Nature
Reviews Neuroscience 8, 893-402.
Josephs, K.A., J.R. Duffy, E.A. Strand, M.M. Machulda, M.L. Senjem, A.V. Master, V.J. Lowe,
C.R. Jack Jr. & J.L. Whitwell (2012). Characterizing a neurodegenerative syndrome:
primary progressive apraxia of speech. Brain 135(5), 1522-1536.
Mesulam, M.M, Christina Wieneke, Cynthia Thompson, Emily Rogalski and Sandra
Weintraub (2012). Quantitative classification of primary progressive aphasia at early
and mild impairment stages. Brain 135, 1537-1553.
Saur, D., B.W. Kreher, S. Schnell, D. Kümmerer, P. Kellmeyer, M.S. Vry, R. Umarova, M. Musso,
V. Glauche, S. Abel, W. Huber, M. Rijntjes, J. Hennig, C. Weiller (2008). Ventral and
dorsal pathways for language. Proceedings of the National Academy of Sciences 105,
18035-18040.
Wilson, S.M., N.F. Dronkers, J.M. Ogar, J. Jang, M.E. Growdon, F. Agosta, M.L. Henry, B.L.
Miller, & M.L. Gorno-Tempini (2010). Neural Correlates of Syntactic Processing in
the Nonfluent Variant of Primary Progressive Aphasia. The Journal of Neuroscience
30(50), 16845-16854.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 139-141
Fronto-temporal pathways and paraphasias

Anna Martínez-Álvarez1 & Silvia Martínez-Ferreiro2
1 Universitat de Barcelona, Spain
Introduction
With the spread of neuroimaging techniques, the existence of pure cortical lesions
has been proved, at best, as anecdotic (Lieberman 2002, Dronkers et al. 2004).
The participation of subcortical structures and, more specifically, neural pathways
turns out to be crucial for the clarification of the complex profiles that the analysis
of acquired deficits tends to provide. Although aphasic syndromes are thought to
be explained away by the classic model of the language network, the dual stream
model of Hickok and Poeppel (2004) combined with tractography results and
data from direct brain stimulation provides arguments in favour of the existence
of a main dorsal subcortical pathway underlying the syntactic system, and a
main ventral subcortical pathway underlying the semantic system within the left
hemisphere (Catani et al. 2005; Saur et al. 2008; Glasser & Rilling, 2008, Friederici
2011).
The present contribution aims at exploring the viability of an extended version of
the dual stream model (Hickock & Poeppel 2007) to account for specific symptoms
of acquired and degenerative forms of aphasia. To do so, we explore the occurrence
of paraphasias traditionally associated to Wernicke’s aphasia and cortical lesions
affecting the temporal lobe (Dell et al. 1997, Fridriksson et al. 2012). While
semantic paraphasias, i.e. the production of words that are semantically linked
to the target word (e.g., chair-table), are traditionally associated to a failure at
the lexical-semantic level, disruptions in the access to the phonological output
lexicon may underlie the appearance of phonemic paraphasias, i.e. non-words
that are phonologically related to the target word (Kay & Ellis, 1987; Biran &
Friedmann, 2005). The present work aims at proving that the proper account
of language processing and language breakdown does not only benefit from but
crucially depends on white matter tracts connecting frontal and temporal language
areas.
Methods
We focus on two different types of paraphasias, more specifically phonological
and semantic paraphasias, which we claim to relate to deficits associated to the
dorsal and ventral circuitry respectively. To do so, we carried out a literature review,
including pathology driven alterations as well as data from direct brain stimulation,
establishing a correlation between specific types of paraphasias and the neural
pathways underlying their occurrence.
140 MARTÍNEZ-ÁLVAREZ & MARTÍNEZ-FERREIRO
Results
The two sources of data analyzed, more specifically abnormalities caused by
electric stimulation and the results from the analysis of cases of primary
progressive aphasia (PPA) shed similar results. According to Suzuki (2012),
electrical stimulation of the arcuate fasciculus induced phonological paraphasias,
whereas the stimulation of the inferior occipitofrontal fasciculus induced semantic
paraphasias. Similar results are reported by Mandonnet et al. (2007) and Duffau et
al. (2009).
Phonological paraphasias are also found as a consequence of logopenic primary
progressive aphasia and non-fluent primary progressive aphasia (Gorno-Tempini
et al. 2011; Mesulam et al. 2012; Joseph et al. 2012). Crucially for the present
proposal, semantic paraphasias are never attested in these deficits. Contrary to
this picture, the presence of semantic paraphasias is common in cases of semantic
dementia (Ogar et al. 2011; Reilly, 2008). Crucially, both non-fluent and logopenic
PPA variants have been related with damage to the superior longitudinal fasciculus
including its arcuate component (Wilson et al. 2011; Galantucci et al. 2011), while
the semantic PPA variant has been related to damage to the ventral pathway, more
concretely, to the extreme capsule fiber and the uncinate fasciulus (Agosta et al.
2010; Galantucci et al. 2011).
Discussion
Studies on degenerative pathologies and language acquisition have shown that
both lesions and developmental stages of subcortical pathways can predict
linguistic behavior (Whitwell et al. 2010, Agosta et al. 2011, Galantucci et al.
2011, Friederici et al. 2011; Brauer et al. 2011). According to recent language
models in the brain, dorsal pathways are associated to sound-to-motor mapping
and higher-level language processes, ventral pathways are related to sound-to-
meaning mapping and lexical semantics (Hickok & Poeppel 2004, Friederici et al.
2006, Friederici 2011).
The present research focuses on the structural connectivity underlying the neural
network of word retrieval, showing how word retrieval proceeds in different
stages that may fail independently. The dissociation of brain damage patterns in
phonemic and semantic paraphasias, once the neural networks connecting frontal
and temporal language regions are taken into account, provides a unique and
useful way to approach language processing.
References
Agosta, F., M. Pievani, S. Sala, C. Geroldi, S. Galluzzi, G.B. Frisoni, M. Filippi (2011). White
matter damage in Alzheimer disease and its relationship to gray matter atrophy.
Radiology, 258, 853-863.
Brauer, J. A., Anwander & A.D. Friederici (2011). Neuroanatomical presequisites for language
functions in the maturing brain. Cerebral Cortex 21, 459-466.
FRONTO-TEMPORAL PATHWAYS AND PARAPHASIAS 141
Catani, M., D.K. Jones & D.H. Ffytche (2005). Perisylvian language networks of the human
brain. Annals of Neurology 57, 8-16.
Dronkers, N.F., D.P. Wilkins, R.D. Van Valin Jr., B.B. Redfern & J.J. Jaeger (2004). Lesion
analysis of the brain areas involved in language comprehension. Cognition 92, 145-
177.
Friederici, A.D. (2002). Towards a neural basis of auditory sentence processing. Trends in
Cognitive Sciences 6, 78-84.
Friederici, A.D., J. Bahlmann, S. Heim, R.I. Shubotz & A. Anwander (2006). The brain
differenciates human and non-human grammars. Proceedings of the National
Academy of Sciences 103, 2458-2463.
Friederici, A.D. (2011). The brain basis of language processing: from structure to function.
Physiological Reviews 91, 1357-1392.
Friederici, A.D., J. Brauer, & G. Lohmann (2011). Maturation of the language network:
From inter- to intrahemispheric connectivities. PloS ONE, 6(6), e20726.doi:
10.1371/journal.pone.0020726.
Galantucci S., M.C. Tartaglia, S.M. Wilson, M.L. Henry, M. Filippi, F. Agosta, N.F. Dronkers,
R.G. Henry, J.M. Ogar, B.L. Miller, M.L. Gorno-Tempini (2011). White matter damage
in primary progressive aphasias: a diffusion tensor tractography study. Brain, 134,
3011-3029.
Glasser, M.F. & J.K. Rilling (2008). DTI Tractography of the human brain’s language pathways.
Cerebral Cortex 18, 2471-2482.
Hickok, G. & D. Poeppel (2004). Dorsal and ventral streams: a framework for understanding
aspects of the functional anatomy of language. Cognition 92, 67-99.
Hickok, G. & D. Poeppel (2007). The cortical organization of speech perception. Nature
Reviews Neuroscience 8, 893-402.
Josephs, K.A., J.R. Duffy, E.A. Strand, M.M. Machulda, M.L. Senjem, A.V. Master, V.J. Lowe,
C.R. Jack Jr. & J.L. Whitwell (2012). Characterizing a neurodegenerative syndrome:
primary progressive apraxia of speech. Brain 135(5), 1522-1536.
Lieberman, P. (2002). On the nature and evolution of the neural bases of human language.
American Journal of Physical Anthropology, 119(S35), 36-62.
Mesulam, M. Marsel, Christina Wieneke, Cynthia Thompson, Emily Rogalski and Sandra
Weintraub (2012). Quantitative classification of primary progressive aphasia at early
and mild impairment stages. Brain 135; 1537-1553. doi:10.1093/brain/aws080.
Ogar, J.M., J.V. Baldo, S.M.Wilson, S.M. Brambati, B.L. Miller, N.F. Dronkers, M.L.
Gorno-Tempini (2011). Semantic dementia and persisting Wernicke’s
aphasia: Linguistic and anatomical profiles. Brain & Language 117, 28-33.
doi:10.1016/j.bandl.2010.11.004.
Saur, D., B.W. Kreher, S. Schnell, D. Kümmerer, P. Kellmeyer, M.S. Vry, R. Umarova, M. Musso,
V. Glauche, S. Abel, W. Huber, M. Rijntjes, J. Hennig, C. Weiller (2008). Ventral and
dorsal pathways for language. Proceedings of the National Academy of Sciences 105,
18035-18040.
Wilson, S.M., S. Galantucci, M.C. Tartaglia, K. Rising, D.K. Patterson, M.L. Henry, J.M. Ogar, J.
DeLeon, B.L. Miller & M.L. Gorno-Tempini (2011). Syntactic processing depends on
dorsal language tracts. Neuron 72, 397-403.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 142-144
Conversational turn length and fluency

measurement in aphasia
Verónica Moreno-Campos1 & Beatriz Gallardo-Paúls2
1 Florida Universitària, Spain
2 University of Valencia, Spain
Introduction
A common assumption regarding fluency is that the difference between a fluent
and non-fluent speaker can be easily stated (Poeck, 1989; Gordon, 1998). However,
there is no objective and valid measure to determinate the level of a person with
aphasia on the fluency continuum. Traditionally, people with aphasia have been
classified as fluent or non-fluent following the cognitive criteria (Uribe, Arana &
Lorenzana, 1969; Goodglass & Kaplan, 1986; Kertesz, 1994; Price et al., 2003).
The present study has attempted to clarify differences between fluent and non-
fluent patterns of speech using analysis data from natural conversation settings.
Methods
Data collection
In order to guarantee validity of linguistic data, they should be collected in
their natural conversational frame, as demanded from clinical practice (Penn,
1985; Ahlsén, 1995; Joanette & Ansaldo, 1999; Perkins, 2005; Gallardo-Paúls,
2009). Conversations by 30 bilingual people with aphasia (Spanish- Catalan)
talking with their key conversational partners (Withworth, Perkins & Lesser, 1997)
were analyzed and compared with interactions between ‘non damaged’ bilingual
(Spanish- Catalan) speakers in order to identify which variable can be relevant for
the fluent/ non-fluent diagnosis.
Data analysis
All analyzed conversations were fragments of 20 minutes chosen at random from
conversation of one hour length. Fluency measurement has to be developed which
can be adapted to the different types of discourse and their components; at the
same time, the formula used has to indicate where the patient in the continuum
of fluency is. The formula evolved was: speaker total words / speaker total speech
turns.
Results
Analyzed the 60 conversation fragments, the average number of words was 1,795
and the average number of turns was 134,33; so the average number of words per
turn was 12,89. The Standard deviation (SD) was 5,59. As it has to be considered
TURN LENGTH AND FLUENCY MEASUREMENT IN APHASIA 143
that fluency is not a dichotomous property of the discourse, but a continuum,

it implies that the rating between -1SD and +1SD can be considered the norm.
Only the ratings more or less than 2SD can be considered as pathological. For
this reason, we have fixed the -1SD (7,3) as the value to differentiate fluent aphasic
speakers from non-fluent.
Discussion
Results of the quantitative analysis allow us to consider that 7,3 words-per-turn
value is the measure to delimit fluent and non-fluent speakers. Our proposed value
adds real conversation data to what we know from less-conversational data context.
It supports other proposals we can find in BDAE-3, where Goodglass, Kaplan and
Barresi (2000) talk about 7-word utterances as the difference to consider a speaker
normal or fluent. Analysis presented in Hernández-Sacristán & Rosell (2009) agree
as well with the measure presented in this work.
These results emphasize the importance of the quantitative analysis of fluency
in speech in its natural environment. The established value of 7,3 words-per-
turn allows a specific evaluation about the fluency level of an aphasic speaker in
their natural linguistic production. Taking into account this natural evaluation the
diagnosis will be closer to the real state of the patient and the rehabilitation will
produce better results. As well, the measure of 7,3 words-per-turn not only can
determinate the difference between fluent and non-fluent speaker, but allows the
diagnosis of severe fluency deficits as logorrhea or mutism.
References
Ahlsén, E. (1995). Pragmatics and Aphasia. An Activity based approach, Monográfico 77
Gothenburg Papers in Theoretical Linguistics.
Gallardo Paúls, B. (2009). Valoración del componente pragmático a partir de datos orales,
Revista de Neurología, 48 (4), 57-61.
Goodglass, H. & Kaplan, E. (1986). Evaluación de la afasia y trastornos relacionados. Madrid:
Editorial Médica Panamericana. Trad. de Carlos Wernicke. Adaptación de J. E.
García-Albea, M. L. Sánchez Bernardos y S. del Viso.
Goodglass, H.; Kaplan, E. & Barelli, B. (2000). Boston Diagnostic Aphasia Examination-Third
Edition (BDAE-3). San Antonio, TX: The Psychological Corporation.
Gordon, J. (1998). The fluency dimensión in aphasia, Aphasiology, 12 7/8, 673-688.
Hernández Sacristán, C. & Rosell, V. (2009). Syntax and conversation in aphasia. A strategic
restrictive use of Spanish and Catalan connector QUE by aphasic speaker, Clinical
Linguistics and Phonetics, 23 (10), 717-741.
Joanette, Y. & Ansaldo, A.I. (1999). Clinical Note: Acquired Pragmatic Impairments and
Aphasia, In Stemmer, B. (Ed.): Pragmatics: Theoretical and Clinical Issues, Brain and
Language, 68,(3), 529-534.
Kertesz, A. (1994). Localization and Neuroimaging in Neuropsychology. San Diego:
Academic Press
Penn, C. (1985). The Profile of Communicative Appropiateness: A clinical tool for the
144 MORENO-CAMPOS & GALLARDO-PAÚLS
assessment of pragmatics, The South African Journal of Communication Disorders,

32, 18-23.
Perkins, M. (2005). Clinical pragmatics: an emergentist perspective, Clinical Linguistics and
Phonetics, 19 (5). pp. 363-366 . DOI: http://dx.doi.org/10.1080/02699200400027080
Price, C.J., Gorno-Tempini, M.L., Graham, K.S., Biggio, N., Mechelli, A., Patterson, K., &
Noppeney, U. (2003). Normal and pathological reading: converging data from lesion
and imaging studies. Neuroimage, 20 (Suppl. 1), S30-S41.
Poeck, K. (1989). Fluency. In C. Code (Ed.). The characteristics of Aphasia (Philadelphia:
Taylor & Francis) p. 23-32.
Uribe, Arana y Lorenzana (1969). Neurología. Medellín, Colombia: Corporación para
investigaciones biológicas.
Whitworth, A.; Perkins, L.; y Lesser, R. (1997). Conversational Analysis Profile for People with
Aphasia CAPPA, London: Whurr Publ.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 145-148
Language mixing in discourse in bilinguals with

aphasia
Avanthi Paplikar1 , Mira Goral1,2,3,4 ,
Martin Gitterman1,4 & Loraine K. Obler1,2,3
1 The Graduate Center of the City University of New York, USA
2 Language in the Aging Brain Lab, VA Boston Healthcare System, USA
3 Boston University School of Medicine, USA
4 Lehman College, City University of New York, USA
There is substantial literature reporting differences in connected language

production in aphasia depending on the elicitation tasks, including differences
in narrative length and sentence complexity (Goral, 2012; Olness 2006). Little is
known about the relationship between language mixing in bilingual aphasia and
the type of task used to elicit the narrative.
In India, language mixing is not uncommon in bilingual and multilingual
individuals. That is, during communication framed in any one language, bilingual
or multilingual speakers mix in various units from a second language, such as
words or morphemes, for effective communication (Bhat & Chengappa, 2003;
Sebastian, Dalvi & Obler, 2012). For bilingual and multilingual individuals with
aphasia, there are studies reporting use of at least two languages when patients
are asked to speak in only one language (Aglioti & Fabbro, 1993; Fabbro, Skrap, &
Aglioti, 2000), but the extent to which bilinguals with aphasia mix their languages
inappropriately is still under debate. Therefore, in the current study we asked the
following questions:
1. Does the amount of language mixing in bilingual individuals with non-fluent

aphasia differ in two different types of discourse: personal narratives and
picture sequence description?
2. Are bilingual individuals with non-fluent aphasia capable of modifying their

language usage based on the monolingual or bilingual status of the listener?
Participants
Five bilingual or trilingual (Kannada-English-Hindi) male participants with non-
fluent aphasia were tested. All the participants were of mild to moderate severity
of aphasia with good comprehension in all their languages. The participants
had an average age of 46 years (29-62 years) and average education of 14.2 years
(12-16 years). The language history and degree of bilingualism was determined
using a questionnaire on language proficiency. Participants were asked to rate
themselves on a 5-point rating scale (0: virtually nothing and 4: excellent) on their
understanding, speaking, reading and writing abilities before and after the stroke in
146 PAPLIKAR ET AL.
Kannada and English (for the four bilinguals) and, in Kannada, English, and Hindi
(for one trilingual). Participants whose average score was 3 or more for their first
(L1) and second language (L2), and 2 or less for their third language (L3) both pre-
and post-stroke on the language questionnaire were selected for the present study.
Participant selection, test administration and language transcription were done by
the first author, a trilingual (Kannada-English-Hindi) certified Speech Language
Pathologist.
Procedure
All the participants also filled out a questionnaire on language mixing. In this
questionnaire, after an explanation of what language mixing is they were asked
questions such as, ‘Do your family members mix languages while speaking to
you?’, ‘What is the preferred direction of language mixing your family uses¿ etc.
The participants received a battery of discourse tasks and a test of language
impairment (WAB). The first four sub-tests of the WAB (Spontaneous Speech
(information content and fluency), Auditory Comprehension, Repetition, and
Naming) were administered in both Kannada and English. The discourse tasks
included three descriptions of two six-picture sequences (Bilingual Aphasia Test-
BAT picture sequence and Husband-Wife Fight), one eight-picture sequences
(Cycle-Car Incident), and personal narratives (about the stroke, a vacation, and an
event when they were happy).
The picture-sequence description and the personal narrative tasks were
administered in three conditions: (A) Monolingual Kannada condition, (B)
Monolingual English condition and (C) Bilingual/Trilingual Kannada-English-
Hindi condition. In each condition, participants were shown a brief video clip of a
person they were told would be the person their discourse responses were being
recorded for. In the first condition the participants viewed a short video clip of
a monolingual Kannada speaker. After the video clip a still photo of the listener
remained and the examiner provided the following instructions: ‘Make up your
own story about what is happening in the picture, with a beginning, a middle,
and an end, only using your first language, i.e., Kannada’. In the second condition
the video clip was of a monolingual English speaker and the instructions were
modified mutatis mutandis. In the last condition, the video clip was of a trilingual
Kannada-English-Hindi speaker and the examiner instructed the participants that
they could use any of their languages for this listener. The first time they mixed in
a word from another language, participants were reminded once to use only L1 or
only L2 in conditions A and B respectively.
Participant responses were orthographically transcribed from the recordings by the
examiner. Each word in the transcript was coded as Kannada, English, or Hindi.
The words were further coded as language-mixed or borrowed words (borrowed
words being words like English ‘car’ that should be considered Kannada as they
are part of the daily speech of non-brain-damaged individuals and may even be
LANGUAGE MIXING IN DISCOURSE IN BILINGUALS WITH APHASIA 147
Table 1: Averages of number of words in each language in Picture Sequence Description and
Personal Narratives. LM: Language Mixed Words; K: Kannada; E: English; H: Hindi
LM in Picture Sequences LM in Personal Narrative

Condition A - K: 28.8 K: 20.4
Kannada Monolingual E: 7.8 E: 7.2
Condition B - K: 2.06 K: 7.6
English Monolingual E: 48.8 E: 34.13
Condition C - K: 25.13 K: 27.7
Kannada-English Bilingual E: 19.5 E: 13.7
H: 0.46
inflected with Kannada affixes, e.g., ‘caru’ for nominative). The number of words
from each language for each discourse type was tallied (Table 1).
Results
The Mann-Whitney U test was used for analysis due to the small N. Not only was
there no significant difference in the number of language-mixed words between
the two discourse tasks; the base-language word usage was also not significantly
different between the two discourse tasks. As well, mixing was greatest in the non-
monolingual condition, suggesting that individuals with aphasia are able to modify
their mixing appropriately for the listener.
Conclusion
It appears that language mixing during narrative production in bilingual
individuals with aphasia is similar regardless of the elicitation task. Additionally,
individuals with aphasia are able to appropriately modify their language mixing
behavior based on the listener, albeit not flawlessly.
References
Aglioti, S., & Fabbro, F. (1993). Paradoxical selective recovery in a bilingual aphasic following
subcortical lesions. NeuroReport, 4, 1359-1362.
Bhat, S., & Chengappa, S. (2005). Code switching in normal and aphasic Kannada-English
bilinguals. Proceedings of the 4th International Symposium on Bilingualism. 306-316.
Fabbro, F., Skrap, M., & Aglioti, S. (2000). Pathological switching between languages after
frontal lesions in a bilingual patient. Journal of Neurology Neurosurgery Psychiatry.
68, 650-652.
Goral, M. (2012). Cross-language treatment effects in multilingual aphasia. In M. Gitterman,
M. Goral, and L.K.Obler (Eds.), Aspects of Multilingual Aphasia. Bristol, UK:
Multilingual Matters, 106 - 121.
148 PAPLIKAR ET AL.
Olness, G.S. (2006). Genre, verb, and coherence in picture elicited discourse of adults with
aphasia. Aphasiology, 20 (2/3/4), 175-187.
Sebastian, D., Dalvi. U., & Obler, L. K. (2012). Language deficits, recovery patterns and
effective intervention in a multilingual 16 years post-TBI. In M. Gitterman, M. Goral,
and L.K.Obler (Eds.), Aspects of Multilingual Aphasia. Bristol, UK: Multilingual
Matters 122-138.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 149-152
When object clitisation and climbing happen

alone, and when they dance cheek to cheek:
Selective impairment in Spanish agrammatism
Andrés Felipe Reyes1 & Roelien Bastiaanse2
1 El Bosque University, Colombia
Introduction
Previous studies have shown that production of clitics in agrammatic Broca’s
aphasia is compromised, either by analyzing spontaneous speech (Nespoulous
et al., 1988; Reznik et al.,1995; Stavrakaki and Kouvava, 2003; Chinellato, 2004;
Rossi, 2007) or by testing clitic production in an experimental setting (Rossi, 2007;
Gavarró, 2008, Nerantzini, 2008; Martínez-Ferreiro, 2010). Clitics in agrammatism
have only been explored in Greek, Italian, French and Ibero-Romance (Spanish,
Catalan and Galician). In Spanish, the study reported by Reznik et al. (1995)
explicitly focused on the production of clitics in aphasic spontaneous speech,
while Martínez-Ferreiro (2010) tested clitic production and comprehension in an
experimental setting. This study focuses on the morpho-syntactic problems of
Spanish agrammatic speakers with emphasis on the production of sentence word
order. Because of the particular flexible word order with which Spanish can be
grammatically produced, and due to the predictions that can be drawn from
the ‘Derived Order Problem Hypothesis’ (DOP-H) (Bastiaanse & van Zonneveld,
2005), the present study focuses on two movement operations, clitic and object
scrambling, in a way that has not been explored before. It is hypothesized
that syntactic complexity, in a linguistic sense, is a critical factor in agrammatic
production, and, therefore, it is predicted that sentences with object movement
and clitic movement will be more difficult than sentences with basic word order
(SVO), regardless of the position in the syntactic tree.
Methods
Participant Age Time Post-Onset Etiology City
(years)
A1 34 8 Hematoma due to Barcelona
diffuse vasculitis
A2 59 3 CVA Barcelona
150 REYES ET AL.
Stimuli
A set of 64 semantically reversible target sentences with transitive verbs and
animate subjects and objects were used, the first 16 employed by Bastiaanse,
Edwards and Rispens (2002) in the development of the Verb and Sentence Test
(VAST), and another 48 sentences derived from them. The set included four target
sentence types: (a) active declarative sentences with a transitive (finite) lexical
verb, that is assumed to stay in its base-generated position in English and Spanish
(the “full object in base position” condition; e.g., “The man carries the woman -
El hombre carga a la mujer”) (n=16); (b) active declarative sentences with a clitic
pronoun, assumed to move from post-verbal to preverbal position (e.g., “The man
la carries - El hombre la carga”) (n=16); (c) active declarative sentences with a finite
lexical verb in which the full object moves to preverbal position (e.g., “The man
to the woman carries - El hombre a la mujer carga”) (n=16); and (d) imperative
sentences with a clitic pronoun in the postverbal position (e.g., “Man carriesla -
Hombre cárgala”) (n=16). In all the sentences the subject and object differed in
gender to avoid correct answers in the second (clitic) condition due to repetition,
as in Spanish the use of third person singular clitics depends on the gender they
refer to (lo-la refer to male or female respectively). For testing each sentence or
item, black-and-white picture pairs developed by Bastiaanse et al. (2002) were
used: the first picture in the pair depicts the prompt sentence and the other depicts
its semantically reversed counterpart, so the two types of clitics (lo, la) were tested
equally as half of the items have a male subject performing the action in the second
part of the phrases. A set of 64 picture pairs (16 original pairs and 3 copies of each)
was used to elicit the three sentence conditions mentioned above. Therefore, the
16 original picture pairs can elicit the 64 types of target sentences. The picture on
the left always depicted the first sentence, and the picture on the right depicted the
sentence which participants had to complete. The order of the items was pseudo-
randomly assigned (each picture occurred only once in the first 16 items, once in
the second 16 items, etc.) but the order was the same for each participant.
Figure 1: Taken from Bastiaanse et al. (2002), with permission.

OBJECT CLITISATION AND CLIMBING IN SPANISH AGRAMMATISM 151
Procedure
Using each pair, the 4 sentence conditions previously described were elicited using
a sentence production priming task in the following way: the experimenter pointed
to the first picture of the pair reading aloud the prompt sentence, after which the
experimenter pointed to the second one, reading only the subject of the sentence
aloud. The participant was expected to complete the sentence depicted in the
second picture using the same sentence structure the examiner used before (e.g.,
the examiner said: “Here the man carries the woman, and here the woman...”, and
according to the 3 sentence conditions, the participant responded: (a) “carries
the man”, (b) “lo (male Spanish clitic) carries”, (c) “to the man carries”), or (d)
“carrieslo”. The experiment was carried out in a quiet room. The participant
sat in front of a computer screen and the experimenter sat next to him/her, and
the experimental procedure was explained to the participant. There were three
trial sentences (the last trial will be different from the first experimental sentence
type). In case the task is not clear, any questions were allowed and additional trial
sentences were added until the task was understood by the participant. All the
responses were transcribed for analysis.
Results
Table 1: Correct sentence production per brain-damaged participant. Row percentages
Condition n A1 A2 A3 A4 A5
(a) -mov -clitic 32 100 96.875 100 84.375 93.75
(b) +mov +clitic 32 15.625 6.25 100 0 0
(c) +mov -clitic 32 0 0 0 0 0
(d) -mov +clitic 32 59.375 3.125 46.875 0 0
The first sentence condition was easier to produce than the other three (p <
0.0001 Fisher’s exact test at both overall and at the individual level, except among
conditions (a) and (b) in the participant A3).
Discussion
The present data supports the hypothesis that it is difficult to move the object,
whether it is a clitic or a full DP.
It is shown that movement and clitics can be affected selectively.
References
Anderson, R. & Centeno, J. (2007). Contrastive Analysis Between Spanish and
English. In: Communication Disorders in Spanish Speakers: Theoretical,
Research and Clinical Aspects. Ed: Centeno, J., Anderson, R. & Obler, L.
152 REYES ET AL.
From: http://books.google.com/books?id=2akmE9kiV3QC&hl=es File recovered on

February 5, 2008.
Ardila, A. (1997). Características en el español de las alteraciones adquiridas en el lenguaje.
Forma y Función, 10, 13-23. Universidad Nacional de Colombia.Bogotá, Colombia.
Bastiaanse, R. & van Zonneveld, R. (2005). Sentence production with verbs of alternating
transitivity in agrammatic Broca’s aphasia. Journal of Neurolinguistics, 18, 59-66.
Bastiaanse, R. & Thompson, C. K. (2003). Verb and auxiliary movement in agrammatic
Broca’s aphasia. Brain and Language, 84, 286-305.
Bastiaanse, R., Edwards, S. & Rispens, J. (2002). The Verb and Sentence Test (VAST). Thames
Valley Test Company.
Burchert, F., Swoboda-Moll, M., & De Bleser, R. (2005). The left periphery in agrammatic
clausal representations: Evidence from German. Journal of Neurolinguistics, 18, 67-
88.
Gavarró, A. (2008). ‘Binding and co-reference in Catalan agrammatism’. The Academy of
Aphasia Meeting, Turku, 20 October.
Martínez-Ferreiro, S. (2010). Towards a characterization of agrammatism in
Ibero-Romance. PhD thesis, Universitat Autònoma de Barcelona. From:
http://webs2002.uab.es/clt/publicacions/tesis/pdf/Martinez_Ferreiro.pdf File
recovered on January 13, 2011.
Martínez-Ferreiro, S. (2003). Verbal in_ectional morphology in Broca’s aphasia. Unpublished
M.A. thesis, Universitat Autònoma de Barcelona.
Nerantzini, M. (2008). Direct Object clitics in Greek Agrammatic Production: A single case
study. Unpublished MA thesis, University of Groningen, The Netherlands.
Pountain, C. J. (2001). A History of the Spanish Language through
Texts. London, England: Routledge. pp. 177, 264-5. From:
http://es.wikipedia.org/wiki/Posici%C3%B3n_de_pronombres_cl%C3%ADticos
File recovered on March 10, 2008.
Reznik, M., Dubrovsky, S. & Maldonado, S. (1995). Agrammatism in Spanish: A Case Study.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 153-156
Syntactic dependency resolution in Broca’s

aphasia
Maria Varkanitsa1 , Dimitrios Kasselimis2,3 , Ioannis Evdokimidis2 ,
Constantin Potagas2 , Judit Druks1 & Hans van de Koot1
1 Department of Linguistics, University College London, UK
2 Neurology Department, University of Athens, Greece
3 Psychology Department, University of Crete, Greece
Introduction
Research on sentence comprehension in aphasia has shown that individuals with
agrammatic Broca’s aphasia often exhibit a highly selective deficit in processing
intra-sentential dependencies; comprehension of sentences that contain filler-
gap dependencies (i.e. A’-Movement) is impaired, whereas comprehension of
sentences that contain Binding relations are relatively spared. This dissociation
has been attributed to the fact that there are important syntactic and processing
differences between the two dependencies (Santi & Grodzinsky, 2007a, 2012).
Syntactically, the antecedent of a reflexive (John in (1a) is in a theta position,
whereas in A’-Movement, the displaced filler (the man, that, in (1b)) is not.
(1) a. Mary knows that John1 loves himself1 Binding

b. Mary knows the man1 that1 David pinched t1 Movement
One could therefore hypothesize that the dissociation should be linked to
predictability, i.e. whether the dependencies can be identified at an early stage of
processing based on syntactic factors. In A’-Movement, the processor encounters
the filler early on and provides a warning that the reader/hearer should store the DP
in memory and go hunting for a gap (predictable dependency). In Binding no such
cues are available; the dependency becomes evident only when processing reaches
the reflexive, which is assigned its anaphoric reference in retrospect (unpredictable
dependency). The two dependencies also make different demands on Working
Memory (WM). A’-Movement puts greater demands on storage processes, but
Binding on retrieval processes (Santi & Grodzinsky, 2012).
The hypothesis that predictability is the key factor distinguishing A’-Movement
from Binding makes the further prediction that relations mediated by leftward LF
movement should behave like Binding. This prediction is supported by the finding
that agrammatic patients perform normally on ambiguous doubly quantified
sentences (e.g. A woman is photographing every child) (Saddy, 1995; Varkanitsa
et al., 2012), whose inverse scope reading involves (leftward) Quantifier Raising in
the LF component (May, 1977).
The present study investigates whether this asymmetry in predictability effects
between overt and covert leftward movement is also manifested in the processing
154 VARKANITSA ET AL.
of sentences with contrastive foci (CF) by Greek-speaking patients with Broca’s

aphasia. As shown in (2), Greek provides an appropriate minimal pair, with the CF
either moved or in situ.
(2) a. Ti GINEKA1 filai o adras t1 ohi to koritsi. Moved CF

The WOMAN1 ACC is kissing the manNOM t1 not the girl.
‘The WOMAN the man is kissing, not the girl.’
b. O andrasNOM filai ti GINEKAACC ohi to koritsi. In situ CF
The man is kissing the WOMAN not the girl
‘The man is kissing the WOMAN, not the girl.’
Methods
Participants
Four chronic Greek-speaking patients with aphasia and four non-brain-damaged
individuals participated in the study. Three patients are also agrammatic, that
is, they exhibited impaired comprehension of movement-derived sentences (i.e.
wh-questions and passives) during background testing, whereas comprehension
of simple semantically reversible sentences was well preserved. Patients’ WM was
assessed with the digit span task and the Corsi block-tapping task (forward and
backward). The agrammatic patients demonstrated WM deficits in both tasks,
whereas the non-agrammatic patient performed within normal limits.
Procedure
Participants performed a picture-selection task, consisted of two experimental
conditions and two filler conditions. The experimental conditions included
sentences with moved object-CF and sentences with in situ object-CF. Sentences
with subject-CF and simple transitives were used as fillers. Each condition included
20 semantically reversible sentences.
Results
Patients’ performance on the picture selection task, presented in Table 1, revealed
a dissociation between processing of sentences containing moved CF, as in (2a),
and sentences containing in situ CF, as in (2b). The agrammatic patients (i.e. AG,
AV, AA) performed significantly lower in the condition with displaced object-CF
compared to the condition with in situ object-CF. This dissociation disappears in
the case of the non-agrammatic patient (MD) who performed relatively well on
both conditions. Controls performed at ceiling and, hence, their results won’t be
discussed here.
Discussion
These findings provide further evidence that predictability and the load it places
on WM is a key factor in Broca’s aphasia. An in situ CF must undergo LF movement
SYNTACTIC DEPENDENCY RESOLUTION IN BROCA’S APHASIA 155
Table 1: Number of correct responses (/total) in the experimental conditions
moved object-CF in situ object-CF

AG 3/20 16/20
AV 8/20 14/20
AA 15/20 20/20
MD 19/20 20/20
to the left periphery of the clause so as to take scope over its background. As was
the case with Quantifier Raising, this covert leftward movement appears spared in
Broca’s aphasics. Taken together, these results suggest that the problems with overt
movement are WM-related.
This conclusion is further supported by the fact that all the agrammatic patients
that participated in this study had lesions that include left IFG, whereas in
the non-agrammatic patient left IFG was intact. Recent neuroimaging studies
of non-brain-damaged individuals have shown that the presence of a syntactic
dependency is not a sufficient condition for activating left Inferior Frontal Gyrus
(IFG). Rather, a predictable displacement, as in movement-derived sentences, is
required (Santi & Grodzinsky, 2007a, 2007b, 2012). This suggests that Broca’s area
hosts a ‘syntactically constrained WM’ (Santi & Grodzinsky, 2012: 830), that is the
component of WM which is responsible for storage processes.
A currently unresolved issue is how the impaired performance on reversible
passives in Broca’s aphasia should be accounted for. Neuroimaging studies report
activation in Broca’s area, however the fact that patients’ performance varies widely
may suggest that the manner of Broca’s area involvement is different from A’-
Movement (Santi & Grodzinsky, 2012).
Acknowledgment
Maria Varkanitsa receives research support from Greek State Scholarship
Foundation (IKY), Alexander Onassis Public Benefit Foundation and A.G Leventis
Foundation. Dimitrios Kasselimis is supported by “IRAKLITOS II - University of
Crete” of the Operational Program for Education and Lifelong Learning 2007-2013
of the NSRF (2007-2013), co-funded by the European Union (European Social
Fund) and National Resources.
References
May, R. C. (1977). The grammar of quantification. MIT.
Saddy, J. D. (1995). Variables and Events in the Syntax of Agrammatic Speech. Brain and
Language, 50(2), 135-150.
Santi, A., & Grodzinsky, Y. (2007a). Taxing working memory with syntax: Bihemispheric
modulations. Human Brain Mapping, 28(11), 1089-1097.
156 VARKANITSA ET AL.
Santi, A., & Grodzinsky, Y. (2007b). Working memory and syntax interact in Broca’s area.
NeuroImage, 37(1), 8-17.
Santi, A., & Grodzinsky, Y. (2012). Broca’s area and sentence comprehension: A relationship
parasitic on dependency, displacement or predictability? Neuropsychologia, 50(5),
821-832.
Varkanitsa, M., Kasselimis, D., Potagas, C., Evdokimidis, I., Van de Koot, H., & Druks, J.
(2012). Processing of Covert Scope Inversion in Broca’s Aphasia. Procedia - Social
and Behavioral Sciences, 61(0), 277-278.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 157-159
The ability of verbal learning and memory in

patients with non-fluent aphasia
Mile Vuković & Irena Vuković
University of Belgrade - Faculty of Special Education and Rehabilitation, Serbia
Introduction
Empirical studies show that aphasic patients have a decline on tests of memory
(Barth et al., 2004; Vukovic, et al, 2008; Christensen & Wright, 2010). However,
most studies were devoted to the working memory research with application tasks
that require language skills. Therefore, it is difficult to say whether the poor
performance on tests of verbal memory is result of linguistic or memory deficits.
On the other hand, data from the literature show that little research has focused
on the ability to learn, although learning is very important for the rehabilitation of
aphasic patients. Bearing these facts in mind, the aims of the present study were:
1) examine the extent of short-term and long-term verbal memory in patients with
non-fluent aphasia; 2) examine the ability to learn verbal material in those patients;
3) examine the presence of proactive and retroactive inhibition.
Methods
Participants and Material
The sample consisted of 20 patients with non-fluent aphasia, aged 45 to 76 years.
The diagnosis of aphasia was done on the Boston Diagnostic Aphasia Examination
(BDAE). The sample included ten respondents with Broca’s aphasia (BA) and
ten respondents with transcortical motor aphasia (TMA). All patients were right-
handed, with a single left hemisphere CVA; they were at least six months post-
onset, without other cognitive deficits, severe motor speech disorders or hearing
impairment. In addition, all patients had ability to repeat single words and short
phrases from the BDAE, as well as good auditory comprehension. Twenty healthy
age-matched respondents (control group) were included in the test, as well.
The Rey Auditory Verbal Learning Test (Rey, 1964) was applied. The test consists of
two word lists (A and B); each list contains 15 words.
Procedure
In the assessment of learning and memory Rey Auditory Verbal Learning Test
(AVLT) was used, according to standard procedure (Lezak, 1995). First, the
examiner reads a list (A) of 15 words five times, at the rate of one word per second
and then requires the respondent to say all the words which she or he memorized
after each trial. On completion of trial V, the examiner reads the second word
list (B), instructing the respondents to say as many words from that list as they
158 VUKOVIĆ & VUKOVIĆ
memorized. During the repetition of the words from the list B, the examiner records
intrusions from list A, and that helps assess proactive inhibition, i.e. the extent to
which the previously learned material interferes with new learning. Following the
B list trial, the examiner asks respondents to recall as many words from the list A as
they can (trial VI). The score for each trial is the number of words correctly recalled.
Based on the number of words, the respondent correctly repeated after first five
readings of list A, the learning curve was created. At the same time, the score, after
the first repetition of the words from the list A, represents the range of short-term
verbal memory. The number of words, which the respondent repeated correctly
after the sixth trial, represents a range of long-term verbal memory.
Results
The findings showed that both groups of aphasic respondents (patients with BA
and those with TMA) have poor performance on the AVLT compared to the control
group. At the same time, patients with BA had better achievement on the task
of short-term and long-term memory compared to patients with TMA. Besides,
differences in terms of the learning curve between BA and TMA were found.
Patients with BA had a more productive learning curve than patients with TMA.
Discussion
Overall, the aphasic patients performed much worse than the control group.
Aphasic respondents scored below average value, during each repetition,
compared to the control group. Lower average values indicate that patients
with non-fluent aphasia have limited capacity of short-term and long-term verbal
memory. Further analysis indicates that aphasic patients differ from the control
group in terms of the learning curve and learning strategies. While the control
group established very productive learning curve without interference of the
contents from the A and B lists, in patients with TMA the fatigue was observed. It is
manifested by a plateau in learning, and/or lack of concentration, which in some
cases leads to a bell-shaped curve of learning. In addition, patients with TMA had
significantly more intrusion and paraphasia than the control group. The degree of
retroactive inhibition is expressed in TMA. Hence, these patients have lower score
on the A6 than on the A5 trial. This finding suggests that learning of the list B
disrupted the reproduction of previously learned list A. Also, patients with TMA had
lower score on B1 than on A1, which indicates the existence of proactive inhibition,
as well as intrusions of list A during repetition of words from list B, which is not
noticed in the control group and in patients with BA.
Regarding the aphasics, the obtained results showed significant difference in
achievement between respondents with BA and those with TMA; patients with TMA
had poorer performance on all trials compared to patients with BA. This finding
shows that learning and verbal memory deficits are more pronounced in TMA
than in BA. No specific learning strategies were observed in patients with TMA,
VERBAL LEARNING AND MEMORY IN NON-FLUENT APHASIA 159
as is the case with the control group respondents who, as expected, remembered
words from the beginning of the list better and had a uniform repetition of the
same sequence. On the other hand, respondents with BA repeated words from the
bottom of the list first. In addition, in patients with BA a high level of proactive
inhibition was noted, which shows that previously learned material distracts the
new learning. Although patients with BA, generally, have a productive learning
curve, it was observed that in these patients plateaus in learning are formed, which
suggests fatigue and limited attention span.
Bearing in mind differences in learning strategies between aphasic patients and
the control group, on one hand, and differences between tested groups of aphasic
respondents, on the other, a further study should be conducted in future to
examine the factors that interfere with learning and memory in aphasia.
References
Bartha, L., Marien P, Poewe, W., & Bbenke, T. (2004). Linguistic and neuropsychological
deficits in conduction aphasia crossed: Report of three cases. Brain and Language,
88, 83-95.
Christensen, SC & Wright, H.H. (2010). Verbal and non-verbal working memory in aphasia:
What there n-back tasks reveal. Aphasiology, 24 (6-8), 752-762.
Lezak, M.D. (1995). Neuropsychological assessment. New York: Oxford University Press.
Rey, A. (1964). L’ examen en psychologie clininque. Paris: Presses Universitaires de France.
Vukovic, M., Vuksanovic, J., Vukovic, I. (2008). Comparison of the recovery patterns
of language and cognitive functions in patients with post-traumatic language
processing deficits and in patients with aphasia following a stroke. Journal of
Communication Disorders, 41: 531-552.
Acknowledgement: This research study was supported by the Ministry of Education and
Science of the Republic of Serbia under project no. 179068 “Evaluation of Treatment of
Acquired Speech and Language Disorders”
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 160-163
Understanding discourse-linked processes in

agrammatic and fluent aphasia: a threefold
study in Russian
Laura S. Bos1,2 , Olga Dragoy3,4 , Ekaterina Iskra3 ,
Sergey Avrutin5 & Roelien Bastiaanse1
2 University of Potsdam, Germany
3 Moscow Research Institute of Psychiatry, Russia
4 National Research University ‘Higher School of Economics’, Moscow, Russia
5 University of Utrecht, The Netherlands
Introduction
Agrammatic aphasic individuals encounter problems with grammatical encoding.
However, not all syntactic processing is equally problematic, which becomes
apparent in studies that involve the relationship between narrow syntax and
discourse structure. Avrutin (2006) described that while agrammatic speakers
have a relatively good performance on who-questions and reflexives, for which
only narrow syntax is needed, their performance on personal pronouns and
referential which-questions is often impaired. For the latter, discourse and access
to information structures play an additional role. These discourse-linked elements
have a specific referent, or set of referents, outside the clause.
Recently, the theory on disturbed discourse linking in agrammatic aphasia (Avrutin,
2006) has been combined with a theory on time reference. Agrammatic aphasic
speakers find it more difficult to produce and comprehend verb forms that refer
to the past than verb forms that refer to the non-past, captured by the Past
DIscourse LInking Hypothesis (PADILIH; Bastiaanse et al., 2011). The PADILIH
predicts that verb forms referring to the past, such as ‘wrote’, are impaired in
agrammatic aphasia, because they are discourse linked: in order to interpret
past time reference, an additional link has to be made to some other event time.
Grammatical encoding, including discourse syntax, is impaired in agrammatic
aphasia. Verb forms referring to the present, such as ‘writes’, are relatively spared,
because they are locally bound: no additional discourse-link is needed because the
event time the verb refers to is in the here-and-now of the moment of speaking.
Also for people with fluent aphasia, discourse-linked structures require additional
processing. They can still refer to the past, however, they tend to resort in less
complex structures with non-finite lexical verbs, such as ‘has written’. Overall,
agrammatic aphasic speakers are less stable in assigning the correct time reference
than fluent aphasic speakers (Dragoy & Bastiaanse, 2013; Bos & Bastiaanse,
subm.) Few studies reported on the performance in the pronominal domain
in fluent aphasic speakers. These speakers have been reported to show general
DISCOURSE-LINKED PROCESSES IN AGRAMMATIC AND FLUENT APHASIA161
problems with referential elements, both for reflexives and (discourse-linked)

personal pronouns (Ruigendijk & Avrutin, 2003.) The performance of fluent
aphasic participants on who- and which-questions is unclear until now.
Aims
The reported differences in locally bound and discourse-linked processing have
been investigated in separate aphasic populations, and mixed results have been
obtained. The current study investigated the three domains in parallel, aiming to
provide answers to the following questions:
(1) Are agrammatic and fluent aphasic speakers both impaired in discourse-linked
processes?
(2) Are discourse-linked processes equally impaired in the domains of time
reference, Wh-questions and pronouns/reflexives?
Methods
Participants
Three groups were tested: 10 healthy, 10 agrammatic aphasic, and 10 fluent aphasic
participants. Aphasic participants were diagnosed within Luria’s classification with
(agrammatic) efferent and/or dynamic aphasia, or (fluent) sensory aphasia.
Materials and analysis

Three sentence-picture matching tasks were administered: (1) the Test for
Assessing Reference of Time (TART) for the Present-Imperfect and the Past-Perfect,
(2) the Wh-Extraction Assessment Tool (WHEAT) for assessing reference of which
- and who-questions in base order, and (3) the Reflexive-Pronoun Test (RePro)
testing reflexive and pronominal reference to a man or a woman. Reflexive
reference is expressed by verb inflection in Russian. Each test had 20 items per
condition. A mixed model regression analysis was performed on the accuracy data
of the three groups on the three tests.
Results
The agrammatic and fluent aphasic speakers were significantly less accurate than
the healthy control participants, who made no errors (z = 2.07, p < .05 and
z = 2.17, p < .05, respectively.) Overall accuracy of the agrammatic versus fluent
aphasic participants did not differ (z = 0.21, p > .05), see the Figure.
The discourse-linking effect in the TART was larger than in the WHEAT (z = 3.
27, p < .001), which was larger than in the RePro (z = 2.62, p < .01.) Furthermore,
the effect of discourse-linkedness was smaller in agrammatic than in fluent aphasic
speakers (z = 2.83, p < .01.) The two groups of aphasic individuals and the three
tests have therefore been analysed separately.
162 BOS ET AL.
* * *
100!
* TART Present Imperfect!
90!
*
80! TART Past Perfect!
70!
Accuracy (%)!
WHEAT Who-question!
60!
50!
WHEAT Which-question!
40!
30! RePro Reflexive!

20!
10! RePro Personal pronoun!
0!
TART WHEAT RePro! TART WHEAT RePro!
Agrammatic aphasic individuals Fluent aphasic individuals!

Figure 1: Accuracy of the agrammatic (left) and fluent (right) aphasic participants on the
three tests.
On the TART, both aphasic groups were less accurate on the past than on the
present (z = 5.62, p < .001 for agrammatic speakers and z = 2.80, p < .01 for
fluent aphasic speakers). Similarly on the WHEAT, both participant groups scored
lower on which-questions than on the who-questions (z = 2.30, p < .05 and
z = 2.08, p < .05 respectively). On the RePro, only fluent aphasic individuals had
a significantly lower accuracy on the discourse linked Pronoun condition than on
the non-discourse linked Reflexive condition (z = 2.74, p < .01; for agrammatic
speaking individuals: z = 1.81, p > .05.)
Discussion
Grammatical decoding disturbances in agrammatic aphasia affect discourse
structure, and as a result, comprehension of discourse-linked elements was
disturbed. Also in fluent aphasia, processing of discourse-linked elements was
affected. For agrammatic speakers, this effect did not become apparent in the
RePro test, where participants made only few errors, unlike the patients in the test
of Ruigendijk and Avrutin (2003). A more complex task should be used to map out
processing differences reflexives versus pronouns. The discourse-linking effect was
largest in the WHEAT, which must be due to differences in sentence structure and
task complexity.
References
Avrutin, S. (2006). Weak Syntax. In Y. Grodzinsky & K. Amunts (Eds.), Broca’s region (pp.
49-62). Oxford: Oxford University Press.
Bastiaanse, R., Bamyaci, E., Hsu, C., Lee, J., Thompson, C.K., & Yarbay Duman, T.
2011. Time reference in agrammatic aphasia: A cross-linguistic study. Journal of
DISCOURSE-LINKED PROCESSES IN AGRAMMATIC AND FLUENT APHASIA163
Neurolinguistics, 24, 652-673.

Bos & Bastiaanse (subm.) Time reference decoupled from tense in agrammatic and fluent
aphasia.
Dragoy, O. & Bastiaanse, R. (2013). Aspects of time: Time reference and aspect production in
Russian aphasic speakers. Journal of Neurolinguistics, 26, 113-128.
Ruigendijk, E. & Avrutin, S. (2003). The comprehension of pronouns and reflexives in
agrammatic and Wernicke’s aphasia. Brain and Language, 87, 17-18.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 164-166
The process of diminutivization in patients

with language impairments and children
Roberta Franceschet1 , Davide Bertocci1 , Giorgio Arcara2 ,
Serena De Pellegrin3 , Wolfgang Dressler4 & Carlo Semenza2,5
1 Department of Language and Literature Studies, University of Padova, Italy
2 IRCCS Fondazione Ospedale San Camillo, Venice, Italy
3 Neurology Clinic, Padua, Italy
4 Department of Linguistics, University of Vienna, Austria
5 Department of Neuroscience, University of Padova, Italy
Introduction
Cross-linguistically, diminutive formation is the first morphological rule that
children acquire. The exceptional status of diminutives makes them suitable for
being acquired very early (Dressler, Savickiene, 2007). Diminutives are derivational
and not inflectional, even if they are not prototypical representatives of derivational
morphology, because they do not change the category of the base. For this
reason, diminutive suffixes are not heads (Scalise, 1994), but still have some head
properties because they can change both the inflectional class and the gender
of the base (Dressler, Merlini, 1994). The representation of morphologically
complex words in the mental lexicon and their neurocognitive processing have
been debated topics in psycholinguistics, in cognitive neuroscience of language
and in neurolinguistics. Despite the fact that the category of diminutives has a wide
distribution across languages and has been largely investigated, except for Luzzatti,
Mondini, Semenza (2001) nothing relevant has been reported on diminutives from
a neuropsychological perspective. An aphasiological study, however can help in
answering the following questions.
How are diminutives represented in the mental lexicon? Which processes are
involved in their production? A comparison was thus made between selected cases
of people with aphasia, healthy young and elderly adults, and preschool children.
The relation between potentiality and illegality and the regression hypothesis were
also investigated.
Methods
Procedure
Three tasks were administered: 1) a diminutive construction task, 2) a picture
naming task and 3) a lexical decision task.
1. The diminutive construction task consisted of questions in this format: If

I want another way to say small chair I would say chair-DIM. What would
DIMINUTIVIZATION IN LANGUAGE IMPAIRMENTS AND CHILDREN 165
be another way to say small house?’. This task included 54 masculine

gender-bases and 50 feminine gender-bases from which different kinds of
diminutives can be derived (simple, e.g. gatt-ino, ‘cat-DIM’; compound, e.g.
radi-ol-ina, ‘radio-DIM-DIM’; interfixed, e.g. pont- ic-ello, ‘bridge-INTERF-
DIM’, transparent, e.g. strad-ina, ‘street-DIM’, semi-trasparent/opaque, e.g.
cane > cagn-ol-ino, ‘dog-DIM-DIM’ and lexicalized diminutives, f. spazzola,
‘brush’ > m. spazzol-ino, ‘toothbrush’).
2. In the picture naming task, two pictures appeared on a computer screen

representing the same entity; the rightmost picture was smaller than the one
on the left, however. Below the picture on the left the name of the entity
portrayed in the picture was written. The participants’ task was to name
the second picture. This task included 43 masculine gender-bases and 38
feminine gender-bases from which different kinds of diminutives could be
derived.
3. The lexical decision task required participants to classify items presented

on the screen as words or non- words. This task included 216 diminutives
(54 existent, high-frequency, e.g. pecor-ella, ‘sheep-DIM’; 54 existent, low-
frequency a, e.g. pecor-etta, ‘sheep-DIM’; 54 potential, e.g. *pecor-uzza and
54 illegal diminutives, e.g. *pecor-ic-etta). Reaction times were recorded.
In all the tasks different diminutives categories were controlled for frequency.
Participants
Three Italian individuals with aphasia (TG, BR, RC) and eleven 4-5 years old
children participated in the study. TG, BR and RC suffered an ischemic stroke. TG
had a lesion to the left frontal lobe and had Broca’s aphasia. BR and RC had a lesion
to the left temporal parietal lobe. BR suffered from mild-minimal anomic aphasia,
while RC suffered from a more severe anomic aphasia. Two control groups were
involved; one included nineteen 24-30 years old subjects, while the other group
included ten 55-67 years old subjects.
Results
In the diminutive construction task and in the picture naming task: (a) both the
control and the experimental groups produced three kinds of errors: 1) the bases
(e.g. antenna, ‘antenna’), instead of antenn-ina, ‘antenna-DIM’, 2) bases different
from the target bases (e.g. cull-etta, ‘cradle-DIM’, instead of lett-ino, ‘bed-DIM’)
and 3) lexicalized diminutives (e.g. spazzolino, ‘toothbrush’, instead of spazzol-ina,
‘brush-DIM’). (b) Only patients produced phonological errors. (c) Only children
produced simple diminutives (e.g. radio > *rad-ina instead of radi-ol-ina, ‘radio-
DIM-DIM’) and augmentatives (e.g. muro, ‘wall’ > *mur- er-one instead of mur-
etto, ‘wall-DIM’). (d) only patients and children produced inadequate responses as
166 FRANCESCHET ET AL.
analytic forms, e.g., insetto, ‘insect’ > piccolo insetto, ‘small insect’, instead of insett-
ino, ‘insect-DIM’).
In the lexical decision task: (a) Control groups and RC were faster with
high-frequency, than with illegal, than with potential, than with low-frequency
diminutives. (2) Children only distinguished existing (faster) vs. non-existing and
were not sensitive to the frequency and to the potentiality/ illegality of diminutive
forms. (3) TG was not sensitive to the difference between potential and illegal
diminutives but, unlike children, he was sensitive to the frequency effect. (4) BR
was faster with high frequency than with low frequency, but was not sensitive to
the difference between potential and illegal diminutives.
Discussion
Multiple neural circuits seem to be involved in the process of diminutivization since
patients with different lesions seem differentially sensitive to different aspects of
the process.
The error distribution in patients differed from the error distribution in children.
This result is not fully consistent with a regression hypothesis.
Frequency plays a role only in adults – both in control subjects and in patients –
whose lexicon is completely formed. On the contrary, it does not yet seem to play a
role in children. The difference between potential and illegal diminutives does not
play any role in the recognition of the items. Children as well as patients are not
sensitive to illegal diminutives. For both groups potential and illegal – and not only
potential – diminutives seem to remain as options and the patterns contained both
in potential and in illegal diminutives are overgeneralized.
References
Dressler W. U., Merlini Barbaresi L. (1994). Morphopragmatics – Diminutives and Intensifiers
in Italian, German and Other Languages. Berlin, Walter de Gruyter & Co.
Dressler W. U., Savickiene I. (2007). The role of diminutives in the acquisition of Italian
morphology, in The Acquisition of Diminutives – A cross-linguistic perspective.
Amsterdam, John Benjamins Publishing Company.
Luzzatti C., Mondini S., Semenza C. (2001). Lexical Representation and Processing of
Morphologically Complex Words: Evidence from a Reading Performance of an Italian
Agrammatic Patient. Brain and Language, 79, 345-359.
Scalise S. (1994). Morfologia. Bologna, Il Mulino.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 167-170
Complex constructions across aphasic

syndromes
Silvia Martínez-Ferreiro1 & Mireia Llinàs-Grau2
2 Universitat Autònoma de Barcelona, Spain
Introduction
The use of complex structures in aphasia has been traditionally linked to fluency.
While grammatical deficits and spared lexicon are traditionally assumed in non-
fluent aphasias (Grodzinsky, 2000), fluent aphasias tend to be characterized by
prominent word retrieval difficulties and spared syntax (Edwards, 2005). The
subordination deficit in non-fluent aphasias has received notable attention,
especially in individuals with agrammatism. Failure to produce complex structures
has been documented for several languages including English, German, Dutch,
Swedish, Polish, Finnish, Japanese and Italian (Bates et al., 1988; Menn & Obler,
1990; Sasanuma, Kamio & Kubota, 1990; Thompson et al., 1994, 1996, 1997;
Hagiwara, 1995). However, individuals with fluent aphasia have also been found
to produce structures that are less complex than those produced by NBDs (Gleason
et al., 1980; Butterworth & Howard, 1987; Niemi, 1990; Bastiaanse, Edwards & Kiss,
1996; Edwards & Bastiaanse, 1998; Edwards, 2005; Bastiaanse, 2011).
The aim of this study is to provide further evidence for the reduction of complexity
in the speech of individuals with fluent aphasia and to determine if the strategies
used to produce these structures vary in NBDs and subjects with aphasia. To
do so, the production of embedded finite clauses in English speaking subjects
with aphasia is analyzed. The claim has been made that embedded finite clauses
in English allow for a simpler option, which does not contain that (Franks,
2005; Llinàs-Grau & Fernández-Sánchez, in press). Reduction of complexity
may be associated with complementizer absence. On the basis of these results,
implications for theoretical linguistic proposals are discussed.
Methods
Subjects & Corpus samples
An analysis of spontaneous speech that included the performance of 200
individuals brought together by the AphasiaBank project (MacWhinney et al., 2011)
was carried out. Two groups were included in the study, an experimental group
comprising 100 individuals diagnosed as aphasic according to the standards of the
Western Aphasia Battery (WAB, Kertesz, 1982), and a control group including 100
non-brain damaged control subjects (NBDs).
The following corpora of subjects with aphasia were analyzed: Adler corpus, BU
168 MARTÍNEZ-FERREIRO & LLINÀS-GRAU
corpus, CMU corpus, Elman corpus, Fridriksson corpus, Garrett corpus, Kansas
corpus, Kempler corpus, Kurland corpus, and the first 21 files of the Scale corpus.
Exclusion criteria only affected those subjects that were not found to be aphasic
according to WAB (n = 8). The sample includes 60 males, 27 females, and 13
subjects for which no information is provided as for gender. The mean age across
deficits is 65;7 ranging from 36;0 to 91;9 years old.
Three corpora of control data were analyzed: Capilouto corpus, Kempler corpus,
and the first 25 files of the Wright corpus. The sample includes 48 males, 50 females
and 2 subjects for whom no information is available (mean age: 69;7, range: 23;0 -
89;6 y.o.).
Scoring
The occurrences of three highly frequent verbs (say, think and know) that select
finite complement clauses optionally introduced by that were counted. The corpus
search provided constructions of different types that were carefully selected on the
basis of their structure. Only constructions which followed the relevant pattern
were chosen for analysis.
Results
The overall results of the 100 subjects with aphasia include a total of 223
subordinate constructions. Out of them, 85.2% (n = 190) were produced without
the complementizer. Significant differences were found between subordinate
structures introduced by that and contexts of that-omission (Wilcoxon Signed
Rank test: Z = -7.144, p = 0.000). As expected, significant differences were also
found between fluent and non-fluent deficits (Mann Whitney U test: Z= -4.934,
p =0.000). While the mean use of embedded finite clauses is 3.02 per fluent
individual (SD = 3.87), it decreases to 0.26 in non-fluent subjects (SD = 0.51).
In addition to subjects with aphasia, the results of 100 NBD subjects were
also analyzed. The overall results include a total number of 514 subordinate
constructions (mean = 5.14, SD = 4.01). The complementizer was absent to a
78.4% (n = 403). Significant differences were found between subordinate structures
introduced by that and contexts of that-omission (Z = -10.848, p = 0.000). Graph 1
summarizes the results across populations.
Discussion
Our results stand in favor of characterizations of fluent aphasia that acknowledge
reduction of syntactic complexity in this group of subjects (Edwards, 2005;
Bastiaanse, 2011; among others). Second, the pattern observed as regards that-
presence vs. that-absence coincides, with the latter as the prevalent option, across
NBDs and subjects with aphasia.
In line with Franks (2005) and Llinàs-Grau & Fernández-Sánchez (in press) the high
percentage of that-absence in both the speech of subjects with aphasia and NBDs
COMPLEX CONSTRUCTIONS ACROSS APHASIC SYNDROMES 169

Figure 1: Embedded finite clauses across populations.
indicate that the embedded finite clauses selected by say/know/think may well be
regarded as selecting a bare Tense Phrase (TP) in unmarked colloquial style.
Acknowledgements
We want to thank Brian MacWhinney for permission to use Aphasia Bank.
References
Bastiaanse, R. (2011). The retrieval and inflection of verbs in the spontaneous speech of
fluent aphasic speakers. Journal of Neurolinguistics, 24, 163-172.
Bastiaanse, R., Edwards, S., and Kiss, K. (1996). Fluent aphasia in three languages: aspects of
spontaneous speech. Aphasiology, 10, 561-575.
Bates, E. A., Friederici, A. D., Wulfeck, B. B., and Juarez, L. A. (1988). On the preservation of
word order in aphasia: Cross-linguistic evidence. Brain and Language, 33, 323-364.
Butterworth, B., and Howard, D. (1987). Paragrammatism. Cognition, 26, 1-37.
Edwards, S. (2005). Fluent aphasia. Cambridge: University Press.
Edwards, S., and Bastiaanse, R. (1998). Diversity in the lexical and syntactic abilities of fluent
aphasic speakers. Aphasiology, 12, 99-117.
Franks, S. (2005). What is that? Indiana University Working Papers in Linguistics, 5, 33-62.
Gleason, J.B., Goodglass, H., Obler, L., Green, E., Hyde, M., and Weintraub, S. (1980).
Narrative strategies of aphasic and normal speaking subjects. Journal of Speech and
Hearing Research, 23, 370-382.
Grodzinsky, Y. (2000). The neurology of syntax. Behavioral and Brain Sciences, 23, 1-71.
Hagiwara, H. (1995). The breakdown of functional categories and the economy of derivation.
Kertesz, A. (1982). Western Aphasia Battery. New York: Grune and Stratton.
Llinàs-Grau, M., and Fernández-Sánchez, J. (in press). Reflexiones en torno a la supresión
del complementante en inglés, español y catalán. Revista Española de Lingüística.
170 MARTÍNEZ-FERREIRO & LLINÀS-GRAU
MacWhinney, B., Fromm, D., Forbes, M., and Holland, A. (2011). AphasiaBank: Methods for
studying discourse. Aphasiology, 25, 1286-1307.
Menn, L., and Obler, L. (1990). Agrammatic Aphasia: A Cross-Language Narrative Source
Book, 3 vols. Philadelphia: John Benjamins.
Niemi, J. (1990). Nonlexical grammatical deviations in ‘paragrammatic’ aphasia. Folia
Linguistica, 24, 389-404.
Sasanuma, S., Kamio, A., and Kubota, M. (1990). Agrammatism in Japanese: Two case
studies. In Agrammatic Aphasia: A cross-language narrative source book, vol. 2, eds.
L. Menn and L. Obler, 1225-1308. Philadelphia: John Benjamins.
Thompson, C. K., Shapiro, L. P., Schneider, S. L., and Tait, M. E. (1994). Morphosyntactic and
lexical analysis of normal and agrammatic aphasic language production. TENNET V.
Montreal, Canada.
Thompson, C. K., Shapiro, L. P., Tait, M. E., Jacobs, B. J., and Schneider, S. L. (1996). Training
Wh-question production in agrammatic aphasia: analysis of argument and adjunct
movement. Brain and Language, 52, 175-228.
Thompson, C. K., Shapiro, L. P., Ballard, K. J., Jacobs, B., Schneider, S. L., and Tait, M. E.
(1997). Training and generalised production of wh and NP movement structures in
agrammatic aphasia. Journal of Speech and Hearing Research, 40, 228-244.
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 171-173
Exploring gender inflection: an insight from

errors in aphasia
Francesca Franzon1 , Davide Bertocci1 & Carlo Semenza2,3
1 Università degli Studi di Padova, Department of Language and Literature Studies
DiSLL, Italy
2 Università degli Studi di Padova, Department of Neurosciences, Italy
3 IRCCS Ospedale S Camillo, Lido di Venezia, Italy
Introduction
In Italian every noun must be morphologically inflected for Gender and Number.
The latter is a feature whose inflection is - most of the times - clearly dependent on
a communicative context: the speaker will choose the singular or the plural form to
refer respectively to one or more entities. For what concerns Gender inflection, the
relationship between form and meaning it is not so clear-cut. In the great majority
of nouns, this feature does not bear any semantic information and surfaces as
a mere formal value, but in some other it seems to be related to the sex of the
referents.
Most grammatical theories (Matthews 1974, Thornton 2005, Zamparelli 2008)
consider Gender as a feature always inherent to nominal lexemes, being dependent
on a fixed property of each noun. Some other theories (Di Domenico 1997)
admit that nouns may be inflected for Gender on the basis of the semantic and
communicative context, as they are for Number.
A set of Italian nouns seems particularly fit for the purpose of exploring the
possibility that Gender may inflect dependent on contextual properties: these
nouns, all denoting animate referents, occur as couples like cavallo-cavalla (horse
- mare) in which the masculine and the feminine forms, referring respectively to
male/unmarked and female individuals, share the same root and differ only for the
morpheme that marks their gender.
Testing people with aphasia may provide some insights on the question.
Methods
Materials
The experimental material consisted of: (a) 74 nouns whose gender might be
contextually assigned (37 masculine, cavallo and 37 matched feminine, cavalla);
in these nouns the opposition of Gender was overt at the morphophonological
level: feminine nouns ended in -a, masculine ones in -o; (b) 26 nouns in which
the opposition of sex is expressed by different lexical roots (13 masculine, toro -
bull, and 13 matched feminine, vacca-cow); (c) 14 control nouns (7 masculine, 7
172 FRANZON ET AL.
feminine) in which gender opposition does not bear the semantic opposition male-
female (collo- neck, colla-glue).
The tasks administered were: Repetition; Reading aloud; Gender Shift (e.g.: S:
cugino? R: cugina; S: vacca? R: toro); Inflection Completion (S: ho visto un
maestr__? R: ho visto un maestro.), Article Completion (S: ho visto __ maestra;
R: ho visto una maestra).
Participants
PW is a 88 year-old right handed female. She suffered a CVA involving the left insula
and was diagnosed as affected by Wernicke’ s aphasia.
AB is a 61-year-old right-handed female. She suffered a CVA involving the territory
of the left Middle Cerebral Artery. She was diagnosed as affected by Broca’s aphasia.
Results
PW was tested only in “Gender Shift”. She was more impaired on the nouns whose
Gender is inherent: she made 4/74 (5,4%) errors in type (a) nouns vs 22/26 (84,6%)
in type (b) nouns. The errors mainly consisted of switching gender of type (b) nouns
by overextending the inflection rule expected for type (a) nouns, giving results such
as uomo (man) >*uoma instead of donna (woman) in 11/22 (50%) cases.
In the same task, AB showed a different pattern: she made 5/74 (6,75%) errors in
type (a) nouns and 11/26 (42,3%) errors in type (b) nouns, mainly omissions: 7/11
(63.6%).
AB’s overall performance was slightly worse in type (a) nouns: she made 60/370
(16.2%) errors; in type (b) nouns the errors were 13/130 (10%). The most interesting
results emerged in her Completion tasks: in “Inflection Completion”, errors were
37/74 (50%) for type (a) and 2/26 (7.7%) for type (b). In most cases (31/37),
she produced the wrong gender morpheme, e.g. *una cavallo. In contrast, her
performance in “Article Completion” was better: the errors were 6/74 (8.1%) for
type (a) and none for type (b).
Discussion
The different performance of the two participants in “Gender Shift” points to
an asymmetry between a Gender assignment of contextual kind and Gender as
depending upon an inherent lexical feature (Atkinson 2012, Slehman & Ihsane
2013). The fact that PW overextended to type (b) nouns the morphophonological
inflection rule of type (a) nouns suggests that such a rule is not only spared, but
that it is also the default operation for animate nouns. Both the contextual and
the lexical Gender are checked in an independent syntactic projection: for what
concerns the former, the feature may assume different values depending on the
semantic context; for what concerns the latter, it may only assume the value fixed
on a lexical basis, depending on a constraint set by the lexical head.
EXPLORING GENDER INFLECTION: AN INSIGHT FROM ERRORS IN APHASIA
173
The sharp discrepancy between AB’s two Completion tasks is unlikely due to a
generic difference in processing resources: it may rather stem from a contextual
operation of gender assignment, independent of proper concord operations that
take place in higher DP positions, which are spared in her performance.
With respect to linguistic theory, we suggest that Gender is assigned in an
autonomous projection in syntax, [class]. More in detail, we hypothesize the
existence of a syntactic layer AgrP (parallel of the verbal IP) whose position is
intermediate between the lexical layer NP and the DP (Giusti 2009). In this layer the
features that must show up at the morphophonological level are encoded: Gender
(and maybe Number) are processed in AgrP projections before rising to the DP.
Both features may inflect according to the semantic and communicative context,
but their formal values must be chosen among the set “allowed” by the lexical
head for what concerns the word inflectional class; their variability depends on a
parameter set by the lexical features as well.
References
Alexiadou, A. (2004) Inflection class, gender and DP internal structure. In G. Müller et al.
(eds.) Exploration in Nominal Inflection. Mouton de Gruyter, 21-50.
Atkinson E. (2012) Gender Features on N and on the root: an account of Gender in French.
Talk at LSRL 42, Cedar City
Corbett G. (1991) Gender, Cambridge University Press, Cambridge
Giusti G. (2009) On feature sharing and feature transfer. University of Venice working papers
in Linguistics, 19: 157-174.
Di Domenico E. (1997) Per una teoria del genere grammaticale. Padova, Unipress
De Vincenzi M. (1999) Differences Between the Morphology of Gender and Number:
Evidence from Establishing Coreferences. Journal of Psycholinguistic Research 28,
5: 537-553
Luzzatti C. - De Bleser R. (1996) Morphological Processing in Italian Agrammatic Speakers:
Eight Experiments in Lexical Morphology”, Brain and Language 54: 26-74.
Luzzatti C.G. - Mondini, S. - Semenza, C. (2012). Lexical impairment in Agrammatism. In
R. Bastiaanse, & C.K. Thompson (eds.), Perspectives on Agrammatism: 60-74. Hove:
Psychology Press
Matthews P. H. (1974) Morphology, Cambridge University Press, Cambridge
Sleeman P. - Ihsane T. (2013) Local and non-local gender agreement in French, Talk at IGG
39, Modena
Thornton A. M. (2005) Morfologia, Carocci, Roma
Zamparelli (2008) On the interpretability of φ-features. In C. De Cat & K. Demuth (eds.) The
Bantu-romance connection: a comparative investigation of verbal agreement, DPs,
and information structure: 167-199
STEM-, SPRAAK- EN

Vol. 18, No. S01, 2013, pp. 174-178
Comprehension of if-conditionals at the

morphosyntax-semantics interface in Turkish
Broca’s aphasia
Tuba Yarbay Duman1 & İlknur Maviş2
1 Amsterdam Center for Language and Communication, University of Amsterdam,
The Netherlands
2 Education, Research & Training Centre for Speech and Language Pathology,
Anadolu University, Turkey
Introduction
Broca’s aphasia is diagnosed in individuals who have significant deficits in
expressive and/or receptive language skills, due to brain injury (Goodglass,
1980). Hallmark features of the disorder include profound impairments in their
morphosyntactic abilities, and specifically in their abilities to use correct tense
/ time-reference marking morphology (e.g. Stavrakaki & Kouvava, 2003; Yarbay
Duman & Bastiaanse, 2009). However, specificity of language impairment is
subject to debate. Recent studies suggest that non-linguistic deficits often co-occur
with linguistic deficits in Broca’s aphasia, implying that individuals with Broca’s
aphasia do not only have impaired language but also suffer from cognitive deficits,
particularly in specific executive functions (e.g. inhibition: Peristeri, Tsimpli, &
Tsapkini, 2011). From this point of view, the morphosyntax-semantics interface
is a highly interesting area to look into because it allows detecting relationships
between morphosyntactic abilities and cognitive abilities of individuals with
Broca’s aphasia. However, although there is extensive work on morphosyntactic
impairments in Broca’s aphasia (e.g. Bastiaanse & Thompson, 2003 for Dutch and
English; Yarbay Duman et. al., 2011 for Turkish), hardly anything is known about
the morphosyntax-semantics interface in this group.
For this investigation, we examined the ability of Turkish-speaking individuals with
Broca’s aphasia to comprehend counterfactuals and compared the results with
those of a control group consisting of individuals with no speech and language
impairment history. Counterfactuals are thoughts about ‘what might have been‘,
that is, imagining what might have happened, but did not. For example, when
a person misses an important interview, he might think ‘if I had taken the bus!’.
In this scenario, the person generated a different event and an outcome than
what has happened in reality, by inhibiting the current state of affairs. That is,
the counterfactual is the scenario in which the person took the bus and made
his interview as opposed to the factual where the person missed the interview.
According to Beck, Riggs, & Gorniak (2009) three executive functions are required
for counterfactual thinking: (1) inhibition – to ignore what has happened; (2)
working memory – to hold two different representations simultaneously in mind;
COMPREHENSION OF IF-CONDITIONALS IN TURKISH BROCA’S APHASIA175
(3) cognitive switching – shifting between those representations. Hence, using, but
also understanding, counterfactuals involve complex cognitive processes. We also
tested the comprehension of nonconditional clauses and factual conditionals, in
which no such cognitive processes are involved, to manipulate morphosyntactic
(the presence/absence of an if-conditional) and semantic (counterfactuality)
variables separately.
Table 1: Summary of the clause types with respect to their relevant characteristics to the
present study
If - embedding Verb morphology Factual Specific EF

NonConditional
Gömleğ–i ütüle–di ve dolab–a as–tı
The shirt-acc iron-past/3sg-conditional the closet-dat hang-aorist no past tense yes no
He ironed the shirt and hung it up in the closet
Interpretation: the shirt was ironed and hung up in the closet
Factual
Gömleğ–i ütüle–diy–se dolab–a as–ar
The shirt–acc iron–past/3sg –conditional the closet–dat hang–aorist yes past tense + -sa yes no
If he has ironed the shirt, he will hang it up in the closet
Interpretation: when the shirt is ironed, it is in the closet
Counterfactual
Gömleğ–i ütüle–sey-di dolab–a asar–dı
The shirt–acc iron–conditional–past/3sg the closet–dat hang–past/3 sg yes -sa + past tense no yes
If he had ironed the shirt, he would have hung it up in the closet
Interpretation: The shirt is certainly not ironed and not hung up in the closet
Individuals with Broca’s aphasia have morphosyntactic deficits. Thus, it was

expected that conditional clauses would be more difficult to comprehend
than nonconditional clauses since the former are morphosyntactically more
complex. Furthermore, in Turkish, counterfactual and factual conditionals are
morphosyntactically equivalent, which allows observing whether counterfactuality
itself adds to difficulties that individuals with Broca’s aphasia experience in
sentence comprehension. Considering the assumption that individuals with
Broca’s aphasia are impaired in executive functions such as inhibitory control,
it was predicted that counterfactual conditionals would be more difficult to
comprehend than factual conditionals. That is, it was expected that individuals
with Broca’s aphasia, unlike the control group, would have more difficulties with
counterfactuals not only because of morphosyntactic complexity but also due to
semantic complexity and the cognitive processes involved in counterfactuals.
Table 1 presents examples of test sentences and their relevant characteristics of
the present study. Note that conditionals in Turkish are marked by a conditional
suffix on the main verb (-(y)sa) ‘if’, which occurs to the right of the past tense
marker – dı in factuals (temporally past) and to its left in counterfactuals. Note
that Turkish, unlike English, has a specialized conditional morpheme used for
both counterfactuals and factuals. Besides, Turkish does not make use of if –
complementizer, modals, auxiliaries, and participle verbs in its formulation of
these structures, which make them grammatically easier and semantically more
transparent than in English.
176 YARBAY DUMAN & MAVIŞ
Methods
Subjects
12 individuals with Broca’s aphasia were tested (7 male / 5 female: MA: 54.2). The
diagnoses were based on the standardized Afazi Dil Değerlendirme Testi (ADD)-
The Test of Language Assessment in Aphasia (Maviş & Toğram, 2009) and the
clinical judgments of a speech therapist. All the patients were at least 3 months
post onset of left CVA (lesion data are available for each patient). Ten non-brain-
damaged Turkish speakers participated (and performed at ceiling) on the test.
Materials
A spoken–sentence–to–picture–matching task was developed with three conditions
(nonconditional, factual conditional, counterfactual conditional), with 15 items
each in the test. A sentence was read aloud by the experimenter and the
participants were asked to point to the picture that matched the spoken sentence
(see Figure 1). In the factual condition, the subjects were always shown pictures
in which the action of the main clause was realized (e.g. ironing was done), as a
consequence of which the action of the other clause had to be realized as well e.g.
the shirt is in the closet.
Results
Factual conditionals were significantly more difficult to comprehend than
nonconditionals (t=2.311; df=1;p=.041) and counterfactual conditionals (t=2.692;
df=11, p=.021). However, there was no difference between factual and
counterfactual conditionals (t =1.351, df =11; p=.204). In this analysis, lexical
errors were ignored (these were cases in which the patients interpreted the target
sentence as a factual or a counterfactual correctly-but, by choosing a semantically
related item e.g. shirt vs. dress) to see the effect of counterfactuality more clearly.
This was the most conservative way to analyze the data since the subjects made
twice as many lexical errors as in factuals than in counterfactuals. Exactly the
opposite pattern was observed for Action Unfulfilled errors (these were the cases
in which the patients interpreted a counterfactual as a factual or vice versa).
Although there was no significant group difference between factuals and
counterfactuals at the group level, 4 patients (out of 12) were significantly impaired
in counterfactuals compared to factuals (chi-square).
Discussion
There are two major findings. First, comprehending factual and counterfactual
conditionals was more difficult than comprehending nonconditionals for Turkish
individuals with Broca’s aphasia. Second, although comprehending factuals and
counterfactuals were equally difficult for them at the group level, some of the
patients were selectively impaired in counterfactual thinking. Apparently, sentence
COMPREHENSION OF IF-CONDITIONALS IN TURKISH BROCA’S APHASIA177
Figure 1: Counterfactual Test Stimuli: Gömleği ütüleseydi dolaba asardı ‘If he had ironed the
shirt, he would have hung it up in the closet’
(1) Target picture; (2) Action Unfulfilled distractor (picture in which the action mentioned
was not fulfilled); (3) Lexical distractor (picture in which action was fulfilled, but the lexical
item was incorrect); and (4) [Action Unfulfilled + Lexical] distractor (picture in which both
the action was not fulfilled and the lexical item was incorrect).
comprehension in Broca’s aphasia is influenced by morphosyntactic properties

of sentence structure: conditionals that were morphosyntactically more complex
than nonconditionals were more difficult to comprehend for them, supporting
earlier data for Turkish (Yarbay Duman et. al., 2011). Furthermore, comprehending
counterfactuals at the morphosyntax-semantics interface is particularly difficult
for some individuals with Broca’s aphasia. Thus, executive functions such as
inhibition required for counterfactual thinking can also be a factor hampering their
sentence comprehension ability.
178 YARBAY DUMAN & MAVIŞ
References
Bastiaanse, R., & Thompson, C.K. (2003). Verb and auxiliary movement in agrammatic
Broca’s aphasia. Brain and Language, 84, 286–305.
Beck, S.R., Riggs, K.J., & Gorniak, S.L. (2009). Relating developments in children’s
counterfactual thinking and executive functions. Thinking and Reasoning, 15, 337-
354.
Goodglass, H. (1980). Disorders of naming following brain injury. American Scientist, 68,
647-655.
Maviş, İ., & Toğram,B. (2009). Language assessment test for aphasia (ADD). Detay Yayıncılık:
Ankara.
Peristeri, E., Tsimpli, I.M., Tsapkini, K. (2011). Linguistic Processing and Executive Control:
Evidence for Inhibition in Broca’s aphasia. Procedia Social and Behavioral Sciences,
23, 213–214.
Stavrakaki, S., & Kouvava, S. (2003). Functional categories in agrammatism: evidence from
Greek. Brain and Language, 89, 129–141.
Yarbay Duman, T., & Bastiaanse, R. (2009). Time reference through verb inflection in Turkish
agrammatic aphasia. Brain and Language, 108, 30–39.
Yarbay Duman, T., Altınok, N., Özgirgin, N, & Bastiaanse, R. (2011). Sentence comprehension
in Turkish Broca’s aphasia: An integration problem. Aphasiology, 25 (8), 908–926.
179
Index
Abel, S., 126 De Somer, C., 31

Ablinger, I., 52 De Surgeloose, D., 34
Aerts, A., 13, 19, 29 De Witte, E., 4, 34, 59
Allamano, N., 48 Dirven, C., 1
Alyahya, R.S.W., 112 Dolenc, B., 132
Arcara, G., 164 Dragoy, O., 8, 98, 160
Archer, B.E., 16 Dressler, W., 164
Arslan, S., 116 Druks, J., 153
Avrutin, S., 160 Dua, G., 34
Duyck, W., 19
Bastiaanse, R., 8, 59, 79, 83, 86, 116, 120, 149, 160
Batens, K., 19 Engelborghs, S., 31
Berry, R., 123 Evdokimidis, I., 153
Bertocci, D., 164, 171
Best, W., 44, 72
Fedina, O.N., 98
Bi, Y., 101
Fedor, A., 72
Blankestijn-Wilmsen, J., 105
Fern-Pollack, L., 72
Boeckx, C., 22
Franceschet, R., 164
Bos, L.S., 160
Franzon, F., 171
Brysbaert, M., 25
Fyndanis, V., 39
Burchert, F., 55
Gallardo-Paúls, B., 142

Cai, Q., 25
Gitterman, M., 145
Capasso, R., 48
Goral, M., 145
Capitani, E., 48
Grassly, J., 44
Carlo Semenza, C., 164, 171
Groenewold, R., 120
Chwalek, V., 126
Gutyrchik, E.F., 98
Code, C., 123
Colle, H., 4
Corthals, P., 109 Haaland-Johansen, L., 123
Costanzo, M., 48 Halm, K., 52
Cosyns, M., 109 Han, Z., 101
Hanne, S., 55
Hartsuiker, R.J., 13
Damen, I., 105
Howard, D., 64
de Aguiar, V., 8
Huber, W., 52, 126
De Bleser, R., 55
Hughes, L., 72
De Coninck, M., 25
Huiskes, M., 120
De Deyn, P.P., 25
Hurkmans, J., 105
de la Paz Cabana, M., 123
De Letter, M., 13, 19, 29, 109
De Pellegrin, S., 164 Iskra, E., 160
180 INDEX
Ivanova, M.V., 98 Papathanasiou, I., 123

Ivaskó, L., 76 Paplikar, A., 145
Patterson, R., 123
Petrushevsky, A.G., 98
Jakab, K., 76
Pirtošek, Z., 132
Jalvingh, F.C., 79
Potagas, C., 153
Jap, B.A., 83
Prizl-Jakovac, T., 123
Jonkers, R., 105
Radach, R., 52
Kapikian, A., 72 Radermacher, I., 126
Kasselimis, D., 153 Raedt, R., 19, 29
Kloet, A., 1 Reis, A., 8
Kok, D. de, 86, 116 Renvall, K., 64
Konstantinopoulou, P., 90 Reyes, A.F., 149
Koot, H. van de, 153 Robert, A., 123
Kosmidis, M.H., 95 Robert, E., 4
Kozintseva, E.G., 98 Rochon, E., 123
Krzok, F., 126 Rofes, A., 59
Kuptsova, S.V., 98 Romanova, A., 64
Roncoli, S., 72
Laiacona, M., 48 Rosci, C., 48
Law, S-P., 101 Rubio-Bruno, S., 123
Leśniak, D., 129
Leivada, E., 22 Sandra, D., 31
Leko, A., 123 Santens, P., 13, 19, 29
Leonard, C., 123 Santens,P., 109
Llinàs-Grau, M., 167 Satoer, D., 1, 4
Lorenzi, L., 48 Sevan, D.A., 98
Smits, M., 1
Soebadi-Haryadi, R.D., 83
Malyutina, S.A., 98
Stavrakaki, S., 90, 95
Manouilidou, C., 90, 132
Strobbe, G., 109
Mariën, P., 4, 25, 31, 34, 59, 67
Marjanovič, K., 132
Martínez-Álvarez, A., 22, 136, 139 Thomas, M., 72
Martínez-Ferreiro, S., 139, 167 Tóth, A., 76
Marvin, T., 132
Masterson, J., 72 Van Borsel, J., 29, 109
Maviş, İ., 174 Van Dam, D., 25
Megari, K., 95 Van der Haegen, L., 25
Miceli, G., 48 Van Hecke, W., 25, 34
Moens, M., 34 van Mierlo, P., 13, 109
Moreno-Campos, V., 142 Van Roost, D., 19, 29
Müller, N., 16 Vandenborre, D., 67
Vanhoutte, S., 29, 109
Varkanitsa, M., 153
Newton, C., 44
Vasishth, S., 55
Nickels, L., 64, 120
Vécsei, L., 76
Niemann, K., 126
Villanueva, M.M., 123
Nobis-Bosch, R., 126
Vincent, A., 1, 4
Visch-Brink, E., 1, 4
Obler, L. K., 145 Voorbraak-Timmerman, V., 105
INDEX 181
Vuković, I., 157

Vuković, M., 157
Willmes, K., 126

Wolthuis, N., 86
Yarbay Duman, T., 174

Yu, X., 101
Zafeiriou, D., 90
Zemva, N., 123
WÖHRMANN
PRINT SERVICE
Printing & Binding on Demand
Boeken, Proefschriften, Prijslijsten,

Verenigingsbladen, Handleidingen, etc.
Garenloos, Genaaid, Gebrocheerd, Hard Gebonden
Zwart/wit, Full-colour, Zwart/wit en Full-colour

gecombineerd.
Loskade 4
7202 CZ Zutphen
Tel: 0575 - 585300

Fax: 0575 - 585308
www.wps.nl
wps@wps.nl
A4-formaat-nl.indd 1 13-1-2010 21:11:21

STEM-, SPRAAK- EN TAALPATHOLOGIE the website complete papers can be
The journal SSTP publishes original, downloaded in pdf-format. The website
scientific contributions in the area of is: http://www.sstp.nl/
voice, speech and language disorders By open access publishing SSTP feels
and aims to reach on both researchers it can best fulfill its goals. This can
and health care and education be accomplished due to the following
professionals in the Netherlands and effects:
Flanders. SSTP covers a challenging,
fast developing multidisciplinary • Free availability, concerning
field. This field includes the costs and reach. Especially for
medical specialties neurology, speech and language therapists
otolaryngology, foniatrics and (8000 in the Netherlands and
revalidation medicine, the behavioral Flanders), who are interested
sciences orthopedagogics and in new developments in their
developmental and neuropyschology, own – for the group as a whole
psycholinguistics and neurolinguistics, very diverse – area of work,
the cognitive sciences, audiology and open access publication offers
last but not least speech- and language the opportunity to follow the
pathology. literature selectively.
In summary, an extremely varied field. • A wider range of availability to

The Dutch language area is not very professionals makes publishing
large, but offers a lot of knowledge in in SSTP more appealing for
the area of research, implementation researchers. SSTP can be the
and application of new developments major channel for results of
in the voice, speech and language especially applied research, in
pathology. SSTP is therefore a source of which specifically the Dutch
information and platform for discussion language plays a large role.
for many readers that want to follow the By open access publishing
newest developments in this language- the authors can be sure that
related field without being forced to rely interested professionals will be
on a large variety of journals. reached.
In order to realize this, SSTP
publishes articles of different scope:
theoretical overview papers, papers on
fundamental and clinical scientific
research, papers about research
methodology and techniques, book
reviews, conference reports and a
conference agenda.
SSTP is digitally available as an
open access journal. The articles
are available for everybody: on

Aphasia

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Aphasia

Uploaded by

Copyright:

Available Formats

Stem-, Spraak- en

14th International Science of Aphasia

This year’s conference is organized by the members of the research group

We wish you an intellectually stimulating conference and a most enjoyable stay in

Jo Verhoeven & Peter Mariën

On behalf of the Local Organising Committee

The 14th International Science of Aphasia Conference is held in Brussels, Belgium,

Local organizing committee 2013

Abstract Selection Committee

Friday, September 20, 2013

Arrival, Registration & Welcome Reception

Saturday, September 21, 2013

Session 1: Language and Cognition

12:30 - 13:30 Lunch

13:30 - 14:30 Contributed Papers I

14:30 - 15:30 Poster Session I

Cedric Boeckx et al.: On the theoretical characterization of agrammatism:

15:30 - 16:00 Coffee Break

16:00 - 17:20 Contributed Papers II

18:00 - 19:00 Committee Meeting

Sunday, September 22, 2013

12:30 - 13:30 Lunch

13:30 - 14:30 Contributed Papers III

14:30 - 15:30 Poster Session II

15:30 - 16:00 Coffee Break

16:00 - 17:20 Contributed Papers IV

Monday, September 23, 2013

11:00 - 11:30 Coffee Break

11:30 - 12:30 Contributed Papers V

12:30 - 13:30 Poster Session III

13:30 - 14:30 Lunch

14:30 Afternoon Excursion and Conference Dinner

Tuesday, September 24, 2013

12:30 - 13:30 Lunch

13:30 - 15:10 Contributed Papers VI

15:10 - 15:30 Coffee Break

15:30 - 16:30 Forward Looking Plenary Discussion & Floor Discussion

Wednesday, September 25, 2013

Long term cognitive functioning after glioma

2. Teixidor P, Gatignol P, Leroy M, Masuet-Aumatell C, Capelle L, Duffau H. Assessment

A standard neurolinguistic approach to awake

Timing of Linguistic level Task Stimuli

Event related potentials of the processing of

(IDEALAB), University of Potsdam, University of Groningen, University of Trento,

Rehabilitation (CeRIN), University of Trento, Rovereto, Italy

Groningen, The Netherlands

Biotechnology & Bioengineering, Centre for Molecular and Structural Biomedicine,

of pronouns at anterior sites, from 500-800ms (F(1,27)=6.805, p=0.015). Figure 1

identity of an antecedent is achieved through conceptual and not syntactic steps.

evidence from German and Dutch. BMC Neuroscience, 7, 23. doi:10.1186/1471-2202-

Gender differences in neurophysiological

Information Systems, Ghent University - IMinds, Belgium

Figure 1: Gender difference during controlled PoA phoneme discrimination displayed in

Word retrieval in aphasic Sesotho-speakers:

Many psycholinguistic studies of naming have been informed by two models

Clinical use of event-related potentials in

Ghent University, Belgium

Neuroanatomical imaging and behavioural language testing cannot provide

Phonology is the most disturbed linguistic modality in aphasia, due to its

On the theoretical characterization of

(1) [CP C’ [AgrP Agr’ [TP T’ [VP [V’]]]]] (Belletti 1990)

Findings from agrammatism suggest that there is a crosslinguistic pattern

(2) [CP C’ [TP T’ [AgrP Agr’ [VP [V’]]]]] (Pollock 1989)