PM R 8 (2016) 254-267

www.pmrjournal.org

Narrative Review

Evidence for Intensive Aphasia Therapy: Consideration of

Theories From Neuroscience and Cognitive Psychology
Jade K. Dignam, BSpPath, Amy D. Rodriguez, PhD, David A. Copland, PhD

Abstract

Treatment intensity is a critical component to the delivery of speech-language pathology and rehabilitation services. Within
aphasia rehabilitation, however, insufficient evidence currently exists to guide clinical decision making with respect to the
optimal treatment intensity. This review considers perspectives from 2 key bodies of research, the neuroscience and cognitive
psychology literature, with respect to the scheduling of aphasia rehabilitation services. Neuroscience research suggests that
intensive training is a key element of rehabilitation and is necessary to achieve functional and neurologic changes after a stroke
occurs. In contrast, the cognitive psychology literature suggests that optimal long-term learning is achieved when training is
provided in a distributed or nonintensive schedule. These perspectives are evaluated and discussed with respect to the current
evidence for treatment intensity in aphasia rehabilitation. In addition, directions for future research are identified, including
study design, methods of defining and measuring treatment intensity, and selection of outcome measures in aphasia
rehabilitation.

Introduction Consequently, it is important to distinguish between

measures of treatment acquisition and long-term main-
Sufficient evidence now exists to support the general tenance when considering rehabilitation outcomes.
efficacy of aphasia therapy [1-6]. However, limited Within the aphasiology literature, studies investi-
evidence exists to guide the optimal scheduling of gating the role of treatment intensity have produced
therapy services for people who have aphasia after a conflicting results [1]. International clinical guidelines
stroke. Evidence from the neurosciences literature, and best practice recommendations for the clinical
which is based primarily on animal models of motor management of stroke advocate for the provision of
recovery after stroke, suggests that intensive therapy is intensive rehabilitation services [17-21]; however, few
necessary to elicit significant neurologic and behavioral direct recommendations are made regarding the
changes [7-11]. In contrast, the cognitive psychology optimal treatment intensity for aphasia rehabilitation.
literature, which primarily consists of studies of healthy In clinical practice, speech-language pathologists are
adults, suggests that nonintensive or distributed frequently required to make decisions regarding the
learning schedules result in superior learning outcomes scheduling of aphasia therapy, including the amount,
[12]. The neuroscience and cognitive psychology litera- intensity, and duration of therapy required. However,
ture seemingly differ with respect to the maintenance of these decisions are often based on service delivery
learning gains, rather than the acquisition of learning. factors, such as staffing and budget constraints, and
Whereas neuroscience research asserts that intensive may not be informed by empirical evidence.
training facilitates acquisition, studies have demon- A recent randomized controlled trial conducted in
strated that treatment gains may not be consistently the United Kingdom, Assessing Communication Therapy
maintained upon the cessation of intensive training in the North West (ACT NoW) [22], has brought into
[13-15]. In contrast, the cognitive psychology literature question the effectiveness of current service delivery
suggests that a distributed training schedule promotes models for aphasia rehabilitation in the early stages of
the long-term retention of learned information [12,16]. stroke recovery and has prompted the need for further

1934-1482/$ - see front matter ª 2016 by the American Academy of Physical Medicine and Rehabilitation
http://dx.doi.org/10.1016/j.pmrj.2015.06.010
 J.K. Dignam et al. / PM R 8 (2016) 254-267 255

research. Bowen et al [22] investigated the effectiveness optimal treatment intensity is an important research
of communication therapy delivered in the first 4 months question in the broader, multidisciplinary rehabilitation
after stroke by comparing “best practice” speech- context, with implications for consumers, clinicians,
language pathology intervention, delivered at an service providers, and policy makers. Consequently, this
average intensity of 1.4 hours per week (a mean total of review also has clinical implications for the multidisci-
18 hours over 13 weeks), with a similar amount of social plinary rehabilitation and management of stroke.
contact from an employed visitor. The study was unable
to differentiate between treatment and control groups Definition of Intensity
on the primary outcome measure of functional commu-
nication ability at 6 months follow-up, suggesting no Therapy intensity is a multifaceted construct that,
additional benefit of speech-language pathology inter- until recently, has been difficult to define in speech-
vention, when delivered at this low intensity, over that language pathology research. Within the aphasiology
of social contact. Although the ACT NoW study did not literature, intensity is commonly used to describe the
explicitly aim to evaluate treatment intensity, the study frequency of therapy, in number of therapy hours per
has initiated debate on the effectiveness of current week. This definition is in contrast to studies of motor
service delivery models in aphasia rehabilitation [23,24]. recovery after stroke, which may consider the amount
Therapy intensity is a fundamental component of the of effort expended during a therapy session or the
delivery of speech and language services and conse- number of times a particular task is repeated. Hinckley
quently is a pertinent area of research. Furthermore, in and Carr [25] describe intensive therapy as “more
view of the significant negative consequences of aphasia treatment provided over a shorter amount of time.”
and the increasing demands on health care services, it is However, there is great variability within the aphasia
important that we address the efficacy of service de- literature as to what constitutes intensive therapy, with
livery models in aphasia rehabilitation. studies ranging from 5 hours per week [26-28] to more
In this review we aim to (1) evaluate and synthesize than 20 hours per week [25,29,30]. Consequently, there
key findings from the neurosciences literature with re- is a need for consistent use of terminology and clear
gard to treatment intensity and its relationship with reporting of treatment variables in aphasiology
functional and neurologic outcomes in rehabilitation, research. Warren et al [31] suggest the use of a stan-
(2) analyze key findings from the cognitive psychology dardized model for defining intensity in which cumula-
literature and consider the effect of training schedules tive treatment intensity consists of dose form (ie, the
on learning outcomes in healthy humans, (3) incorpo- task in which the teaching episode is delivered), dose
rate these perspectives with our knowledge and un- (ie, the number of teaching episodes per session), dose
derstanding of service delivery models and treatment frequency (ie, the number of times a dose is provided
intensity in aphasia rehabilitation, and (4) identify lim- per day and per week) and total intervention duration
itations in the current evidence base for treatment in- (ie, the period over which an intervention is provided).
tensity in aphasia rehabilitation and propose a research To date, few clinical studies have provided the infor-
agenda for future studies. mation required to be able to calculate cumulative
A comprehensive review of the stroke and aphasia treatment intensity based on this model [32,33]. For the
rehabilitation literature, as well as literature pertaining purpose of this review, the amount of therapy provided,
to learning theory and neuroplasticity, was undertaken. or therapy dosage, is defined as the total number of
Studies evaluating treatment intensity in adults with therapy hours, whereas intensity of treatment is defined
poststroke aphasia were considered. Articles were as the number of therapy hours per unit time. The
accessed via multiple databases (the Cochrane library, duration of therapy is defined as the total period of
Web of Science, Scopus, and CINAHL), and search terms intervention, measured in weeks or months. It is
included “aphasia,” “intensity,” “neuroplasticity,” acknowledged that, increasingly, clinical studies in
“therapy,” “rehabilitation,” “distributed practice,” aphasia and stroke rehabilitation are calling for greater
“spacing effect,” and “learning theory.” In addition, the control and reporting of treatment variables. This
bibliographies of relevant studies were reviewed to increased rigor is necessary to more accurately delin-
identify further research articles. Relevant articles eate the effects of treatment intensity on communica-
published in English prior to November 2014 were tion outcomes [32].
included in the review.
This is the first narrative style review to consider key Neuroscience and Learning
findings from both the neurosciences and cognitive psy-
chology literature and interpret these findings with Principles of Experience-Dependent
respect to the current evidence base for treatment in- Neuroplasticity
tensity in aphasia rehabilitation. This review has impli-
cations for clinical practice and service delivery models It has been argued that rehabilitation is in essence a
in aphasia rehabilitation. Furthermore, establishing learning experience [34]. Consequently, an increased
256 Intensity of Aphasia Therapy

knowledge and understanding of the neurologic reorganization in response to intensive training has been
mechanisms underlying learning may help to develop replicated in a number of animal studies [36,37]. In
new and effective treatment techniques and to inform contrast, a study conducted by Luke et al [38] failed to
our models of rehabilitation. Neuroplasticity describes find evidence of neuroplasticity in adult rats when
the adaptive capacity of the central nervous system skilled reaching training was provided nonintensively
in response to internal and external influences and is (approximately 60 repetitions per day). Furthermore,
the means through which we encode new experiences Luke et al [38] found that the rate of acquisition and
and learn new skills and behaviors [11]. It is also the level of behavioral proficiency achieved with non-
means through which the damaged brain relearns intensive training was less than that reported in other
skills and behaviors, for example, in response to studies of skilled reaching, which employed intensive
rehabilitation [11]. training regimes [36,37].
Keefe [14] suggests that the functional outcome of MacLellan et al [39] suggest that there is a critical
stroke is a combination of the individual’s premorbid intensity of rehabilitation that must be met to evoke
function, the neurologic changes initiated by the stroke cortical reorganization and behavioral changes.
itself, and the effect of environmental and behavioral MacLellan et al [39] evaluated the functional changes
influences on brain function and organization. Conse- that occurred in adult male rats in response to a
quently, the ultimate goal of rehabilitation is to behavioral training program. These researchers also
manipulate the environmental and behavioral experi- measured levels of brain-derived neurotrophic factor, a
ences of the individual after the stroke to guide protein thought to assist neural reorganization, in the
neurologic reorganization in a way that facilitates re- motor cortex and the hippocampus of rats before and
covery. Kleim and Jones [11] proposed 10 principles of after the intervention. Rats were randomly assigned to
experience-dependent neuroplasticity to inform reha- an unlimited environmental enrichment group, which
bilitation, and these principles have been further eval- involved unlimited reaching training over a 4-hour
uated by Raymer et al [35] with respect to the period, 5 days per week for 6 weeks, or a limited envi-
rehabilitation of aphasia. These principles outline the ronmental enrichment group, which involved restricted
critical features of the learning experience required to reaching training over the same intervention duration.
drive cortical reorganization and facilitate recovery. For The study found significantly greater improvements on
example, Kleim and Jones [11] suggest that rehabilita- functional measures of reaching for the unlimited
tion be task-specific, salient, repetitive, and intensive. rehabilitation group at 3 and 5 weeks postinfarct in
Of the 10 principles, “Intensity Matters,” “Repetition comparison with the limited rehabilitation group.
Matters,” and “Use It or Lose It” have significant clinical Furthermore, brain-derived neurotrophic factor levels
implications for the scheduling and distribution of re- were found to be significantly higher for the unlimited
habilitations services. Unfortunately, a review of each rehabilitation group. MacLellan et al [39] concluded
of these principles is beyond the scope of this article. that a sufficient dose and intensity of training is
Consequently, the following section will discuss the required to meet the threshold for stimulation and that
evidence for “Intensity Matters” and consider its impli- below this threshold, recovery does not occur.
cations for aphasia rehabilitation. A growing body of clinical research has emerged to
investigate the principles of experience-dependent
Intensity Matters neuroplasticity in stroke rehabilitation; specifically,
Evidence supporting the role of intensity in the in- several studies have investigated the effect of treat-
duction of neuroplasticity has been derived from basic ment intensity. A series of meta-analyses, systematic
neuroscience and clinical studies of stroke rehabilita- reviews, and experimental and observational studies
tion. Animal studies have revealed that intensive and have provided support for the benefits of intensive
repetitive behavioral training results in significant stroke rehabilitation [40-48], with improvements noted
neurologic and functional changes. Although the dose of in activities of daily living [42,43,48], mobility [42,48],
therapy required to initiate these central changes re- functional outcomes [40,41,43,47,48], and reduced
mains unknown, Kleim and Jones [11] noted that in length of stay in the hospital [47]. In contrast to the
many animal studies, intervention is provided in an aphasia literature, these studies have measured the
intensive format and that this intensity appears neces- benefits of intensive training, with intensity defined as
sary to achieve rehabilitation gains. For example, in a the amount of therapy provided or the number of
study conducted by Wang et al [8], adult owl monkeys therapeutic repetitions. Therefore, many of these
were trained on a tactile stimulation task that involved studies do not differentiate between therapy intensity
4-6 weeks of behavioral training with several hundred and amount and consequently provide support for a
trials per day. Wang et al [8] found that intensive, dose-dependent rehabilitation effect (i.e., the total
repetitive training resulted in significant changes in amount of therapy provided corresponds with therapy
the topographical representation of stimulated regions gains). Furthermore, for many disciplines, specific in-
in the somatosensory cortex. This finding of neural formation about optimal treatment intensity and the
 J.K. Dignam et al. / PM R 8 (2016) 254-267 257

effect of treatment intensity at different stages of re- impairment-based therapy in the acute or chronic
covery, for example, during the acquisition versus stages of aphasia recovery. In addition to these studies,
maintenance of a skill, is yet to be determined. a Cochrane review was conducted by Brady et al [1],
Consequently, treatment intensity is a pertinent field of who also found mixed evidence for the benefits of in-
research across rehabilitation disciplines, and further tensity. Although overall aphasia severity scores and
investigation into the parameters of optimal treatment written language scores improved more with intensive
intensity in the multidisciplinary rehabilitation of stroke therapy, no significant difference was found in measures
is warranted. of receptive or expressive (spoken) language. Brady
et al [1] also revealed that significantly more persons
Intensity and Aphasia Therapy withdrew from intensive therapy programs than from
nonintensive therapy programs. These studies highlight
Several studies have attempted to address the issue the need for further objective evaluation into the
of intensity of practice in aphasia therapy. A meta- effects of the intensity of aphasia therapy.
analysis conducted by Robey [6] found that aphasia
therapy delivered at an intensity of more than 2 hours Constraint-Induced Aphasia Therapy
per week resulted in greater communication changes
than less-intensive aphasia therapy. Likewise, a review The development of new, intensive treatment
conducted by Bhogal et al [49] found that when an approaches in aphasia rehabilitation, such as constraint-
average of 8.8 hours of speech and language therapy induced aphasia therapy (CIAT), have further contrib-
was provided per week for 11.2 weeks, significant uted to the evidence base for therapy intensity. A
communication gains were achieved, whereas when an greater understanding of neuroplasticity and the
average of 2 hours per week of speech and language neurologic processes underlying stroke recovery has
therapy were provided for 22.9 weeks, positive thera- stimulated the development of a group of techniques
peutic effects did not occur. These reports support the termed “constraint-induced therapy.” The original
benefits of intensive aphasia therapy; however, they constraint-induced protocol, constraint-induced move-
may also be confounded by a dosage effect. Basso [50] ment therapy, was developed for the rehabilitation of
found that a greater amount of therapy is more likely upper limb paresis [56,57]. Constraint-induced move-
to facilitate recovery than a smaller amount of therapy. ment therapy is an intensive aphasia therapy program,
In both of the reviews conducted by Robey [6] and adapted from the CIMT protocol, that incorporates the
Bhogal et al [49], as well as a number of clinical studies neuroplasticity principles of forced use, intensive and
investigating treatment intensity [25,27,30], the total repetitive practice, and specificity of training. In a
amount of therapy has not been consistent between pioneer study of CIAT, Pulvermuller et al [58] found that
groups. For example, Bhogal et al [49] compared the massed practice of verbal communication in
intensive therapy studies (total therapy time ¼ 98.4 conjunction with the restraint of nonverbal methods of
hours) with nonintensive therapy studies (total therapy communication (ie, gesture) for 3 hours per day for 10
time ¼ 43.6 hours). Therefore, it is possible that the days resulted in significant communication gains on
increased amount of therapy, rather than the intensity standard clinical tests. The positive therapeutic benefits
of treatment, was responsible for the positive thera- of CIAT have been replicated in a number of studies
peutic outcomes. In addition to this design limitation, [53,59,60]. However, the role of individual elements in
several studies investigating treatment intensity in the CIAT protocol (ie, constraint, intensity, and repeti-
aphasia rehabilitation have failed to incorporate a low- tion), as well as optimal treatment parameters
intensity control group or to compare the effects of (ie, timing, dosage, and schedule), remains unknown.
different levels of intensity [51-53]. Consequently, it is A number of studies have compared CIAT with other
difficult to draw definitive conclusions regarding the intensive aphasia therapy programs and have found
true effect of treatment intensity. comparable clinical gains, suggesting that intensity may
Additional studies have provided more equivocal re- be a common, active treatment component [51,61,62].
sults for the benefits of intensive aphasia therapy. For example, Barthel et al [51] evaluated the active
Cherney et al [54] conducted a systematic, evidence- therapeutic components of CIAT by comparing CIAT with
based review and found that increased intensity of model-oriented aphasia therapy, delivered at the same
therapy was associated with positive changes in out- high intensity. The study found that model-oriented
comes at the impairment level; however, mixed results aphasia therapy resulted in significant communication
were found for changes in communication activity/ gains, comparable to the gains achieved with CIAT. The
participation and in the maintenance of communication authors concluded that intensity and repetition appear
gains. With new evidence available, Cherney et al [55] to contribute to improvements in language function;
conducted a more recent review, and in contrast to however, constraint may not necessarily be an essential
their previous findings, they failed to distinguish element of the treatment protocol. Consequently,
between the effects of intensive and nonintensive further research is required to identify the active
258 Intensity of Aphasia Therapy

treatment components and optimal treatment parame- direct consequence of the number of repetitions.
ters of CIAT [51,63,64]. Although this study contributes to our understanding of
the effect of repetition and treatment intensity, un-
Intensive Comprehensive Aphasia Programs fortunately it does not inform us of the optimal distri-
bution of therapy sessions over time.
Another novel aphasia rehabilitation approach that To the best of our knowledge, only 4 studies have
has recently emerged and incorporates the principle of controlled the type and amount of aphasia therapy
intensive training is the development of intensive provided while systematically varying the intensity of
comprehensive aphasia programs (ICAPs). ICAPs adopt a treatment with respect to therapy schedule (Table 1).
comprehensive approach to aphasia rehabilitation that Martins et al [71] conducted a randomized, controlled
includes a combination of individual therapy, group trial to compare the effect of intensive aphasia therapy
therapy, computer-based therapy, and patient/family (2 hours per day, 5 days per week, for 10 weeks) with
education [65]. This service delivery model, which is standard practice (2 hours per week for 50 weeks).
based on a biopsychosocial approach to illness and Participants were stratified for aphasia severity, and
disability [65], provides intensive aphasia therapy, treatment groups were comparable for age, years of
defined as a minimum of 3 hours of therapy per day, 5 education, time after onset, aphasia severity and
days per week, for at least 2 weeks [65]. The develop- baseline language measures. The treatment provided in
ment of ICAPs represents an exciting progression in the the study was based on the Multimodal Stimulation
provision of aphasia therapy services, and preliminary Approach [72], and although therapists utilized the
research indicates that ICAPs can have a positive effect same treatment materials, therapy tasks were individ-
on individual’s language impairment and functional ualized for participants. The study found a trend fa-
communication [66-68]. However, the concept of an voring intensive training, as indicated by higher aphasia
ICAP is still relatively new, and although these programs quotient scores for the intensive group at 10 weeks, 50
are promising, further clinical research is required to weeks, and 62 weeks; however, this result did not reach
evaluate their efficacy and to determine optimal significance. No significant difference was found be-
treatment parameters [69]. tween the 2 groups on the primary or secondary
outcome measures at the completion of the study. One
Dosage-Controlled Aphasia Studies of the key limitations of this research was the small
sample size. Although 102 participants fulfilled the in-
Investigation into the effect of treatment intensity, clusion criteria for the study, only 30 participants were
independent of therapy dosage, is necessary to identify included and only 18 participants remained at the pri-
the fundamental parameters of aphasia rehabilitation. mary end point (50 weeks). Consequently, because of
Pulvermuller and Berthier [7] reviewed and interpreted the limited sample size, few meaningful conclusions can
the neurosciences literature with respect to the be drawn from the results.
scheduling of aphasia rehabilitation and proposed the Ramsberger and Marie [28] considered the intensity
Massed Practice Principle. This principle suggests that question using a computer-based anomia therapy pro-
for a constant number of therapy hours, a large amount gram. Treatment was delivered intensively (5 sessions
of aphasia therapy in a short amount of time is more per week for a total of 15-20 sessions) and non-
effective than the same or lesser amount of therapy intensively (2 sessions per week for a total of 15-20
delivered over an extended duration. Consequently, sessions) in a repeated measures crossover design with a
Pulvermuller and Berthier [7] argued for the benefits of sample of 4 participants. Mixed results for the benefits
intensive therapy, independent of the total amount of of intensive treatment were found. For 3 participants,
therapy provided. no difference was observed in the outcomes for inten-
Laganaro et al [70] investigated the effect of dosage- sive versus nonintensive therapy. However, one partic-
controlled treatment intensity on word retrieval out- ipant demonstrated a strong preference for the
comes in adults with acute aphasia. However, this study intensive therapy regime, with greater gains in word
defined treatment intensity with respect to the number retrieval made during intensive treatment. These re-
of repetitions of treated items. The number of expo- sults suggest that internal factors and specific patient
sures to treated items was a function of the size of characteristics may influence an individual’s response to
treatment sets (a small set includes 48 items; a large set different treatment schedules.
includes 96 items). Both training conditions involved a Raymer et al [73] investigated the effect of treat-
daily therapy (30 to 60 minutes) for 1 week. Thus, the ment intensity on measures of auditory comprehension
therapist contact time was controlled. At the conclusion and word retrieval in 5 persons with aphasia. The
of therapy, the proportional therapeutic gains for both repeated measures crossover design study contained
conditions were comparable. Interestingly, the large two 12-session training phases: an intensive therapy
treatment set resulted in numerically higher word condition (3-4 sessions per week) and a nonintensive
learning gains, suggesting that improvement was not a therapy condition (1-2 sessions per week). The study
 J.K. Dignam et al. / PM R 8 (2016) 254-267 259

found moderate effect sizes for the 2 participants

primary or secondary outcome

IT and RT; advantage of IT not

Improvement in naming for both

Improvement in naming for both

maintained at 1 mo follow-up
Improvement in naming for 3/4
between IT or RT groups on
completing the auditory comprehension training. Both

participants demonstrated
measures; trend of higher

advantage for RT at 1 mo
participants regardless of
treatment schedule; 1/4
participants made significant gains during the first phase

strong preference for IT

No significant difference

of auditory comprehension therapy, regardless of the

IT and RT; significant

therapy schedule (intensive/nonintensive). For word
Outcomes of Study

retrieval training, during the acquisition phase, there

scores for IT

was an advantage of intensive therapy over non-

follow-up
intensive therapy. All 5 participants demonstrated large
effect sizes in response to intensive therapy, whereas 2
of 5 participants revealed large effect sizes and 2 of 5
participants revealed moderate effect sizes in response
50 and 62 wk TPO

to nonintensive therapy. Interestingly, in this study it

Stability between

1 mo follow-up

1 mo follow-up
maintenance

was found that at the 1-month follow-up assessment,

Maintenance

although therapy gains were maintained above baseline,

the advantage of intensive therapy compared with
Minimal

data

nonintensive therapy was not maintained. Based on

these results, the authors suggested that nonintensive
therapy may be equally as effective as intensive therapy
45-60 min, 2 d/wk
(15-20 sessions)

and recommended that future research consider the

1-2 sessions/wk
(12 sessions,

(10 sessions)
2 h/wk, 50 wk

2 d/wk, 5 wk

relationship between training schedules and the main-

22-24 h)

tenance of therapy gains.

(100 h)

Finally, Sage et al [74] utilized a repeated-measures

crossover design to compare the outcomes of intensive
RT

anomia therapy (5 sessions per week for 2 weeks) with

45-60 min, 5 d/wk

the outcomes of nonintensive anomia therapy (2 ses-

(15-20 sessions)
10 wk (100 h)

3-4 sessions/wk
2 h/d, 5 d/wk,

sions per week for 5 weeks) in 8 participants with

(12 sessions,

(10 sessions)
5 d/wk, 2 wk

chronic aphasia. The investigators study found that

22-24 h)

immediately after therapy, both intensive and non-

Summary of dosage-controlled studies investigating treatment intensity in aphasia rehabilitation

intensive therapy conditions resulted in a significant

IT

improvement in naming ability. At the 1-month follow-

comprehension

up assessment, however, there was a small but statis-

N ¼ sample size; TPO ¼ time post onset; IT ¼ intensive therapy; RT ¼ regular therapy.
Anomia therapy

Anomia therapy
and anomia

tically significant advantage for items learned

stimulation
approach
Aphasia Quotient, Multimodal

nonintensively. Thus the authors found that whereas

therapy
Auditory
Therapy

both treatment schedules resulted in positive language

outcomes during the acquisition phase, the nonintensive
therapy schedule resulted in the superior maintenance
comprehension
Lisbon Aphasia

of therapy gains at 1-month follow-up.

and naming
Assessment

The findings of these dosage-controlled aphasia

Impairment:

Impairment:

Impairment:
accuracy

accuracy

accuracy
auditory

therapy studies suggest that when considering the

Battery

naming

naming
Measures
Outcome

maintenance of therapy gains, nonintensive therapy

may result in equivalent or even superior clinical out-
comes compared with intensive therapy. These findings
58.25 mo

are somewhat in contrast to the relevant neurosciences

Mean ¼
>6 mo

>4 mo

literature, which suggests an advantage of intensive

18 3 mo
TPO

training. However, the neurosciences literature has

frequently utilized animal models of stroke rehabilita-
N

tion to investigate the recovery of sensory and motor

functions after stroke. Although these models have
crossover design

crossover design

crossover design
Ramsberger Single participant,

Single participant,

Single participant,
controlled trial

played a pivotal role in beginning to understand the

Randomized

neural changes involved in stroke recovery, it is impor-

tant to acknowledge that language is a uniquely human
Design

cognitive faculty, thus suggesting inherent limits on how

easily evidence from animals can be applied to human
language rehabilitation. Furthermore, it is possible that
and Marie
et al [71]

et al [73]

Sage et al

important distinctions exist between the processes un-

Table 1

Raymer
Martins

[28]

[74]
Study

derlying the recovery of motor skills in humans after

stroke and the relearning of high-level cognitive
260 Intensity of Aphasia Therapy

functions, such as language. As such, the following learning underlies the therapeutic process and is an
section will review the cognitive psychology literature essential element in determining how therapy affects
with respect to theories of learning in healthy humans behavioral change [80]. Although currently no complete
and will consider how theories of verbal learning may model of human learning exists, research has investi-
translate to language recovery in aphasia. gated individual components within this model, such as
the organization of learning schedules and selection of
Cognitive Psychology and Learning training stimuli, to identify factors that promote
optimal learning.
Advancements in basic neuroscience research have Roediger and Karpicke [81] identify retrieval practice
contributed to our understanding of the biological and as one of the most important determinants of successful
physiological processes of learning and neuroplasticity. learning. Retrieval practice, or the testing effect, refers
In addition, research in the field of psychology has hel- to the robust and long-lasting advantage of testing over
ped to identify the nature of learning and memory additional study time. A pioneer study conducted by
mechanisms in healthy adults and to identify the con- Tulving [82] demonstrated that the process of actively
ditions required to promote successful learning. recalling information from long-term memory results in
Learning can be defined as “. any system or process, the superior learning and retention of information when
whether explicit or implicit, that results in the modifi- compared with an equivalent or even an increased
cation of behavior by experience” [75]. Learning in amount of study time. The testing effect has since been
healthy adults is a complex cognitive process that may replicated in a number of studies in cognitive psychol-
be influenced by a number of personal factors (eg, ogy (for a review, see Roediger and Karpicke [81]).
motivation and level of education) and cognitive factors Retrieval practice has been found to have differential
(eg, attention and executive function), as well as vari- effects across the continuum of the learning process,
ables pertaining to the learning experience itself (eg, and although repeated testing tends to slow initial
provision of feedback, errorless versus errorful produc- learning, it has been found to result in greater long-term
tion, and learning schedule). This process may be retention.
further complicated in persons with acquired neurologic The testing effect has been demonstrated in healthy
impairments due to the underlying pathophysiology, the adults, regardless of accuracy of recall [83]. However,
presence of concomitant cognitive deficits, and the the production of errors may have differential effects
potential interaction of these factors. Although it is on learning outcomes for adults with neurologic im-
acknowledged that important distinctions exist between pairments. A body of literature has emerged to inves-
learning in healthy persons and relearning in persons tigate the effect of errorless versus errorful training in
with acquired neurologic impairments, a greater un- persons with language and cognitive impairments
derstanding of the learning process in healthy persons [83,84]. Errorless learning paradigms aim to prevent the
may help to inform models of rehabilitation in the production of errors in training and thus reduce the
stroke population [76,77]. reinforcement of negative learning patterns [84]. Pre-
liminary research indicates that errorless and errorful
Theory of Learning therapy techniques result in comparable therapeutic
outcomes for adults with aphasia [85-88]. However, this
Detailed assessment of a person’s cognitive and is an emerging field of research, and further consider-
physical functioning after he or she sustains a stroke ation into the effect of error production on therapeutic
enables clinicians to carefully identify what is to be outcomes in persons with cognitive and language im-
learned in the therapeutic process. For example, in pairments is required.
aphasia rehabilitation, the cognitive neuropsychological The provision of feedback is another variable thought
framework for language processing enables speech- to influence the success of learning outcomes. Research
language pathologists to determine which aspects of a conducted in healthy adults has demonstrated success-
person’s language processing are impaired and conse- ful learning for verbal information, irrespective of the
quently require treatment. In contrast, a theory of presence or absence of feedback [89,90]. However,
learning informs us as to how this skill or behavior may additional studies suggest that both the nature and the
best be learned [78]. Baddeley [75] suggests that any timing of feedback may be integral [91,92]. For
treatment that aims to remediate cognitive functions, example, in a study of second language acquisition in
such as language, must incorporate an understanding of healthy persons, Pashler et al [91] found that specific
the impaired, underlying system, as well as the princi- feedback in response to an incorrect answer facilitated
ples of learning. This suggestion is further supported by learning and retention, whereas general feedback (ie,
the work of Robertson and Murre [79], who suggest that correct/incorrect) or feedback in response to a correct
broader cognitive factors, such as sustained attention, answer did not affect the learning outcomes. Interest-
are important considerations in the development and ingly, some reviews have suggested that frequent
implementation of rehabilitation programs. A theory of feedback may actually inhibit deeper levels of cognitive
 J.K. Dignam et al. / PM R 8 (2016) 254-267 261

processing and, consequently, may facilitate perfor- included in the meta-analysis with a retention interval
mance during acquisition yet reduce long-term reten- of greater than 1 month demonstrated superior learning
tion [93,94]. Few studies have specifically investigated when study sessions were spaced weeks to months apart
the role of feedback in learning outcomes for people in comparison with study sessions spaced 1 day apart.
with aphasia. In a study of nonverbal learning conducted Cepeda et al [107] conducted a series of subsequent
by Vallila-Rohter and Kiran [89], it was found that adults experiments to further investigate this effect and to
with aphasia differentially responded to the presence of determine the optimal gap between training sessions
feedback in training. Whereas some participants in this required to maximize learning. Experiment 2 of the
study demonstrated superior learning for items trained study required healthy participants to learn the names
in the feedback condition, other participants demon- of unfamiliar, visually presented items over 2 learning
strated a preference for paired-associate learning. sessions, scheduled either 0 days, 1 day, 1 week, 1
Alternatively, Fillingham et al [86] found no significant month, 3 months, or 6 months apart. Participants were
effect of the presence or absence of feedback on assessed 6 months after the second training session to
anomia therapy outcomes in adults with aphasia. evaluate learning and retention of the information.
Studies have also indicated that retrieval difficulty Analysis of the data revealed that optimal learning
may influence learning and the long-term retention of occurred when sessions were spaced 28 days apart.
information. Bjork [95] described the term “desirable From their research findings, Cepeda et al [107]
difficulties,” where optimal learning occurs when concluded that the optimal interval between learning
retrieval is maximally difficult yet successful. Methods sessions is a function of the retention interval and
of increasing the level of difficulty of retrieval may consequently, to achieve long-term retention of the
include manipulating the learning schedule, for learned information, significant temporal gaps between
example, by increasing the interval between learning learning sessions are required.
sessions, or by providing delayed versus immediate A greater understanding of the mechanisms underly-
feedback. The influence of retrieval difficulty on ing verbal learning and the environment and conditions
learning emerged from the finding that techniques that that promote it in healthy persons may have significant
enhance the short-term recall of information (ie, implications for the relearning of verbal information in
massed schedules) often result in inferior long-term aphasia rehabilitation [76,90]. Furthermore, techniques
retention, whereas techniques that reduce initial demonstrated to promote verbal learning in healthy
learning during the acquisition stage (ie, distributed persons may help inform therapeutic methods for adults
schedules and testing) often enhance the long-term with aphasia [76]. The use of nonwords in experimental
retention of information. Consequently, conditions paradigms provides an excellent opportunity to evaluate
that facilitate acquisition of a skill or behavior may not verbal learning processes. Callan and Schweighofer
necessarily facilitate the maintenance of that skill or [104] considered the learning of nonwords as a function
behavior. of training schedule (ie, massed versus distributed) in a
paired-associates learning task in healthy adults. The
The Distributed Practice Effect stimuli used in the study consisted of novel words (ie,
nonwords) paired with familiar lexical aspects of known
The distributed practice effect, also known as the words (ie, known meanings). Callan and Schweighofer
spacing effect, is a phenomenon that has received [104] found that performance on a delayed cued-recall
much attention in cognitive psychology. The distrib- test was superior for the distributed learning schedule
uted practice effect suggests that for a given amount in comparison with the massed schedule, replicating the
of practice, distributed presentations yield signifi- distributed practice effect. The advantage of distrib-
cantly better learning than massed presentations uted learning for nonwords has been demonstrated in a
[16,96]. This effect has been demonstrated for a number of studies in the adult population [101,106] and
range of tasks including simple and complex motor has even been demonstrated for novel vocabulary
tasks [97-99], memory recognition tasks for words, learning in school-aged children [105].
sentences, and pictures [100-102], training of cognitive Some evidence exists that distributed training is more
functions, such as working memory [103], and novel effective in clinical populations with impaired learning,
word learning [104-106]. such as in adults with dementia or in children with
Several studies have specifically investigated the learning disabilities [108,109]. The distributed practice
distributed practice effect in the learning and recall of effect has been investigated in clinical rehabilitation
verbal information. In a meta-analysis of the verbal studies of adults with traumatic brain injury, amnesia,
learning literature conducted by Cepeda et al [12] it was and multiple sclerosis [110-113]. These studies each
found that of 271 comparisons between massed versus found superior recall for information that was learned in
distributed learning schedules, only 12 comparisons a distributed schedule when compared with massed
showed no effect or reduced performance for a training. However, many of these studies evaluated
distributed practice schedule. In addition, every study the effect of the therapy schedule (ie, massed versus
262 Intensity of Aphasia Therapy

distributed) using relatively short intervals between interval between sessions is shorter (ie, reduced amount
training, and as such their clinical applicability may be of time for priming to diminish between sessions). The
limited. residual priming of stimuli learned in the intensive
A number of theories have been proposed to account condition reduces the difference between the learning
for the distributed practice effect. The 2 main ap- goal and the current state and, according to the delta
proaches discussed in the literature include encoding rule, therefore limits the rate of learning for those
variability theories and deficient processing theories stimuli. For items learned nonintensively, the amount of
(for a review see Dempster [16] or Delaney et al [114]). residual priming is diminished because of the longer
Encoding variability theories suggest that over a time between learning sessions. Consequently, the dif-
distributed period of time the number of subjective ference between the learning goal and the current state
contexts in which information is encoded increases and is large and learning is facilitated. To date, no one
consequently the number of potentially effective theory has been able to comprehensively account for
retrieval routes also increases [16]. Alternatively, in a the advantage of distributed training across the diverse
massed schedule the number of encoding paths is range of tasks and conditions that are reported in the
reduced, thus limiting cues to facilitate recall. The literature. Sage et al [74] provide one account for the
deficient processing theories suggest that massed pre- benefit of distributed training on word retrieval out-
sentations receive reduced cognitive processing in comes in aphasia rehabilitation. Thus, theoretically
comparison with distributed presentations and that motivated experimental studies that investigate the
recall is a function of the amount of processing the item distributed practice effect in the clinical rehabilitation
received [16]. Consequently, distributed presentations of adults with acquired neurologic disorders such as
are learned and recalled more effectively. Several aphasia are warranted.
mechanisms have been proposed to account for the
reduced cognitive processing of massed presentations Summary and Future Directions
[97,100,115,116]. For example, the voluntary attention
hypothesis suggests that participants pay less attention Determining the optimal treatment intensity for
to massed stimuli in comparison with spaced stimuli neurorehabilitation poses an important research ques-
because of the nature of the task (ie, boredom and fa- tion, with clinical implications for consumers, clinicians,
tigue effects) [16]. The rehearsal hypothesis suggests policy makers, and service providers. The neuroscience
that spaced presentations allow for greater mental literature suggests that intensive treatment promotes
rehearsal of stimuli between presentations and greater neurologic changes underpinning recovery from stroke
consolidation of learning. Consequently, stimuli are and is an essential component of rehabilitation. How-
encoded more deeply and retrieved more effectively ever, much of this research has been based on animal
[16]. and motor models of stroke rehabilitation. Conse-
Although a complete discussion of these theories is quently, it is unclear how these principles translate to
beyond the scope of this review, one pertinent account aphasia rehabilitation. Within cognitive psychology,
by Sage et al [74] will be discussed. Sage et al [74] used models of learning in healthy humans suggest that
learning algorithms, such as the Rescorla-Wagner theory distributed training promotes optimum long-term
[117] and the delta rule [117], to account for the learning and retention of trained skills and behaviors.
advantage of distributed practice over massed practice Although there are important distinctions between
in their study of confrontation naming in adults with learning in healthy persons and relearning in persons
aphasia. Put simply, these algorithms suggest that the with neurologic impairments, knowledge of the cogni-
rate of learning is dependent on the difference between tive processes underlying healthy learning may help
the learning goal (ie, target performance) and the cur- inform models of rehabilitation. Therefore, both the
rent state (suboptimal performance). When this differ- neuroscience and cognitive psychology literature have
ence is large, learning is enhanced. Performance on important implications for treatment intensity and ser-
verbal learning tasks, such as confrontation naming, vice delivery models in stroke and aphasia rehabilita-
may be influenced by repetition priming. Repetition tion. Given the findings of previous dosage-controlled
priming is a phenomenon in which the identification or studies of treatment intensity in aphasia rehabilitation
naming of an object or word is improved by previous and the evidence for the distributed practice effect in
exposure to that object or word [118]. For example, the cognitive psychology literature, it is important to
repetition priming may result in faster or more accurate avoid assumptions that high-intensity aphasia treatment
confrontation naming for stimuli previously seen is superior for all patients. Further empirical investiga-
compared with new stimuli. Priming effects have been tion is required to establish the true effect of treatment
shown to be relatively long-lasting and diminish over intensity and to determine how these 2 bodies of
time [118]. With respect to the study of Sage et al [74], research translate to language recovery in aphasia. The
the authors suggest that the amount of residual priming next section outlines a research agenda to further
for stimuli learned intensively is higher because the clarify this important clinical question.
 J.K. Dignam et al. / PM R 8 (2016) 254-267 263

Study Design Considerations therapy provided [25,29,58,120]. Dosage-controlled

studies considering the effect of treatment intensity
Much research has been conducted to investigate the across the impairment, activity, and participation levels
role of treatment intensity in the recovery of language of the World Health Organization’s International Clas-
function in aphasia rehabilitation. Unfortunately, sification of Functioning, Disability and Health (WHO-
methodologic limitations, such as reduced control for ICF) [121] are lacking. In recent years there has been a
treatment type [58] and dosage [25,27,30], have pre- shift in models of rehabilitation to consider the effects
vented us from identifying the true effect of treatment of aphasia therapy on both impairment and functional
intensity. Furthermore, within the aphasiology litera- outcome measures, as well as on quality of life
ture, there is limited consensus as to the definition of [122,123]. The need to objectively demonstrate rele-
intensive versus nonintensive therapy. A systematic vant and functional therapeutic gains is driven by
approach to defining parameters of treatment intensity increasing demands for accountability from health care
with respect to dose, dose frequency, and total inter- providers and policy makers; however, it is also central
vention duration is required. Tighter control of design to the holistic rehabilitation of people with aphasia.
variables, as well as application of the treatment in- Consequently, consideration of the effect of treatment
tensity model proposed by Warren et al [31], will allow intensity using a range of outcome measures across
for greater comparisons across studies and will help WHO-ICF domains is required.
delineate the role of treatment intensity in future
research. Consideration of the Maintenance of Treatment
A small number of studies have systematically Effects
investigated the effect of treatment intensity while
controlling the amount of therapy in adults with aphasia The ultimate aim of aphasia rehabilitation is to
[28,71,73,74]. However, to date the majority of these facilitate positive and enduring changes to individuals’
studies have provided a limited amount of therapy to a communication. Interestingly, however, few clinical
relatively small sample of persons. Although Martins studies have specifically considered which service de-
et al [71] aimed to deliver 100 hours of aphasia therapy, livery factors or treatment techniques optimize the
because of the challenges with recruitment and a high maintenance of therapeutic gains in aphasia rehabili-
attrition rate, this study was inadequately powered to tation. The neuroscience and cognitive psychology
differentiate between the 2 treatment conditions at the literature, as previously discussed, has contributed to
primary completion point. Furthermore, Martins et al our understanding of the factors that enhance the
[71] investigated the effect of treatment intensity in the acquisition and maintenance of newly learned skills and
subacute phase of stroke, and as such, the effects of behaviors. However, further clinical research is required
spontaneous recovery cannot be excluded. The to determine how these factors may be applied to
remaining dosage-controlled studies have delivered optimize the learning and retention of language skills in
relatively small amounts of therapy ranging from aphasia. A review of the stroke rehabilitation literature
approximately 10 hours [74] to 22-24 hours [73], with conducted by Teasell et al [124] found that although
sample sizes ranging from 4 participants [28] to 8 par- intensive rehabilitation resulted in significant gains in
ticipants [74]. Consequently, it is difficult to determine the short term, the long-term effects for many studies
whether an increased amount of aphasia therapy would were not maintained consistently over time. Within
influence the relationship between therapy schedule the aphasia literature, several studies investigating the
(ie, intensive versus nonintensive) and communication effect of therapy intensity on communication out-
outcomes. Evaluating the effect of treatment intensity comes have failed to collect maintenance data
when provided with a greater overall amount of therapy [25,27,58,119,120]. The few studies that have
is recommended. Furthermore, given the heteroge- controlled the amount of aphasia therapy while varying
neous nature of aphasia, a larger sample of participants the intensity have predominantly utilized a repeated-
is required. measures crossover design [28,73,74]. Robey and
Schultz [125] suggest this design is confounded in
Consideration of Outcome Measures aphasia research because of the influence of potential
carryover effects and/or a period effect. Consequently,
Previous dosage-controlled studies investigating the reliable data currently are limited regarding the main-
effects of treatment intensity in aphasia rehabilitation tenance of communication outcomes after intensive
have primarily evaluated impairment-based treatments, versus nonintensive aphasia therapy. Further studies
such as auditory comprehension and/or word retrieval investigating factors that may influence the mainte-
therapy [28,73,74,119]. Although some studies have nance of therapy gains, including consideration of the
considered activity/participation level outcomes, in relationship between therapy schedules and the long-
each of these studies, conclusions regarding intensity term retention of therapy outcomes, are required
have been confounded by the type or the amount of [28,29,58,73,74].
264 Intensity of Aphasia Therapy

Conclusions 7. Pulvermuller F, Berthier ML. Aphasia therapy on a neuroscience

basis. Aphasiology 2008;22:563-599.
8. Wang XQ, Merzenich MM, Sameshima K, Jenkins WM. Remodeling
Knowledge regarding the optimal treatment intensity of hand representation in adult cortex determined by timing of
in aphasia rehabilitation is necessary to provide clini- tactile stimulation. Nature 1995;378:71-75.
cally beneficial and cost-effective aphasia services. 9. Langhorne P, Bernhardt J, Kwakkel G. Stroke rehabilitation.
However, the optimal treatment intensity for aphasia Lancet 2011;377:1693-1702.
rehabilitation remains unknown. A comprehensive 10. Teasell R, Kalra L. What’s new in stroke rehabilitation: Back to
basics. Stroke 2005;36:215-217.
search of the literature pertaining to intensity and 11. Kleim JA, Jones TA. Principles of experience-dependent neural
aphasia rehabilitation was undertaken as part of this plasticity: Implications for rehabilitation after brain damage.
narrative review. The neurosciences literature suggests J Speech Lang Hear Res 2008;51:S225-S239.
that intensive training is a key and necessary component 12. Cepeda NJ, Pashler H, Vul E, Wixted JT, Rohrer D. Distributed
of rehabilitation (ie, “intensity matters”). However, the practice in verbal recall tasks: A review and quantitative syn-
thesis. Psychol Bull 2006;132:354-380.
cognitive psychology literature suggests that although 13. Jenkins WM, Merzenich MM, Recanzone G. Neocortical represen-
intensive training may facilitate the acquisition of a tational dynamics in adult primates: Implications for neuropsy-
skill, optimal long-term learning is achieved with chology. Neuropsychologia 1990;28:573-584.
distributed or nonintensive training schedules. This re- 14. Keefe KA. Applying basic neuroscience to aphasia therapy: What
view has discussed and evaluated both of these per- the animals are telling us. Am J Speech Lang Pathol 1995;4:88-93.
15. Teasell R, Foley N, Salter K, et al. The Stroke Rehabilitation Ev-
spectives with respect to the provision of aphasia idence Based Review. 16th ed. Available at: http://www.ebrsr.
therapy services. Although efforts were made to incor- com/. Accessed July 2, 2015.
porate all relevant studies, it is acknowledged that a 16. Dempster FN. Spacing effects and their implications for theory
systematic review would have provided stronger and practice. Educ Psychol Rev 1989;1:309-330.
methodology. 17. Lindsay MP, Gubitz G, Bayley M, Phillips S. Canadian Best Practice
Recommendations for Stroke Care. Ontario, Canada: Stroke
In addition, this review has evaluated the current Rehabilitation Best Practices Writing Group; 2013.
state of evidence for treatment intensity in aphasia 18. National Stroke Foundation. Clinical Guidelines for Stroke Man-
intervention and has identified directions for future agement. Melbourne, Australia: National Stroke Foundation;
research. We have highlighted the need for well- 2010.
designed research projects to systematically evaluate 19. Intercollegiate Stroke Working Party. National Clinical Guideline
for Stroke. 4th ed. London, United Kingdom: Royal College of
parameters of treatment intensity such as dose, dose Physicians; 2012.
frequency, and total intervention duration in aphasia 20. Duncan PW, Zorowitz R, Bates B, et al. Management of adult
rehabilitation. Furthermore, future studies are required stroke rehabilitation care: A clinical practice guideline. Stroke
to investigate the relationship between treatment in- 2005;36:E100-E143.
tensity and the acquisition and maintenance of therapy 21. Hacke W, Ringleb PA, Bousser MG, et al. Guidelines for manage-
ment of ischaemic stroke and transient ischaemic attack
gains. Finally, it is recommended that future studies 2008dThe European Stroke Organisation (ESO) Executive Com-
consider the effect of treatment intensity on commu- mittee and the ESO Writing Committee. Cerebrovasc Dis 2008;25:
nication outcomes across WHO-ICF domains. Resolving 457-507.
the importance of intensity is a critical issue in the field 22. Bowen A, Hesketh A, Patchick E, et al. Effectiveness of enhanced
that will have broad implications for the clinical out- communication therapy in the first four months after stroke for
aphasia and dysarthria: A randomised controlled trial. BMJ 2012;
comes of consumers and for service delivery models in 345:e4407.
stroke and aphasia rehabilitation. 23. Godecke E, Worrall L. Peer commentary on the paper “Effec-
tiveness of communication therapy in the first four months after
stroke for aphasia and dysarthria: A randomised controlled trial”
by A. Bowen et al. Retrieved from http://www.bmj.com.ezproxy.
References library.uq.edu.au/content/345/bmj.e4407?tab¼responses.
24. Code C. Peer commentary on the paper “Effectiveness of
1. Brady MC, Kelly H, Godwin J, Enderby P. Speech and language enhanced communication therapy in the first four months after
therapy for aphasia following stroke. Cochrane Database Syst Rev stroke for aphasia and dysarthria: A randomised controlled trial”
2012;5:CD000425. by A. Bowen et al. Retrieved from http://www.bmj.com.ezproxy.
2. Cicerone KD, Dahlberg C, Kalmar K, et al. Evidence-based library.uq.edu.au/content/345/bmj.e4407?tab¼responses.
cognitive rehabilitation: Recommendations for clinical practice. 25. Hinckley JJ, Carr TH. Comparing the outcomes of intensive and
Arch Phys Med Rehabil 2000;81:1596-1615. non-intensive context-based aphasia treatment. Aphasiology
3. Nickels L, Best W. Therapy for naming disorders (Part 1): Princi- 2005;19:965-974.
ples, puzzles and progress. Aphasiology 1996;10:21-47. 26. Bakheit AMO, Shaw S, Barrett L, et al. A prospective, randomized,
4. Wisenburn B, Mahoney K. A meta-analysis of word-finding treat- parallel group, controlled study of the effect of intensity of
ments for aphasia. Aphasiology 2009;23:1338-1352. speech and language therapy on early recovery from poststroke
5. Whurr R, Lorch MP, Nye C. A meta-analysis of studies carried out aphasia. Clin Rehabil 2007;21:885-894.
between 1946 and 1988 concerned with the efficacy of speech 27. Denes G, Perazzolo C, Piani A, Piccione F. Intensive versus regular
and language therapy treatment for aphasic patients. Eur J Disord speech therapy in global aphasia: A controlled study. Aphasiology
Commun 1992;27:1-17. 1996;10:385-394.
6. Robey RR. A meta-analysis of clinical outcomes in the treatment 28. Ramsberger G, Marie B. Self-administered cued naming therapy:
of aphasia. J Speech Lang Hear Res 1998;41:172-187. A single-participant investigation of a computer-based therapy
 J.K. Dignam et al. / PM R 8 (2016) 254-267 265

program replicated in four cases. Am J Speech Lang Pathol 2007; 50. Basso A. How intensive/prolonged should an intensive/prolonged
16:343-358. treatment be? Aphasiology 2005;19:975-984.
29. Brindley P, Copeland M, Demain C, Martyn P. A comparison of the 51. Barthel G, Meinzer M, Djundja D, Rockstroh B. Intensive language
speech of ten chronic Broca’s aphasics following intensive and therapy in chronic aphasia: Which aspects contribute most?
non-intensive periods of therapy. Aphasiology 1989;3:695-707. Aphasiology 2008;22:408-421.
30. Hinckley JJ, Craig HK. Influence of rate of treatment on the 52. Code C, Torney A, Gildea-Howardine E, Willmes K. Outcome of a
naming abilities of adults with chronic aphasia. Aphasiology 1998; one-month therapy intensive for chronic aphasia: Variable indi-
12:989-1006. vidual responses. Semin Speech Lang 2010;31:21-33.
31. Warren SF, Fey ME, Yoder PJ. Differential treatment intensity 53. Maher LM, Kendall D, Swearengin JA, et al. A pilot study of use-
research: A missing link to creating optimally effective communi- dependent learning in the context of constraint induced lan-
cation interventions. Ment Retard Dev Disabil Res Rev 2007;13:70-77. guage therapy. J Int Neuropsychol Soc 2006;12:843-852.
32. Cherney LR. Aphasia treatment: Intensity, dose parameters, and 54. Cherney LR, Patterson JP, Raymer A, Frymark T, Schooling T.
script training. Int J Speech Lang Pathol 2012;14:424-431. Evidence-based systematic review: Effects of intensity of treat-
33. Harnish S, Morgan J, Lundine JP, Bauer A, Singletary F. Dosing of a ment and constraint-induced language therapy for individuals
cued picture-naming treatment for anomia. Am J Speech Lang with stroke-induced aphasia. J Speech Lang Hear Res 2008;51:
Pathol 2014;23:S285-S299. 1282-1299.
34. Helm-Estabrooks N. Cognition and aphasia: A discussion and a 55. Cherney LR, Patterson JP, Raymer AM. Intensity of aphasia ther-
study. J Commun Disord 2002;35:171-186. apy: Evidence and efficacy. Curr Neurol Neurosci Rep 2011;11:
35. Raymer AM, Beeson P, Holland A, et al. Translational research in 560-569.
aphasia: From neuroscience to neurorehabilitation. J Speech 56. Taub E, Miller NE, Novack TA, et al. Technique to improve chronic
Lang Hear Res 2008;51:S259-S275. motor deficit after stroke. Arch Phys Med Rehabil 1993;74:
36. Kleim JA, Barbay S, Cooper NR, et al. Motor learning-dependent 347-354.
synaptogenesis is localized to functionally reorganized motor 57. Taub E, Uswatte G, Pidikiti R. Constraint-induced movement
cortex. Neurobiol Learn Mem 2002;77:63-77. therapy: A new family of techniques with broad application to
37. Kleim JA, Hogg TM, VandenBerg PM, Cooper NR, Bruneau R, physical rehabilitationdA clinical review. J Rehabil Res Dev 1999;
Remple M. Cortical synaptogenesis and motor map reorganization 36:237-251.
occur during late, but not early, phase of motor skill learning. J 58. Pulvermuller F, Neininger B, Elbert T, et al. Constraint-induced
Neurosci 2004;24:628-633. therapy of chronic aphasia after stroke. Stroke 2001;32:
38. Luke LM, Allred RP, Jones TA. Unilateral ischemic sensorimotor 1621-1626.
cortical damage induces contralesional synaptogenesis and en- 59. Meinzer M, Djundja D, Barthel G, Elbert T, Rockstroh B. Long-
hances skilled reaching with the ipsilateral forelimb in adult male term stability of improved language functions in chronic aphasia
rats. Synapse 2004;54:187-199. after constraint-induced aphasia therapy. Stroke 2005;36:
39. MacLellan CL, Keough MB, Granter-Button S, Chernenko GA, 1462-1466.
Butt S, Corbett D. A critical threshold of rehabilitation involving 60. Meinzer M, Elbert T, Wienbruch C, Djundja D, Barthel G,
brain-derived neurotrophic factor is required for poststroke re- Rockstroh B. Intensive language training enhances brain plasticity
covery. Neurorehabil Neural Repair 2011;25:740-748. in chronic aphasia. BMC Biol 2004;2:20-29.
40. Huang HC, Chung KC, Lai DC, Sung SF. The impact of timing and 61. Sickert A, Anders LC, Munte TF, Sailer M. Constraint-induced
dose of rehabilitation delivery on functional recovery of stroke aphasia therapy following sub-acute stroke: A single-blind,
patients. J Chin Med Assoc 2009;72:257-264. randomised clinical trial of a modified therapy schedule. J Neu-
41. Kwakkel G, Wagenaar RC, Twisk JWR, Lankhorst GJ, Koetsier JC. rol Neurosurg Psychiatry 2014;85:51-55.
Intensity of leg and arm training after primary middle-cerebral- 62. Kurland J, Baldwin K, Tauer C. Treatment-induced neuroplasticity
artery stroke: A randomised trial. Lancet 1999;354:191-196. following intensive naming therapy in a case of chronic Wer-
42. Kwakkel G, van Peppen R, Wagenaar RC, et al. Effects of nicke’s aphasia. Aphasiology 2010;24:737-751.
augmented exercise therapy time after stroke: A meta-analysis. 63. Meinzer M, Rodriguez AD, Rothi LJG. First decade of research on
Stroke 2004;35:2529-2536. constrained-induced treatment approaches for aphasia rehabili-
43. Kwakkel G, Wagenaar RC, Koelman TW, Lankhorst GJ, tation. Arch Phys Med Rehabil 2012;93:S35-S45.
Koetsier JC. Effects of intensity of rehabilitation after stroke: A 64. Attard MC, Rose ML, Lanyon L. The comparative effects of multi-
research synthesis. Stroke 1997;28:1550-1556. modality aphasia therapy and constraint-induced aphasia
44. Lohse KR, Lang CE, Boyd LA. Is more better? Using metadata to therapy-plus for severe chronic Broca’s aphasia: An in-depth pilot
explore dose-response relationships in stroke rehabilitation. study. Aphasiology 2013;27:80-111.
Stroke 2014;45:2053-2058. 65. Cherney LR, Worrall L, Rose M. What is an ICAP?: An international
45. Birkenmeier RL, Prager EM, Lang CE. Translating animal doses of survey of intensive, comprehensive aphasia programs. Interna-
task-specific training to people with chronic stroke in 1-hour tional Aphasia Rehabilitation Conference; 2012; Melbourne, Vic-
therapy sessions: A proof-of-concept study. Neurorehabil Neural toria, Australia.
Repair 2010;24:620-635. 66. Rodriguez A, Worrall L, Brown K, et al. Aphasia LIFT: Exploratory
46. Cooke EV, Mares K, Clark A, Tallis RC, Pomeroy VM. The effects of investigation of an intensive comprehensive aphasia programme.
increased dose of exercise-based therapies to enhance motor Aphasiology 2013;27:1339-1361.
recovery after stroke: A systematic review and meta-analysis. 67. Persad C, Wozniak L, Kostopoulos E. Retrospective analysis of
BMC Med 2010;8:13. outcomes from two intensive comprehensive aphasia programs.
47. Jette DU, Warren RL, Wirtalla C. The relation between therapy Top Stroke Rehabil 2013;20:388-397.
intensity and outcomes of rehabilitation in skilled nursing facil- 68. Winans-Mitrik RL, Hula WD, Dickey MW, Schumacher JG,
ities. Arch Phys Med Rehabil 2005;86:373-379. Swoyer B, Doyle PJ. Description of an intensive residential
48. Bode RK, Heinemann AW, Semik P, Mallinson T. Relative impor- aphasia treatment program: Rationale, clinical processes, and
tance of rehabilitation therapy characteristics on functional outcomes. Am J Speech Lang Pathol 2014;23:S330-S342.
outcomes for persons with stroke. Stroke 2004;35:2537-2542. 69. Hula WD, Cherney LR, Worrall LE. Setting a research agenda to
49. Bhogal SK, Teasell R, Speechley M. Intensity of aphasia therapy: inform intensive comprehensive aphasia programs. Top Stroke
Impact on recovery. Stroke 2003;34:987-992. Rehabil 2013;20:409-420.
266 Intensity of Aphasia Therapy

70. Laganaro M, Di Pietro M, Schnider A. Computerised treatment of 93. Schmidt RA, Bjork RA. New conceptualizations of practice:
anomia in acute aphasia: Treatment intensity and training size. Common principles in 3 paradigms suggest new concepts for
Neuropsychol Rehabil 2006;16:630-640. training. Psychol Sci 1992;3:207-217.
71. Martins IP, Leal G, Fonseca I, et al. A randomized, rater-blinded, 94. Rosenbaum DA, Carlson RA, Gilmore RO. Acquisition of intellec-
parallel trial of intensive speech therapy in sub-acute post-stroke tual and perceptual-motor skills. Annu Rev Psychol 2001;52:
aphasia: The SP-I-R-IT study. Int J Lang Commun Disord 2013;48: 453-470.
421-431. 95. Bjork RA. Memory and metamemory considerations in the training
72. Duffy JR, Coelho C. Schuell’s stimulation approach to rehabili- of human beings. In: Metcalfe J, Shimamura A, eds. Metacogni-
tation. In: Chapey R, ed. Language intervention strategies in tion: Knowing about knowing. Cambridge, MA: MIT Press; 1994;
aphasia and related neurogenic communication disorders. 4th ed. 185-205.
Baltimore, MD: Lippincott Williams & Wilkins; 2001; 341-382. 96. Donovan JJ, Radosevich DJ. A meta-analytic review of the dis-
73. Raymer AM, Kohen FP, Saffell D. Computerised training for im- tribution of practice effect: Now you see it, now you don’t. J Appl
pairments of word comprehension and retrieval in aphasia. Psychol 1999;84:795-805.
Aphasiology 2006;20:257-268. 97. Dail TK, Christina RW. Distribution of practice and metacognition
74. Sage K, Snell C, Lambon Ralph MA. How intensive does anomia in learning and long-term retention of a discrete motor task. Res
therapy for people with aphasia need to be? Neuropsychol Rehabil Q Exerc Sport 2004;75:148-155.
2011;21:26-41. 98. Mackay S, Morgan P, Datta V, Chang A, Darzi A. Practice distri-
75. Baddeley A. A theory of rehabilitation without a model of learning bution in procedural skills training: A randomized controlled trial.
is a vehicle without an engine: A comment. Neuropsychol Rehabil Surg Endosc 2002;16:957-961.
1993;3:235-244. 99. Moulton CAE, Dubrowski A, MacRae H, Graham B, Grober E,
76. Basso A, Marangolo P, Piras F, Galluzzi C. Acquisition of new Reznick R. Teaching surgical skills: What kind of practice makes
“words” in normal subjects: A suggestion for the treatment of perfect? A randomized, controlled trial. Ann Surg 2006;244:
anomia. Brain Lang 2001;77:45-59. 400-409.
77. Breitenstein C, Knecht S. Development and validation of a lan- 100. Challis BH. Spacing effects on cued-memory tests depend on
guage learning model for behavioral and functional-imaging level of processing. J Exp Psychol Learn Mem Cogn 1993;19:
studies. J Neurosci Methods 2002;114:173-179. 389-396.
78. Howard D. Learning theory is not enough. Aphasiology 1999;13: 101. Russo R, Mammarella N. Spacing effects in recognition memory:
140-143. When meaning matters. Eur J Cogn Psychol 2002;14:49-59.
79. Robertson IH, Murre JMJ. Rehabilitation of brain damage: Brain 102. Xue G, Mei LL, Chen CS, Lu ZL, Poldrack R, Dong Q. Spaced
plasticity and principles of guided recovery. Psychol Bull 1999; learning enhances subsequent recognition memory by reducing
125:544-575. neural repetition suppression. J Cogn Neurosci 2011;23:
80. Ferguson A. Learning in aphasia therapy: It’s not so much what 1624-1633.
you do, but how you do it! Aphasiology 1999;13:125-132. 103. Penner I-K, Vogt A, Stöcklin M, Gschwind L, Opwis K, Calabrese P.
81. Roediger HL, Karpicke JD. The power of testing memory basic Computerised working memory training in healthy adults: A
research and implications for educational practice. Perspect comparison of two different training schedules. Neuropsychol
Psychol Sci 2006;1:181-210. Rehabil 2012;22:716-733.
82. Tulving E. The effects of presentation and recall of material 104. Callan DE, Schweighofer N. Neural correlates of the spacing ef-
in free-recall learning. J Verbal Learning Verbal Behav 1967;6: fect in explicit verbal semantic encoding support the deficient-
175-184. processing theory. Hum Brain Mapp 2010;31:645-659.
83. Middleton EL, Schwartz MF. Errorless learning in cognitive 105. Sobel HS, Cepeda NJ, Kapler IV. Spacing effects in real-world
rehabilitation: A critical review. Neuropsychol Rehabil 2012;22: classroom vocabulary learning. Appl Cogn Psychol 2011;25:
138-168. 763-767.
84. Fillingham JK, Hodgson C, Sage K, Ralph MAL. The application of 106. Mammarella N, Russo R, Avons SE. Spacing effects in cued-
errorless learning to aphasic disorders: A review of theory and memory tasks for unfamiliar faces and nonwords. Mem Cognit
practice. Neuropsychol Rehabil 2003;13:337-363. 2002;30:1238-1251.
85. Fillingham JK, Sage K, Ralph MAL. The treatment of anomia using 107. Cepeda NJ, Coburn N, Rohrer D, Wixted JT, Mozer MC, Pashler H.
errorless learning. Neuropsychol Rehabil 2006;16:129-154. Optimizing distributed practice theoretical analysis and practical
86. Fillingham JK, Sage K, Ralph MAL. Treatment of anomia using implications. Exp Psychol 2009;56:236-246.
errorless versus errorful learning: Are frontal executive skills and 108. Camp CJ, Foss JW, Ohanlon AM, Stevens AB. Memory in-
feedback important? Int J Lang Commun Disord 2005;40:505-523. terventions for persons with dementia. Appl Cogn Psychol 1996;
87. Conroy P, Sage K, Lambon Ralph MA. Errorless and errorful ther- 10:193-210.
apy for verb and noun naming in aphasia. Aphasiology 2009;23: 109. Riches NG, Tomasello M, Conti-Ramsden G. Verb learning in
1311-1337. children with SLI: Frequency and spacing effects. J Speech Lang
88. Abel S, Schultz A, Radermacher I, Willmes K, Huber W. Decreasing Hear Res 2005;48:1397-1411.
and increasing cues in naming therapy for aphasia. Aphasiology 110. Cermak LS, Verfaellie M, Lanzoni S, Mather M, Chase KA. Effect of
2005;19:831-848. spaced repetitions on amnesia patients’ recall and recognition
89. Vallila-Rohter S, Kiran S. Non-linguistic learning and aphasia: performance. Neuropsychol 1996;10:219-227.
Evidence from a paired associate and feedback-based task. 111. Hillary FG, Schultheis MT, Challis BH, et al. Spacing of repetitions
Neuropsychologia 2013;51:79-90. improves learning and memory after moderate and severe TBI.
90. Breitenstein C, Kamping S, Jansen A, Schomacher M, Knecht S. J Clin Exp Neuropsychol 2003;25:49-58.
Word learning can be achieved without feedback: Implications for 112. Goverover Y, Arango-Lasprilla JC, Hillary FG, Chiaravalloti N,
aphasia therapy. Restor Neurol Neurosci 2004;22:445-458. DeLuca J. Application of the spacing effect to improve learning
91. Pashler H, Cepeda NJ, Wixted JT, Rohrer D. When does feedback and memory for functional tasks in traumatic brain injury: A pilot
facilitate learning of words? J Exp Psychol Learn Mem Cogn 2005; study. Am J Occup Ther 2009;63:543-548.
31:3-8. 113. Goverover Y, Hillary FG, Chiaravalloti N, Arango-Lasprilla JC,
92. Pashler H, Rohrer D, Cepeda NJ, Carpenter SK. Enhancing learning DeLuca J. A functional application of the spacing effect to
and retarding forgetting: Choices and consequences. Psychon Bull improve learning and memory in persons with multiple sclerosis.
Rev 2007;14:187-193. J Clin Exp Neuropsychol 2009;31:513-522.
 J.K. Dignam et al. / PM R 8 (2016) 254-267 267

114. Delaney PF, Verkoeijen P, Spirgel A. Spacing and testing effects: A 120. Lee JB, Kaye RC, Cherney LR. Conversational script performance
deeply critical, lengthy, and at times discursive review of the in adults with non-fluent aphasia: Treatment intensity and
literature. In: Ross BH, ed. Psychology of Learning and Motiva- aphasia severity. Aphasiology 2009;23:885-897.
tion: Advances in Research and Theory, Vol 53. San Diego, CA: 121. World Health Organization. The International Classification of
Elsevier Academic Press; 2010; 63-147. Functioning, Disability and Health. Geneva, Switzerland: World
115. Wahlheim CN, Dunlosky J, Jacoby LL. Spacing enhances the Health Organization; 2001.
learning of natural concepts: An investigation of mechanisms, 122. Cruice M, Worrall L, Hickson L, Murison R. Measuring quality of
metacognition, and aging. Mem Cognit 2011;39:750-763. life: Comparing family members’ and friends’ ratings with those
116. Janiszewski C, Noel H, Sawyer AG. A meta-analysis of the spacing of their aphasic partners. Aphasiology 2005;19:111-129.
effect in verbal learning: Implications for research on advertising 123. Kagan A, Simmons-Mackie N, Rowland A, et al. Counting what
repetition and consumer memory. J Consum Res 2003;30:138-149. counts: A framework for capturing real-life outcomes of aphasia
117. Anderson JR. Learning and memory: An integrated approach. intervention. Aphasiology 2008;22:258-280.
New York, NY: Wiley; 1995. 124. Teasell R, Foley N, Salter K, Bhogal S, Jutai J, Speechley M.
118. Wiggs CL, Martin A. Properties and mechanisms of perceptual Evidence-based review of stroke rehabilitation: Executive
priming. Curr Opin Neurobiol 1998;8:227-233. summary, 12th edition. Top Stroke Rehabil 2009;16:463-488.
119. Harnish SM, Neils-Strunjas J, Lamy M, Eliassen J. Use of fMRI in 125. Robey RR, Schultz MC. A model for conducting clinical-outcome
the study of chronic aphasia recovery after therapy: A case study. research: An adaptation of the standard protocol for use in
Top Stroke Rehabil 2008;15:468-483. aphasiology. Aphasiology 1998;12:787-810.

Disclosure

J.K.D. UQ Centre for Clinical Research, Building 71/918, RBWH Campus, The D.A.C. UQ Centre for Clinical Research, The University of Queensland, Herston,
University of Queensland, Herston, QLD 4029, Australia. Address correspondence Australia; and National Health and Medical Research Council Centre of Clinical
to: J.K.D.; e-mail: j.dignam@uq.edu.au Research Excellence in Aphasia Rehabilitation, Herston, Australia
Disclosure: nothing to disclose Disclosures related to this publication: grant, Australian Research Council Future
Fellowship, NHMRC Centre of Clinical Research Excellence in Aphasia
A.D.R. National Health and Medical Research Council Centre of Clinical Research Rehabilitation
Excellence in Aphasia Rehabilitation, St Lucia, Australia; and School of Health
This work was supported by the National Health and Medical Research Council
and Rehabilitation Sciences, The University of Queensland, Herston, Australia
Centre of Clinical Research Excellence in Aphasia Rehabilitation (grant 569935).
Disclosure: nothing to disclose
Submitted for publication February 17, 2015; accepted June 16, 2015.