This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Available online at www.sciencedirect.com
Cognition 108 (2008) 243–262
www.elsevier.com/locate/COGNIT
Brief article
Morpheme-based reading aloud: Evidence
from dyslexic and skilled Italian readers
Cristina Burani a,*, Stefania Marcolini a,c, Maria De Luca b,
Pierluigi Zoccolotti b,d
a
Institute for Cognitive Sciences and Technologies, ISTC-CNR, Via S. Martino della Battaglia 44,
00185 Rome, Italy
b
Neuropsychology Unit, IRCCS Fondazione Santa Lucia, Rome, Italy
c
University of Padua, Italy
d
Sapienza University of Rome, Italy
Received 8 November 2006; revised 22 December 2007; accepted 24 December 2007
Abstract
The role of morphology in reading aloud was examined measuring naming latencies to
pseudowords and words composed of morphemes (roots and derivational suffixes) and corresponding simple pseudowords and words. Three groups of Italian children of different ages
and reading abilities, including dyslexic children, as well as one group of adult readers participated in the study. All four groups read faster and more accurately pseudowords composed of
root and suffix than simple pseudowords (Experiment 1). Unlike skilled young and adult readers, both dyslexics and younger children benefited from morphological structure also in reading aloud words (Experiment 2). It is proposed that the morpheme is a unit of intermediate
grain size that proves useful in processing all linguistic stimuli, including words, in individuals
with limited reading ability (dyslexics and younger readers) who did not fully develop mastering of whole-word processing. For skilled readers, morphemic parsing is useful for reading
those stimuli (i.e., pseudowords made up of morphemes), for which a whole-word lexical unit
does not exist; where such whole-word lexical units do exist, skilled readers do not need to rely
*
Corresponding author.
E-mail address: cristina.burani@istc.cnr.it (C. Burani).
0010-0277/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.cognition.2007.12.010
Author's personal copy
244
C. Burani et al. / Cognition 108 (2008) 243–262
on morphological parsing because they can rely on a lexical (whole-word) reading unit that is
larger than the morpheme.
Ó 2008 Elsevier B.V. All rights reserved.
Keywords: Morphology; Reading; Developmental dyslexia
1. Introduction
In languages as different as English, Finnish and Italian, elementary school
children use the morphemic constituents of words in performing tasks on written
stimuli such as fragment completion (Feldman, Rueckl, DiLiberto, Pastizzo, &
Vellutino, 2002), word definition (Bertram, Laine, & Virkkala, 2000; Burani,
Bimonte, Barca, & Vicari, 2006) and lexical decision (Burani, Marcolini, & Stella,
2002). Sensitivity to word morphology develops early in childhood (Carlisle &
Nomabody, 1993; Casalis & Louis-Alexandre, 2000) and is present to some extent
also in impaired readers (Casalis, Sopo, & Colé, 2004; Elbrö & Arnbak, 1996;
Leong & Parkinson, 1995).
Morphological knowledge in word comprehension and production tasks does
not necessarily imply that morphemes also play a role as processing units in a
print-to-sound decoding task such as reading aloud. Little information on this
topic is available on children. Studies in opaque orthographies, such as English,
Danish and French, followed the assumption that in such orthographies (in
which word spelling is to some degree morphologically governed) knowledge
of morphemes may help the child in assigning the correct word pronunciation
(Seymour, 1997; Verhoeven & Perfetti, 2003). With untimed stimulus presentation, the presence of known morphemes in a word, such as stems and affixes,
may affect young readers’ accuracy in reading aloud, mainly when morphologically complex words are phonologically and semantically transparent with respect
to the base word (Carlisle & Stone, 2003; Elbrö & Arnbak, 1996; Laxon, Rickard, & Coltheart, 1992), or when suffixes are frequent and productive (Mann
& Singson, 2003).
Up-to-date, only one study has assessed the role of morphology in children’s
reading aloud a transparent language. Burani et al. (2002) showed that young
Italian readers in third to fifth grades could benefit from the presence of morphemes similarly to adult readers (see review in Burani & Laudanna, 2003). In
a naming task, pseudowords made up of a root and a derivational suffix in a
combination not existing in Italian (e.g., DONNISTA, ‘womanist’) were read faster and more accurately than simple pseudowords matched for orthographic
familiarity (e.g., DENNOSTO).
To our knowledge, no study has investigated whether in transparent orthographies morphemes are effective units also in reading aloud words. For an experienced reader, parsing a word into morphemic sub-parts may be an efficient
strategy when it is not familiar. For low-frequency words, the recourse to
Author's personal copy
C. Burani et al. / Cognition 108 (2008) 243–262
245
higher-frequency constituents (morphemes) may facilitate processing (Burani &
Thornton, 2003).
Apart from frequency, the adoption of morphemes as reading units may
also be constrained by the joint effect of word length and reader’s ability.
For both adult and young skilled readers, 7–10 letter words can be processed
in one shot (Hutzler & Wimmer, 2004; Rayner & McConkie, 1976). However,
developmental dyslexics experience difficulties in processing such stimuli as
whole units. Italian dyslexics’ eye fixations reveal fractionated text scanning,
with a prevalence of small amplitude saccades (De Luca, Di Pace, Judica, Spinelli, & Zoccolotti, 1999). These eye-fixation patterns result in an extremely
slow and analytical reading strategy and in marked stimulus length effects
(Spinelli et al., 2005; Zoccolotti et al., 1999, 2005a) affecting similarly words
and non-words (De Luca, Borrelli, Judica, Spinelli, & Zoccolotti, 2002; see
also Hutzler & Wimmer, 2004). This pattern of reading performance resembles
that of children at an early stage of learning to read (Zoccolotti et al.,
2005a).
The present study assessed reading aloud of morphologically complex words
and pseudowords in Italian children of different reading abilities, including
developmental dyslexics. Our hypothesis was that morphemic constituents (roots
and suffixes) could help dyslexics to read aloud both pseudowords and words.
Morphemes may be efficient reading units for dyslexics because they have an
intermediate size between graphemes, which lead to extremely slow and analytical processing, and words, which for dyslexics are too large units to be processed as a whole. Similar benefits from morpheme-based reading aloud of
familiar words may be present in younger readers, but may not occur for
skilled readers due to their ability to process larger reading units as a whole.
Word-based reading avoids parsing and assembling costs connected to morpheme-based reading (for costs of morphemic parsing in lexical decision, see
Laine, Vainio, & Hyönä, 1999; Traficante & Burani, 2003).
The reading aloud of both pseudowords and words composed of morphemes
(roots and derivational suffixes) was compared to that of simple pseudowords
and words with no root + suffix structure. Three groups of Italian children of different ages, with and without reading difficulties, were tested along with a group
of adult readers. The aim was to show that, in a transparent orthography, readers
of different skills may take advantage of reading units (morphemes) of larger than
the single grapheme grain size. All groups of readers, irrespective of reading skill,
were expected to take advantage of morphemic units in reading aloud pseudowords: the presence of morphemes in a pseudoword would result in shorter reading latencies and higher reading accuracy than grapheme-based reading. Only less
skilled readers, i.e., younger readers and dyslexics, who have not developed efficient whole-word reading ability, were expected to rely on morphemic constituents when reading aloud words. In contrast, skilled readers, both children and
adults, should read as fast and accurately both morphologically complex and simple words, because of their capacity to process both types of words as whole
units.
Author's personal copy
246
C. Burani et al. / Cognition 108 (2008) 243–262
2. Experiment 1. Reading aloud pseudowords
In the first experiment, the reading aloud of pseudowords made up of morphemes
was contrasted with the reading of simple pseudowords.
2.1. Method
2.1.1. Participants
Four groups of participants were included in the study: dyslexics, chronologicallymatched skilled readers, reading-matched younger normally-developing children and
adults.
Criterion for inclusion in the dyslexic group was marked reading delay on a standard reading achievement test (MT test, Cornoldi, Colpo, & Gruppo, 1981). The
child reads aloud a passage with a 4-min time limit; speed (second per syllable)
and accuracy (number of errors, adjusted for the amount of text read) are scored.
Dyslexics scored below 1.5 z-scores in either speed or accuracy. A total of 17 dyslexics was examined. Of these, 7 were below the cut-off for both speed and accuracy and
10 for accuracy only.
Dyslexics were compared to 34 skilled readers of the same chronological age. Performance of these children on the MT test was well within normal limits (with mean
z-scores near-zero) for both accuracy and speed. Dyslexics and same-age skilled
readers were matched for sex and non-verbal intelligence (Raven’s Colored Progressive Matrices). Summary statistics and mean scores on screening tests are given in
Table 1.
Dyslexics were also compared to 17 younger typically developing children
matched for sex and reading speed on the MT test (0.45 and 0.41 second per syllable
for the dyslexic and typically developing children, respectively; t < 1, ns). Also these
children performed within normal limits on the MT test (with mean z-scores nearzero) according to their respective age norms (Table 1).
Table 1
Summary statistics (mean age in years and months, with range in parentheses
Chronological age Male
Dyslexic 6th 11;3 (10;10–12;3)
graders
2nd–3rd
8;3 (7;1–9;2)
graders
Skilled 6th 11;1 (10;4–11;8)
graders
Female Raven test
N = 11 N = 6
30.12 (SD = 3.4)
N = 10 N = 7
25.6 (SD = 2.9)
Reading speed
1.41 (SD = 1.2)
.19 (SD = .17)
N = 22 N = 12 30.15 (SD = 3.1) .30 (SD = .33)
Reading
accuracy
2.27 (SD = .67)
0.04 (SD = .38)
.01 (SD = .49)
N of male and female participants), mean scores at Raven test (with standard deviation in parentheses),
mean z-scores on reading speed and accuracy (with standard deviation in parentheses) for dyslexics,
reading-matched young children and skilled children. Values are based on the respective age norms.
Author's personal copy
247
C. Burani et al. / Cognition 108 (2008) 243–262
All children had normal or corrected-to-normal vision. Performance on the
Raven’s Colored Progressive Matrices was well within normal limits for all
three groups according to normative Italian data (Pruneti, 1985).
Finally, 30 adults (15 male and 15 female students at Universities of Rome), 20–32
years old, participated in the experiment.
The characteristics of the dyslexic participants’ reading disturbance were also
examined by additional tasks. Four lists of words (varying for frequency and
length) and two of non-words (varying for length) were administered (Words
and Non-words Reading test; Zoccolotti, De Luca, Di Filippo, Judica, & Spinelli, 2005b). Thirty stimuli per list were given; number of errors and reading
speed were scored. The dyslexic children were severely affected: their mean
standardised scores were less than two SDs in most conditions (see Table 2).
The children were similarly impaired for both reading speed and accuracy.
Word and non-word reading were similarly impaired. This latter finding is similar to previous observations on Italian children (Zoccolotti et al., 1999). A
Blending test measured phonological awareness. Words (and non-words) were
presented phoneme by phoneme at a rate of 1 per second. Then, the participants repeated aloud the whole-word (or non-word). Nineteen words and 19
non-words (5–6 letters) were presented (for details see Angelelli, Judica, Spinelli, Zoccolotti, & Luzzatti, 2004). The number of correctly blended items was
counted. Overall, the dyslexics’ mean performance was well within normal limits (see Table 2), consistently with previous observations indicating limited metaphonological deficits among Italian dyslexics (Brizzolara et al., 2006).
2.1.2. Materials
Two sets of 16 three-syllable pseudowords (morphological and simple) were
constructed. Morphological pseudowords were composed of a root (e.g.,
DONN-, ‘woman’) plus a derivational suffix (e.g., -ISTA, ‘-ist’) resulting in a
combination not existent in Italian (e.g., DONNISTA, ‘womanist’). Simple
pseudowords (e.g., DENNOSTO) did not include any existing morpheme.
The roots were of high-frequency (Marconi, Ott, Pesenti, Ratti, & Tavella,
Table 2
Performance of dyslexic children on the Words and Non-words Reading test and on the Blending test
Words and Non-words Reading test
HF short words
HF long words
LF short words
LF long words
Short non-words
Long non-words
Blending test
Words
Non-words
Speed
2.14
2.34
2.46
2.77
2.37
2.58
Accuracy
(SD = 3.65)
(SD = 2.73)
(SD = 2.77)
(SD = 2.64)
(SD = 2.72)
(SD = 2.84)
Whole target repetition
0.23 (SD = .80)
0.78 (SD = .88)
Values indicate z-scores based on normative samples.
0.98
2.37
2.16
1.52
2.03
1.37
(SD = 2.58)
(SD = 3.23)
(SD = 2.26)
(SD = 2.26)
(SD = 2.90)
(SD = 1.71)
Blended phoneme pairs
0.17 (SD = 0.79)
0.68 (SD = 0.58)
Author's personal copy
248
C. Burani et al. / Cognition 108 (2008) 243–262
1993), and suffixes were among the most frequent and productive in Italian
nominal and adjectival derivatives. The non-suffix final strings in simple
pseudowords had the same average frequency in word-final position as the suffixes in morphological pseudowords. The two pseudoword sets (Appendix A)
were matched for initial phoneme, syllabic structure, length, bigram frequency,
orthographic neighborhood size and orthographic complexity (Barca, Ellis, &
Burani, 2007; Burani, Barca, & Ellis, 2006).
To favour lexical reading, 16 filler words were included in the list. The total 48
stimuli were presented in two blocks in different random orders. The order of block
presentation varied. A short practice including both pseudowords and words preceded the experimental session.
2.1.3. Procedure
The stimuli, presented in lower case in the centre of a PC screen, disappeared at
the response or after 6000 or 1000 ms, for children and adults, respectively. Participants read the stimuli aloud as quickly and accurately as possible. Using a voice key
connected to the computer, in the SuperLab Pro-2.0 software, reaction times (RT)
were collected from the stimulus presentation to the onset of vocalization. The experimenter noted mispronunciations.
2.2. Results and discussion
Invalid trials due to registration errors or responses exceeding the deadline (7.5%
of total data points for adults, 11.2% for skilled children, 11.4% for younger children
and 16.9% for dyslexics) were excluded from the analyses. RTs for correct responses
in the two pseudoword sets and pronunciation errors are shown in Fig. 1, separately
for the four participating groups.
For children, by-participant ANOVAs with group (dyslexic, younger and skilled)
as unrepeated factor and pseudoword type (morphological vs. simple) as repeated
factor were carried out both on log transformed RTs data and number of pronunciation errors. In the by-item ANOVAs, pseudoword type was the unrepeated factor
and readers’ group was the repeated factor.
In the RTs analyses, there were main effects of group (F1 (2, 65) = 24.07,
p < .0001, MSE = 0.052; F2 (2, 60) = 291.92, p < .0001, MSE = 0.003), with slower
naming times for both dyslexic and younger children than for skilled children, and
no speed differences between dyslexic and young children. There was a significant
effect of pseudoword type (F1 (1, 65) = 51.32, p < .0001, MSE = 0.002; F2
(1, 30) = 10.35, p < .005, MSE = 0.008), with no group pseudoword interaction
(F1 (2, 65) = 2.28, p > .1, MSE = 0.002; F2 (2, 60) = 0.68, p > .1, MSE = 0.003).
All groups of children took advantage of morphemic units to speed up pseudoword
processing.
The analyses on errors showed significant effects of group (F1 (2, 65) = 6.47,
p < .005, MSE = 5.10; F2 (2, 60) = 10.63, p < .001, MSE = 87.7), with dyslexics less
accurate than both skilled and younger children, pseudoword type (F1 (1, 65) =
71.31, p < .0001, MSE = 2.79; F2 (1, 30) = 54.49, p < .0001, MSE = 111.78) and
Author's personal copy
C. Burani et al. / Cognition 108 (2008) 243–262
a
249
Dyslexic children
Young children
1850
Skilled children
Adults
RT(ms)
1500
1150
800
450
Morphological
Simple
Pseudoword type
b
Dyslexic children
Young children
Skilled children
Adults
40
Error %
32
24
16
8
0
Morphological
Simple
Pseudoword type
Fig. 1. Mean by-item naming times in ms (a) and percent pronunciation errors (b) as a function of
pseudoword type (morphological and simple) and reader’s ability (dyslexic children, young children,
skilled children, adults) Experiment 1.
group pseudoword interaction (F1 (2, 65) = 3.45, p < .05, MSE = 2.79; F2
(2, 60) = 3.95, p < .05, MSE = 87.71). All groups of children benefited from morphemic units to obtain better pseudoword pronunciation accuracy (p < .005 in all cases),
with a bigger advantage for dyslexics. No between-group difference was present for
morphological pseudowords; in contrast, dyslexics made more errors on simple
pseudowords than both skilled (p < .005) and younger (p < .005) children who did
not differ from each other.
Author's personal copy
250
C. Burani et al. / Cognition 108 (2008) 243–262
Based on their nearly flawless performance, only RTs were analyzed in the case of
adult readers. The ANOVA showed faster RTs for morphological than simple
pseudowords (F1 (1, 29) = 67.24, MSE = 644.77, p < .0001; F2 (1, 30) = 10.7,
MSE = 1739.8, p < .003).
Irrespective of reading skill, all groups of readers read faster and more
accurately pseudowords made up of morphemes than simple pseudowords.
The effect of morphemic constituency on reading accuracy was larger for
dyslexic readers than for all the other groups. Thus for all readers, and specifically for dyslexics, morphemes resulted in more effective reading units than
graphemes.
3. Experiment 2. Reading aloud words
In Experiment 1, pseudowords were considered, contrasting pseudowords made
up of morphemes and control pseudowords. In Experiment 2, words were considered, contrasting derived words and simple words; the aim was to answer whether
morphemes are effective processing units (and equally effective reading units for
readers of different reading skills) also when a larger reading unit (i.e., the word)
is present. Morphemic constituents (roots and suffixes) in a word may affect positively the reading performance of less skilled readers (i.e., younger and dyslexic children), for which the word represents a difficult unit to be processed as a whole.
Skilled readers, both children and adults, who are better at processing words as
whole units, should read as fast and accurately morphologically complex and simple
words.
3.1. Method
3.1.1. Participants and procedure
Same as in Experiment 1.
3.1.2. Materials
Two sets of 38 derived and 38 simple words each were selected. Derived
words were composed of a root and a derivational suffix (e.g., CASS-IERE,
‘cashier’). Simple words were not parsable in root + derivational suffix (e.g.,
CAMMELLO, ‘camel’). The words, 7–10 letters long, had medium-to-low frequency in the child written frequency count (Marconi et al., 1993), were phonologically and semantically transparent, and included frequent roots and
suffixes. All words had the most frequent Italian stress, on the penultimate syllable. The two sets (Appendix B) were matched on initial phoneme, word frequency, familiarity, length, bigram frequency, orthographic neighborhood size
and orthographic complexity.
Thirty simple and nine derived word-fillers were added to balance the number of
repetitions of suffixes and simple word endings and increase the variety of suffixes.
The materials were divided into five blocks of 23 words each, for a total of 115
Author's personal copy
C. Burani et al. / Cognition 108 (2008) 243–262
251
words. The blocks, and the words within a block, were presented in a random order.
A brief practice session preceded the experiment.
3.2. Results and discussion
Missing data accounted for 2.4% of data points for adults, 8.5% for skilled children, 10% for younger children and 16.6% for dyslexics. The RTs and pronunciation
a
Dyslexic children
Young children
1450
Skilled children
Adults
RT(ms)
1250
1050
850
650
450
Derived
Simple
Word type
Dyslexic children
Young children
Skilled children
Adults
b
20
Error %
15
10
5
0
Derived
Simple
Word type
Fig. 2. Mean by-item naming times in ms (a) and percent pronunciation errors (b) as a function of word
type (derived and simple) and reader’s ability (dyslexic children, young children, skilled children, adults)
Experiment 2.
Author's personal copy
252
C. Burani et al. / Cognition 108 (2008) 243–262
errors for the two word sets are shown in Fig. 2 separately for the four participating
groups.
For children, ANOVAs on log transformed RTs showed main effects of group (F1
(2, 65) = 26.73, p < .0001, MSE = 0.031; F2 (2, 148) = 723.76, p < .0001, MSE =
0.002), word type (derived vs. simple) (F1 (1, 65) = 17.72, p < .0001, MSE = 0.001;
F2 (1, 74) = 5.34, p < .05, MSE = 0.002) and group word type interaction (F1
(2, 65) = 6.54, p < .005, MSE = 0.001; F2 (2, 148) = 6.5, p < .005, MSE = 0.002).
Skilled children were faster than both dyslexic and younger readers, who did not differ from each other. Both dyslexic and younger children were faster in reading aloud
words including morphemic units (p < .005). No effect of word morphology was
present in skilled children.
The analyses on pronunciation errors showed a significant effect of group only (F1
(2, 65) = 18.99, p < .0001, MSE = 31.29; F2 (2, 148) = 22.65, p < .0001, MSE =
41.04), with dyslexics more accurate than young readers (p < .001) and less accurate
than skilled readers (p < .05). The group word type interaction was not significant.
The ANOVA on RTs on adults showed no differences in reading aloud derived vs.
simple words.
As expected, only less skilled readers (i.e., younger and dyslexic children) displayed faster latencies and higher accuracy in reading aloud words made up of
morphemes than simple words. In contrast, skilled readers read as fast and accurately derived and simple words, for which whole-word lexical units are
available.
4. General discussion
Confirming previous evidence (Burani, Arduino, & Marcolini, 2006; Burani &
Laudanna, 2003; Burani et al., 2002), both adults and skilled children read faster
and more accurately pseudowords composed of a root and a suffix. Less skilled (dyslexic and younger) readers also engaged in morphemic processing to supplement
grapheme-phoneme decoding. These results add up to recent data showing that lexical units are available to Italian developmental dyslexics (Barca, Burani, Di Filippo,
& Zoccolotti, 2006) as well as to children in the first grades (Orsolini, Fanari, Tosi,
De Nigris, & Carrieri, 2006). Morphemes provide lexical reading units of a largergrain size than graphemes thus reducing the limitations owed to the analytical reading processing of less skilled readers. Pseudoword reading in dyslexic and younger
readers was similar to that of skilled children and adults: Whenever possible, all
readers exploit the possibility of segmenting a new polysyllabic stimulus at a larger
size than the letter/grapheme.
Unlike skilled readers, only dyslexic and younger readers benefited from
morphological structure in reading words aloud. This new finding may appear
surprising. According to developmental models, the morphographic level of
representation is an advanced phase of development of literacy that is established on top of orthographic knowledge (Seymour, 1997); thus it should be
available to skilled readers (see reviewed evidence in Section 1). However,
Author's personal copy
C. Burani et al. / Cognition 108 (2008) 243–262
253
the characteristics of both the task and the orthography should be taken into
account. In word comprehension tasks, morphological decomposition can be
useful to understand the meaning of a newly encountered root + suffix combination (provided that the root and the suffix are known by the reader). In
contrast, in reading aloud, the reader must quickly assign the correct pronunciation to a word and morphemic parsing does not necessarily speed up processing. When a larger reading unit (the whole-word) is available, parsing a
word into smaller reading units (the morphemes) may entail costs, as well
as benefits. Parsing a familiar word into morphemic sub-parts can be more
laborious and time-consuming than full-form activation, entailing processing
costs at stages such as morpheme segmentation and composition (Schreuder
& Baayen, 1995). Processing costs consequent to morphemic parsing may also
occur at the production stage that is involved in reading aloud. In Italian,
assembling the pronunciation of a (bound) root and a suffix to obtain
whole-word pronunciation implies re-assigning to the root + suffix combination
a different stress than the stress of the root alone, with consequent planning
of a new co-articulation of the morphemic combination. In contrast, reading
aloud English compounds or suffixed words does not entail re-assigning stress
to the constituent word morphemes. Thus, English adult readers may produce
lower error rates in naming compound and suffixed words in comparison to
monomorphemic words, as well as faster naming latencies for compound than
for monomorphemic words (Inhoff, Briihl, & Schwartz, 1996), or for compounds with a high-frequency second constituent (Juhasz, Starr, Inhoff, &
Placke, 2003). Overall, for skilled readers an headstart to the morphological
reading route may occur only in those cases in which the advantages connected to parsing prevail on the parsing costs (Bertram & Hyönä, 2003;
Caramazza, Laudanna, & Romani, 1988).
In contrast to skilled readers, morpheme-based reading is efficient for less
skilled readers in processing different types of linguistic stimuli, including words,
because low-frequency words may not yet be in their orthographic lexicon, or
may be too long to be processed as a whole. This is in agreement with results
obtained on deeper orthographies. Carlisle and Stone (2005) showed that only
younger English speaking elementary students (second and third graders) had faster reading times on derived than on monomorphemic words, whereas speed of
reading the two word types did not differ for fifth and sixth graders. Using an
untimed presentation, Elbrö and Arnbak (1996) found that Danish adolescent
dyslexics read words with a semantically transparent morphological structure
(e.g., sunburn) better than words with an opaque structure (e.g., window), an
advantage not found for the control group.
Elbrö and Arnbak (1996) suggested that morpheme recognition is a compensatory reading strategy in word decoding and comprehension in dyslexia.
In a lexical decision task, Colè, Leuwers, and Sprenger-Charolles (2005)
reported stronger morphological priming effects in college dyslexics than in
controls. All these authors concluded that the morphologically based reading
strategy in dyslexics is semantic, because it involves meaning extraction from
Author's personal copy
254
C. Burani et al. / Cognition 108 (2008) 243–262
the smallest units (stems and affixes) constituting morphologically complex
words.
On the basis of the present results, we cannot adjudicate whether the advantage shown by dyslexic and younger readers is due to accessing meaningful
units (morphemes). Whereas a semantic strategy may account for performance
in tasks such as primed lexical decision (Colè et al., 2005) or untimed reading
aloud (Elbrö & Arnbak, 1996), speeded naming in a transparent orthography is
not likely to be affected by semantic variables (Burani, Arduino, & Barca,
2007). Both in adult and child Italian readers the degree of semantic interpretability of new root–suffix combinations affects lexical decision but has no
impact on reading aloud (Burani, Dovetto, Spuntarelli, & Thornton, 1999;
Burani et al., 2002).
A recent theoretical proposal focuses on the reading units’ grain size that is used
during reading acquisition. According to the psycholinguistic grain size theory (Ziegler & Goswami, 2005), children learning to read a transparent orthography rely on
small grain size units of processing such as single letters and phonemes, even when
large-unit information is available. In contrast, readers of inconsistent orthographies
are ‘‘forced” to develop multiple grain size mappings. This view does not consider
morphemes in explaining the grain sizes used in decoding, although the authors
acknowledge that salient units of different grain size may emerge as bigger units that
are phonologically more accessible than single graphemes (Goswami & Ziegler,
2006).
The present results indicate that morphemes may develop as orthographic
and phonological salient reading units not only when smaller grain sizes are
inconsistent or unavailable, as in unpointed Hebrew spelling (Frost, 2006),
but also when smaller reading units are easily available as in a transparent
orthography. However, even in a transparent orthography, whole-word units
can be more salient and efficient reading units than morphemes for readers
not suffering from processing limitations. Further work on readers of different
abilities of Italian and other languages is required to confirm the present findings as well as to extend our knowledge of the role of morphemes in reading
aloud across different orthographies.
Acknowledgments
This study was supported by a grant from the International Dyslexia Association.
Cristina Burani and Pierluigi Zoccolotti are members of the Marie Curie RTN ‘‘Language and Brain” (European Commission, MRTN-CT-2004-512141). The authors
thank Despina Paizi for useful comments.
Author's personal copy
Appendix A
Pseudowords used in Experiment 1
BIGR
FREQ
N-SIZE
CONT
RULE
RT
(% ERR)
DYSL
RT
(% ERR)
YOUNG
RT
(% ERR)
SKILLED
RT
(% ERR)
ADULTS
Morphological
bagnezza (bathness)
cartismo (paperism)
codismo (tailism)
corpezza (bodyness)
cuoroso (heartous)
donnista (womanist)
erbista (herbist)
gitista (hikist)
guerroso (warous)
mammista (motherist)
pallismo (ballism)
sonnezza (sleepyness)
stelloso (starrous)
stradoso (streetous)
vetrezza (glassness)
zampismo (pawism)
8
8
7
8
7
8
7
7
8
8
8
8
8
8
8
8
9.97
10.61
10.48
10.30
10.84
11.04
10.69
10.95
10.62
10.79
10.65
10.51
11.15
10.98
10.42
9.77
0
1
0
0
1
0
0
0
0
1
1
0
0
0
0
0
1
1
1
1
1
0
0
1
1
0
0
0
0
0
0
0
1885 (.18)
1360 (.06)
1513 (.06)
1368 (.24)
1796 (.18)
1073 (.06)
949 (.12)
1648 (.12)
1440 (0)
1283 (.12)
1311 (0)
1688 (.18)
1584 (.18)
1910 (.12)
1509 (.24)
1220 (.06)
1686
1092
1725
1928
1903
1129
1246
1892
1543
1579
1432
1420
1671
1730
1568
1456
821 (.09)
806 (0)
765 (.06)
857 (.03)
837 (.24)
808 (0)
629 (0)
1063 (.09)
773 (.03)
762 (.03)
701 (.06)
799 (.09)
864 (.03)
853 (0)
705 (.12)
807 (.03)
594
590
586
624
673
599
568
714
658
607
576
656
640
661
592
654
Mean
SD
7.75
0.45
10.61
0.37
0.25
0.45
0.44
0.51
1471 (.12)
262 (.07)
1562 (.11)
258 (.08)
803 (.06)
94 (.06)
625 (.01)
42 (.03)
Simple
bognezzo
cudosta
cuonede
cuprezzo
8
7
7
8
9.70
10.67
10.87
9.98
0
0
0
0
1
1
1
1
1133
1553
1691
1562
1550
1903
2514
1438
905 (.21)
903 (.18)
1049 (.18)
873 (.06)
640
680
707
646
(.59)
(.47)
(.18)
(.29)
(.06)
(.12)
(.24)
(.06)
(.24)
(.12)
(0)
(.24)
(.24)
(.06)
(.06)
(0)
(.06)
(.06)
(.18)
(.12)
(.18)
(.18)
(.24)
(.18)
(.03)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(.03)
(0)
(0)
(0)
(.13)
(0)
C. Burani et al. / Cognition 108 (2008) 243–262
LETT
(.13)
(.03)
(.07)
(0)
255
(continued on next page)
Author's personal copy
256
Appendix A (continued)
BIGR
FREQ
N-SIZE
CONT
RULE
RT
(% ERR)
DYSL
RT
(% ERR)
YOUNG
RT
(% ERR)
SKILLED
RT
(% ERR)
ADULTS
curtosta
dennosto
ermosto
getosto
guarrede
memmosto
pillosta
sannezzo
stollede
strodasa
vatrezzo
zemposta
8
8
7
7
8
8
8
8
8
8
8
8
10.83
11.19
10.99
10.86
10.62
10.71
11.07
10.34
11.18
10.75
10.37
10.08
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
1
0
0
0
0
0
0
0
1717
2012
1587
2160
1383
2222
1670
1617
2071
1658
1639
1863
2010
1798
1532
1710
1617
1552
1600
2059
1675
2073
2145
1842
991 (.12)
806 (.32)
698 (0)
791 (.44)
864 (.18)
927 (.18)
838 (.15)
756 (.21)
1001 (.26)
1070 (.32)
823 (.21)
811 (.18)
672
666
585
658
701
660
622
703
719
765
673
664
Mean
SD
7.75
0.45
10.64
0.44
0
0
0.44
0.51
1721 (.35)
286 (.13)
882 (.20)
105 (.1)
673 (.06)
42 (.08)
(.41)
(.53)
(.24)
(.18)
(.47)
(.41)
(.35)
(.41)
(.29)
(.18)
(.41)
(.18)
(.29)
(.06)
(.06)
(.35)
(.29)
(.18)
(.18)
(.41)
(.35)
(.29)
(.18)
(.06)
1814 (.22)
289 (.11)
(.13)
(.03)
(0)
(.3)
(.03)
(0)
(0)
(.07)
(0)
(0)
(.07)
(.07)
Note. LETT, word length in letters; BIGR, mean bigram frequency, log transformed (natural logarithm); N-SIZE, orthographic neighborhood size; CONT
RULE, number of c, g and sc letters, that need the following letter context to assign the correct pronunciation; DOUBLE LETT, number of double letters;
RT, mean pseudoword naming reaction time; % ERR, mean % error rate; DYSL, dyslexic children; YOUNG, younger children; SKILLED, skilled children;
ADULTS, adult readers.
C. Burani et al. / Cognition 108 (2008) 243–262
LETT
Author's personal copy
Appendix B
Words used in Experiment 2
BIGR
FREQ
LETT
N-SIZE
CONT
RULE
RT
(% ERR)
DYSL
RT
(% ERR)
YOUNG
RT
(% ERR)
SKILLED
RT
(% ERR)
ADULTS
27
12
10
59
17
6
47
33
19
5
7
6.27
6.10
6.23
6.63
6.30
6.27
5.93
6.85
6.40
6.53
6.63
10.81
10.88
10.93
11.44
11.19
11.15
10.92
11.30
10.22
10.90
10.85
7
8
9
8
7
8
10
8
8
7
8
2
6
3
2
5
2
1
2
1
1
0
0
0
1
1
1
1
3
0
1
0
1
1207 (.29)
1265 (.06)
1160 (.12)
892 (0)
1320 (0)
1142 (.06)
1076 (.12)
1098 (.18)
1236 (0)
1102 (0)
1201 (.18)
1173 (.35)
1269 (.06)
1172 (0)
1144 (0)
1089 (.06)
1362 (0)
1343 (.06)
1195 (.24)
1079 (0)
1192 (.06)
986 (.24)
718
643
621
593
615
636
708
685
658
716
630
553
588
585
557
598
587
573
582
560
533
626
23
33
41
11
18
6
29
32
14
6
12
22
11
34
14
6.85
6.10
6.95
6.95
6.07
5.43
6.73
5.67
6.30
6.43
6.53
5.83
6.13
6.70
6.55
10.74
10.65
10.81
10.88
10.45
10.58
11.00
10.81
10.44
10.71
10.97
11.08
11.22
11.20
11.22
9
10
10
8
10
7
8
8
10
7
10
8
9
9
8
3
0
1
2
0
1
1
1
1
6
0
1
1
2
1
1
0
1
0
3
0
2
1
1
0
2
0
1
2
0
1125 (.06)
1105 (0)
1214 (0)
1054 (.06)
1229 (0)
1057 (0)
1072 (0)
1042 (.12)
945 (.12)
1123 (.06)
1231 (.29)
925 (.12)
1517 (0)
1183 (.06)
934 (.12)
1093 (.12)
1372 (.06)
1142 (0)
1358 (.12)
1336 (0)
985 (.06)
1016 (0)
1157 (.06)
1236 (0)
1190 (.06)
1584 (.18)
1108 (.06)
1264 (.24)
1002 (0)
1183 (.24)
671 (0)
749 (0)
807 (0)
744 (.09)
716 (0)
669 (.03)
606 (0)
656 (0)
667 (.03)
718 (0)
872 (0)
604 (.12)
731 (.03)
714 (.03)
690 (.12)
(continued on
(.12)
(.03)
(.03)
(.03)
(0)
(0)
(.06)
(.09)
(0)
(0)
(.03)
(0)
(.03)
(.03)
(.03)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
613 (0)
592 (0)
614 (0)
618 (0)
592 (0)
544 (0)
611 (0)
571 (0)
577 (0)
603 (0)
627 (0)
606 (0)
635 (0)
542 (0)
618 (.03)
next page)
257
guerriero
importanza
insegnante
libreria
linguaggio
lupetto
maglione
mancanza
negoziante
nuotata
parolaccia
passante
pecorella
pescatore
piattino
driver
ballet
stroll
singer
map
cashier
knowledge
dentist
sweetness
florist
ice cream
man
warrior
importance
teacher
bookstore
language
cub scout
pullover
lack
shopkeeper
swim
dirty word
passer-by
little sheep
fisherman
saucer
FAM
C. Burani et al. / Cognition 108 (2008) 243–262
Derived
autista
balletto
camminata
cantante
cartina
cassiere
conoscenza
dentista
dolcezza
fioraio
gelataio
WORD
FREQ
Author's personal copy
258
Appendix B (continued)
terriccio
trenino
tristezza
vecchiaia
vetrina
vicinanza
villetta
little ladder
signal
miss
hope
instrument
player
mold
small train
sadness
oldness
shop
window
proximity
cottage
Mean
SD
Simple
assassino
battaglia
cammello
castagna
coccinella
continente
corteccia
discorso
documento
funerale
ginocchio
murderer
battle
camel
chestnut
ladybug
continent
bark
discourse
document
funeral
knee
FAM
BIGR
FREQ
LETT
N-SIZE
CONT
RULE
RT
(% ERR)
DYSL
RT
(% ERR)
YOUNG
RT
(% ERR)
SKILLED
RT
(% ERR)
ADULTS
5
42
56
38
5
5.97
6.17
6.63
6.40
6.70
11.11
10.96
11.02
10.93
11.18
8
7
9
8
9
1
5
3
1
1
2
1
1
0
0
1254
1155
1252
1098
1453
(.06)
(0)
(.12)
(0)
(.18)
1173
1148
1127
1245
1356
(.12)
(.06)
(.12)
(.06)
(0)
784
703
779
723
706
(.03)
(0)
(0)
(.03)
(0)
572
559
605
564
550
(0)
(0)
(0)
(0)
(0)
6
32
28
8
43
4.63
6.37
7.00
6.43
6.50
10.92
11.25
10.52
10.64
11.02
9
7
9
9
7
0
1
1
0
3
2
0
0
2
0
1014 (.12)
924 (.12)
1441 (.06)
1213 (0)
840 (.06)
1268
1414
1144
1417
1034
(.06)
(.12)
(.06)
(.12)
(0)
677
691
788
676
604
(0)
(.03)
(0)
(.03)
(0)
625
570
576
594
548
(0)
(0)
(0)
(0)
(0)
11
7
6.55
6.65
10.71
10.98
9
8
1
3
1
0
1476 (.24)
1153 (.12)
1467 (.06)
1087 (.24)
788 (.15)
741 (.03)
612 (0)
589 (0)
22
15
6.35
0.46
10.91
0.27
8.37
1
1.74
1.59
0.84
0.89
1151 (.08)
159 (.08)
1208 (.09)
142 (.09)
697 (.03)
64 (.04)
586 (0)
27 (.01)
5
38
7
29
19
38
16
56
15
11
50
6.85
6.95
6.13
6.53
5.97
6.70
5.43
6.33
6.33
6.33
6.53
10.99
10.71
10.86
10.86
11.10
11.40
10.96
11.01
10.68
11.03
10.8
9
9
8
8
10
10
9
8
9
8
9
3
1
3
3
1
2
0
1
2
1
3
0
1
1
2
3
1
3
1
0
0
1
1135 (0)
897 (.12)
1071 (0)
1150 (0)
1519 (.18)
1843 (.06)
1272 (.06)
1026 (0)
1250 (0)
1448 (.06)
1252 (0)
1076 (.18)
989 (.06)
923 (0)
1175 (0)
1267 (.06)
1511 (.12)
1202 (.12)
1432 (0)
1030 (.06)
1498 (.06)
1606 (.12)
614
605
617
680
755
768
713
746
713
721
702
589
568
586
588
620
595
575
586
576
553
610
(0)
(0)
(0)
(0)
(.03)
(.03)
(.06)
(.03)
(0)
(0)
(0)
(0)
(0)
(.03)
(0)
(0)
(0)
(0)
(0)
(.03)
(0)
(.03)
C. Burani et al. / Cognition 108 (2008) 243–262
scaletta
segnale
signorina
speranza
suonatore
WORD
FREQ
Author's personal copy
Mean
SD
gorilla
address
interval
labyrinth
blackboard
earthworm
matress
metal
daffodil
nostalgia
landscape
paradise
wig
property
papyrus
priest
sausage
shelf
syringe
salary
turtle
tragedy
sunset
avalanche
suitcase
shame
event
5
19
30
16
34
8
7
22
5
24
17
24
12
8
5
16
8
15
23
10
53
23
18
11
28
22
37
6.85
6.57
6.41
5.73
6.85
5.90
6.77
6.33
4.30
5.80
6.17
6.90
6.10
5.67
5.07
5.93
6.50
5.83
6.40
6.30
6.57
5.93
6.60
5.97
6.95
6.20
6.30
10.98
10.41
10.98
10.86
10.83
10.64
11.12
11.10
10.84
10.73
11.23
11.02
10.36
10.98
10.74
10.75
10.63
10.83
10.75
10.98
10.43
10.73
11.11
10.82
10.87
10.31
10.82
7
9
10
9
7
8
9
7
7
9
8
8
8
10
9
9
9
8
7
9
9
8
8
7
7
8
7
0
2
1
1
2
0
1
1
3
1
1
1
0
1
1
1
0
1
0
1
0
1
2
1
1
3
2
1
0
0
0
1
1
0
0
1
1
0
0
2
0
1
1
2
2
1
0
1
1
0
1
1
2
1
1396 (0)
1145 (.06)
1061 (0)
1299 (.06)
923 (0)
1229 (.06)
2107 (.29)
1391 (.24)
1249 (.06)
1325 (.06)
1249 (.06)
1082 (.06)
1109 (.18)
1409 (0)
1510 (.06)
1634 (.06)
1395 (0)
1262 (0)
1382 (.12)
1613 (.06)
1124 (.06)
1918 (.24)
1135 (.06)
1242 (.29)
1040 (0)
1291 (0)
1304 (.12)
1188 (.06)
1162 (0)
1164 (0)
1209 (.18)
1008 (.06)
1343 (0)
1305 (.12)
1300 (.29)
1510 (.12)
1304 (.12)
1314 (.12)
1308 (.12)
1078 (.18)
1474 (.41)
1447 (.24)
2348 (.29)
1240 (.47)
1156 (0)
2014 (.24)
2229 (.12)
903 (0)
1361 (.18)
1082 (.06)
1214 (.24)
1174 (.06)
1222 (0)
1334 (.06)
652
673
728
639
612
726
686
653
681
799
648
605
618
691
807
760
721
742
753
860
606
813
678
644
657
655
645
(0)
(0)
(0)
(.09)
(0)
(.03)
(0)
(.12)
(.12)
(.06)
(0)
(0)
(0)
(.24)
(.09)
(.09)
(.09)
(.03)
(.03)
(0)
(0)
(.06)
(0)
(.12)
(.03)
(0)
(0)
21
14
6.24
0.54
10.85
0.23
8.37
0.94
1.29
0.96
0.89
0.83
1308 (.07)
259 (.08)
1318 (.12)
312 (.11)
694 (.03)
65 (.05)
569
565
608
610
581
588
599
583
601
587
570
570
588
632
655
570
587
582
618
609
611
586
604
595
577
570
572
(0)
(0)
(0)
(0)
(.03)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(.03)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
C. Burani et al. / Cognition 108 (2008) 243–262
gorilla
indirizzo
intervallo
labirinto
lavagna
lombrico
materasso
metallo
narciso
nostalgia
panorama
paradiso
parrucca
patrimonio
pergamena
sacerdote
salsiccia
scaffale
siringa
stipendio
tartaruga
tragedia
tramonto
valanga
valigia
vergogna
vicenda
590 (0)
21 (.01)
Note. WORD FREQ, word frequency out of 1 million occurrences; FAM, rated word familiarity (7-point scale); LETT, word length in letters; BIGR FREQ,
mean bigram frequency, log transformed (natural logarithm); N-SIZE, orthographic neighborhood size; CONT RULE, number of c, g and sc letters, that
need the following letter context to assign the correct pronunciation; RT, mean pseudoword naming reaction time; % ERR, mean% error rate; DYSL,
dyslexic children; YOUNG, younger children; SKILLED, skilled children; ADULTS, adult readers.
259
Author's personal copy
260
C. Burani et al. / Cognition 108 (2008) 243–262
References
Angelelli, P., Judica, A., Spinelli, D., Zoccolotti, P., & Luzzatti, C. (2004). Characteristics of writing
disorders in Italian dyslexic children. Cognitive and Behavioral Neurology, 17, 18–31.
Barca, L., Burani, C., Di Filippo, G., & Zoccolotti, P. (2006). Italian developmental dyslexic and proficient
readers: Where are the differences? Brain and Language, 98, 347–351.
Barca, L., Ellis, A. W., & Burani, C. (2007). Context-sensitive rules and word naming in Italian children.
Reading and Writing: An Interdisciplinary Journal, 20, 495–509.
Bertram, R., & Hyönä, J. (2003). The length of a complex word modifies the role of morphological
structure: Evidence from eye movements when reading short and long Finnish compounds. Journal of
Memory and Language, 48, 615–634.
Bertram, R., Laine, M., & Virkkala, M. M. (2000). The role of derivational morphology in vocabulary
acquisition: Get by with a little help from my morpheme friends. Scandinavian Journal of Psychology,
4, 2–15.
Brizzolara, D., Pecini, C., Chilosi, A., Cipriani, P., Gasperini, F., Mazzotti, S., et al. (2006). Do
phonological and rapid automatized naming deficits differentially affect dyslexic children with and
without a history of language delay? A study on Italian dyslexic children. Cognitive and Behavioural
Neurology, 19, 141–149.
Burani, C., Arduino, L. S., & Barca, L. (2007). Frequency, not age of acquisition affects Italian word
naming. European Journal of Cognitive Psychology, 19, 828–866.
Burani, C., Arduino, L. S., & Marcolini, S. (2006). Naming morphologically complex pseudowords: A
headstart for the root? The Mental Lexicon: A Journal of Cross-Disciplinary Research, 1, 299–327.
Burani, C., Barca, L., & Ellis, A. W. (2006). Orthographic complexity and word naming in Italian: Some
words are more transparent than others. Psychonomic Bulletin and Review, 13, 346–352.
Burani, C., Bimonte, D., Barca, L., & Vicari, S. (2006). Word morphology and lexical comprehension in
Williams Syndrome. Brain and Language, 99, 112–113.
Burani, C., Dovetto, F. M., Spuntarelli, A., & Thornton, A. M. (1999). Morpho-lexical naming of new
root-suffix combinations: The role of semantic interpretability. Brain and Language, 68, 333–339.
Burani, C., & Laudanna, A. (2003). Morpheme-based lexical reading: Evidence from pseudo-word
naming. In E. Assink & D. Sandra (Eds.), Reading complex words: Cross-language studies
(pp. 241–264). Kluwer: Dordrecht.
Burani, C., Marcolini, S., & Stella, G. (2002). How early does morpho-lexical reading develop in readers of
a shallow orthography? Brain and Language, 81, 568–586.
Burani, C., & Thornton, A. M. (2003). The interplay of root, suffix and whole-word frequency in
processing derived words. In H. R. Baayen & R. Schreuder (Eds.), Morphological Structure in
Language Processing (pp. 157–208). Mouton de Gruyter: Springer.
Caramazza, A., Laudanna, A., & Romani, C. (1988). Lexical access and inflectional morphology.
Cognition, 28, 297–332.
Carlisle, J. F., & Nomabody, D. M. (1993). Phonological and morphological awareness in first grades.
Applied Psycholinguistics, 14, 177–195.
Carlisle, J. F., & Stone, C. A. (2003). The effects of morphological structure on children’s reading of
derived words in English. In E. Assink & D. Sandra (Eds.), Reading complex words: Cross-language
studies (pp. 26–49). Kluwer: Dordrecht.
Carlisle, J. F., & Stone, C. A. (2005). Exploring the role of morphemes in word reading. Reading Research
Quarterly, 40, 428–449.
Casalis, S., & Louis-Alexandre, M.-F. (2000). Morphological analysis, phonological analysis and
learning to read French: A longitudinal study. Reading and Writing: An Interdisciplinary Journal,
12, 303–335.
Casalis, S., Sopo, D., & Colé, P. (2004). Morphological awareness in developmental dyslexia. Annals of
Dyslexia, 54, 114–138.
Colè, P., Leuwers, C., & Sprenger-Charolles, L. (2005). A morphology-based compensatory reading
strategy in adult dyslexics. In XIV conference of the European society for cognitive psychology. Leiden,
31st August–3rd September.
Author's personal copy
C. Burani et al. / Cognition 108 (2008) 243–262
261
Cornoldi, C., Colpo, G., & Gruppo, M. T. (1981). Prove di lettura MT. Guida all’uso. Firenze:
Organnizzazioni Speciali.
De Luca, M., Borrelli, M., Judica, A., Spinelli, D., & Zoccolotti, P. (2002). Reading words and pseudowords: An eye movement study of developmental dyslexia. Brain and Language, 80, 617–626.
De Luca, M., Di Pace, E., Judica, A., Spinelli, D., & Zoccolotti, P. (1999). Eye-movements patterns in
linguistic and non-linguistic tasks in developmental surface dyslexia. Neuropsychologia, 37, 1407–1420.
Elbrö, C., & Arnbak, E. (1996). The role of morpheme recognition and morphological awareness in
dyslexia. Annals of Dyslexia, 46, 209–240.
Feldman, L. B., Rueckl, J., DiLiberto, K., Pastizzo, M., & Vellutino, F. R. (2002). Morphological analysis
by child readers as revealed by the fragment completion task. Psychonomic Bulletin and Review, 9,
529–535.
Frost, R. (2006). Becoming literate in Hebrew: The grain size hypothesis and Semitic orthographic
systems. Developmental Science, 9, 439–440.
Goswami, U., & Ziegler, J. C. (2006). Fluency, phonology and morphology: A response to the
commentaries on becoming literate in different languages. Developmental Science, 9, 451–453.
Hutzler, F., & Wimmer, H. (2004). Eye movements of dyslexic children when reading in a regular
orthography. Brain and Language, 89, 235–242.
Inhoff, A. W., Briihl, D., & Schwartz, J. (1996). Compound word effects differ in reading, on-line naming,
and delayed naming tasks. Memory and Cognition, 24, 466–476.
Juhasz, B. J., Starr, M. S., Inhoff, A. W., & Placke, L. (2003). The effects of morphology on the processing
of compound words: Evidence from naming, lexical decision and eye fixations. British Journal of
Psychology, 94, 223–244.
Laine, M., Vainio, S., & Hyönä, J. (1999). Lexical access routes to nouns in a morphologically rich
language. Journal of Memory and Language, 40, 109–135.
Laxon, V., Rickard, M., & Coltheart, V. (1992). Children read affixed words and non-words. British
Journal of Psychology, 83, 407–423.
Leong, C. K., & Parkinson, M. E. (1995). Processing of English morphological structure by poor readers.
In C. K. Leong & R. M. Joshi (Eds.), Developmental and Acquired Dyslexia (pp. 237–261). Dordrecht:
Kluwer.
Mann, V., & Singson, M. (2003). Linking morphological knowledge to English decoding ability: Large
effects of little suffixes. In E. Assink & D. Sandra (Eds.), Reading complex words: Cross-language
studies (pp. 1–25). Dordrecht: Kluwer.
Marconi, L., Ott, M., Pesenti, E., Ratti, D., & Tavella, M. (1993). Lessico elementare. Dati statistici
sull’italiano letto e scritto dai bambini delle elementari. Bologna: Zanichelli.
Orsolini, M., Fanari, R., Tosi, V., De Nigris, B., & Carrieri, R. (2006). From phonological recoding to
lexical reading: A longitudinal study on reading development in Italian. Language and Cognitive
Processes, 576–607.
Pruneti, C. A. (1985). Dati normativi del test P M. 47 Coloured su un campione di bambini italiani.
Bollettino di psicologia applicata, 176, 27–35.
Rayner, K., & McConkie, G. W. (1976). What guides a reader’s eye movements? Vision Research, 16,
829–837.
Schreuder, R., & Baayen, R. H. (1995). Modelling morphological processing. In L. B. Feldman (Ed.),
Morphological aspects of language processing (pp. 131–154). Hove: Erlbaum.
Seymour, P. H. K. (1997). Foundations of orthographic development. In C. Perfetti, L. Rieben, & M.
Fayol (Eds.), Learning to Spell (pp. 319–337). Hillsdale, NJ: Erlbaum.
Spinelli, D., De Luca, M., Di Filippo, G., Mancini, M., Martelli, M., & Zoccolotti, P. (2005). Length effect
in word naming latencies: Role of reading experience and reading deficit. Developmental Neuropsychology, 27, 217–235.
Traficante, D., & Burani, C. (2003). Visual processing of Italian verbs and adjectives: The role of
inflectional family size. In H. R. Baayen & R. Schreuder (Eds.), Morphological structure in language
processing (pp. 45–64). Berlin: Mouton de Gruyter.
Verhoeven, L., & Perfetti, C. (2003). Introduction to this special issue: The role of morphology in learning
to read. Scientific Studies of reading, 7, 209–217.
Author's personal copy
262
C. Burani et al. / Cognition 108 (2008) 243–262
Ziegler, J. C., & Goswami, U. (2005). Reading acquisition, developmental dyslexia and skilled reading
across languages: A psycholinguistic grain size theory. Psychological Bulletin, 131, 3–29.
Zoccolotti, P., De Luca, M., Di Pace, E., Gasperini, F., Judica, A., & Spinelli, D. (2005a). Word length
effect in early reading and in developmental dyslexia. Brain and Language, 93, 369–373.
Zoccolotti, P., De Luca, M., Di Filippo, G., Judica, A., & Spinelli, D. (2005b). Prova di lettura di parole e
non parole. (Words and Non-words Reading test). Roma: IRCCS Fondazione Santa Lucia, (<http://
www.hsantalucia.it/modules.php?name=Content&pa=showpage&pid = 1032>).
Zoccolotti, P., De Luca, M., Di Pace, E., Judica, A., Orlandi, M., & Spinelli, D. (1999). Markers of
developmental surface dyslexia in a language (Italian) with high grapheme-phoneme correspondence.
Applied Pyscholinguistics, 20, 191–216.