Clinical Linguistics & Phonetics

ISSN: 0269-9206 (Print) 1464-5076 (Online) Journal homepage: http://www.tandfonline.com/loi/iclp20

Improved word comprehension in Global aphasia
using a modified semantic feature analysis
treatment
Philippa Munro & Samantha Siyambalapitiya
To cite this article: Philippa Munro & Samantha Siyambalapitiya (2016): Improved word
comprehension in Global aphasia using a modified semantic feature analysis treatment,
Clinical Linguistics & Phonetics, DOI: 10.1080/02699206.2016.1198927
To link to this article: http://dx.doi.org/10.1080/02699206.2016.1198927

Published online: 22 Aug 2016.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=iclp20
Download by: [Northern Illinois University]

Date: 23 August 2016, At: 23:52

CLINICAL LINGUISTICS & PHONETICS
http://dx.doi.org/10.1080/02699206.2016.1198927

Improved word comprehension in Global aphasia using a
modified semantic feature analysis treatment
Philippa Munroa,b and Samantha Siyambalapitiyaa,c
a
Discipline of Speech Pathology, James Cook University, Townsville, Australia; bSpeech Pathology, Mackay
Base Hospital, Queensland Health, Mackay, Australia; cMenzies Health Institute Queensland, Griffith University,
Gold Coast, Australia

ABSTRACT

ARTICLE HISTORY

Limited research has investigated treatment of single word comprehension in people with aphasia, despite numerous studies examining
treatment of naming deficits. This study employed a single case
experimental design to examine efficacy of a modified semantic
feature analysis (SFA) therapy in improving word comprehension in
an individual with Global aphasia, who presented with a semantically
based comprehension impairment. Ten treatment sessions were conducted over a period of two weeks. Following therapy, the participant demonstrated improved comprehension of treatment items and
generalisation to control items, measured by performance on a spoken word picture matching task. Improvements were also observed
on other language assessments (e.g. subtests of WAB-R; PALPA subtest 47) and were largely maintained over a period of 12 weeks without further therapy. This study provides support for the efficacy of a
modified SFA therapy in remediating single word comprehension in
individuals with aphasia with a semantically based comprehension
deficit.

Received 4 November 2015
Revised 26 May 2016
Accepted 4 June 2016
KEYWORDS

Aphasia treatment; auditory
comprehension; Global
aphasia; semantic feature
analysis; treatment intensity

While numerous treatment studies have investigated remediation of naming difficulties
associated with aphasia, research regarding treatment of auditory comprehension is
comparatively limited (Morris, Franklin, Ellis, Turner, & Bailey, 1996; Francis, Riddoch,
& Humphreys, 2001; Tessier, Weill-Chounlamountry, Michelot, & Pradat-Diehl, 2007).
This is to the detriment of individuals diagnosed with aphasia, as auditory comprehension
is almost always affected to some degree following the onset of aphasia (Bastiaanse,
Nijboer, & Taconis, 1993; Morris et al., 1996). While it is acknowledged that in many
cases comprehension deficits may improve spontaneously over time, it is important to
recognise that full recovery of comprehension does not always occur, especially in more
severe cases such as individuals diagnosed with Global aphasia (Alexander & Loverso,
1992; Sarno, 1998).
Thus far, only a small number of studies have explored remediation of comprehension
deficits in individuals with aphasia (Behrmann & Lieberthal, 1989; Bastiaanse et al., 1993;
Francis et al., 2001; Morris et al., 1996; Tessier et al., 2007; Morris & Franklin, 2012).
These studies employed a cognitive neuropsychological approach which aims to utilise
language processing models to identify the underlying level of language deficit and target
CONTACT Samantha Siyambalapitiya
s.siyambalapitiya@griffith.edu.au
Speech Pathology, School of Allied
Health Sciences, Gold Coast Campus, Griffith University, Gold Coast, QLD 4222, Australia.
© 2016 Taylor & Francis

2

P. MUNRO AND S. SIYAMBALAPITIYA

it directly through therapy (Behrmann & Lieberthal, 1989; Hall & Riddoch, 1997; Albert,
1998; Best & Nickels, 2000). This approach allows explanations to be posited regarding
how different therapeutic approaches can influence a specific level of language impairment
(Best & Nickels, 2000).
According to cognitive neuropsychological theory, auditory comprehension disorders
can be classified into three main types based on the level of impairment (Bastiaanse et al.,
1993; Hall & Riddoch, 1997; Francis et al., 2001). These levels can be defined as an
individual’s ability to: a) identify and process phonemes (auditory processing); b) recognise strings of sounds as real and familiar words (lexical-phonological processing); or c) to
comprehend or access the meaning of words (semantic processing) (Bastiaanse et al., 1993;
Whitworth, Webster, & Howard, 2005; Tessier et al., 2007). Impairments that impede
semantic processing are typically treated using a semantically based treatment that focuses
on meaning, whereas impairments that affect the preceding levels may be targeted using a
phonologically based treatment that focuses on processing of sounds or phonemes.
Several studies have focused on remediation of semantically based comprehension
impairments; those resulting from a breakdown in the access to, or processing within, the
semantic system (Behrmann & Lieberthal, 1989; Bastiaanse et al.,1993; Francis et al., 2001;
Morris & Franklin, 2012). Francis et al. (2001) employed explicit and implicit therapy
approaches, utilising semantically based tasks, to target a comprehension impairment in a
participant who demonstrated difficulty accessing the semantic system. Therapy was conducted over two separate phases to enable comparison between the two approaches. The
first phase, implicit therapy, required the participant to read definitions and match synonyms with the target word. While results indicated improved comprehension following the
implicit therapy, maintenance of these improvements was poor with a significant decline
noted in the individual’s performance two weeks post-treatment. Explicit therapy, however,
where the participant was required to reproduce definitions from written and auditory
models and produce synonyms for target words, resulted in improved comprehension that
was maintained across the two weeks without therapy. The authors suggested that the
difference in maintenance of improvement could be attributed to the design of explicit
therapy activities, combined with the participant’s strengths and weaknesses. It could also be
suggested that the active participation required by the participant during the explicit therapy
may have assisted with improved maintenance of results (Francis et al., 2001; Hickin, Best,
Herbert, Howard, & Osborne, 2002; Leonard, Rochon, & Laird, 2008). No generalisation of
improved comprehension was observed on untreated items.
Behrmann and Lieberthal (1989) conducted a single case experimental design study
investigating remediation of a semantically based comprehension deficit using therapy
that employed analysis of semantic features. In semantic feature analysis (SFA)
approaches, the patient is encouraged to explore semantic features of the target word,
such as the properties, category, location, shape, and colour that contribute to the target
word’s meaning (see Boyle & Coelho, 1995; Conley & Coelho, 2003; Boyle, 2004; Davis,
2005; Rider, Wright, Marshall, & Page, 2008). In their study (Behrmann & Lieberthal,
1989), the participant, a 57-year-old individual who had Global aphasia and was three
months post-onset of a left middle cerebral artery infarct, demonstrated severe expressive
and receptive deficits secondary to a breakdown at the level of semantic processing. Unlike
SFA approaches typically discussed within naming studies, Behrmann and Lieberthal
(1989) explored the semantic features of target words with a two-stage treatment; first,

CLINICAL LINGUISTICS & PHONETICS

3

teaching the general concept of a category and then teaching specific details of items
within a category. Stimuli consisted of primarily spoken and written words with pictorial
stimuli used for some items only.
Following therapy, results indicated improved comprehension of treated single words
with some generalisation to control items; however, no follow-up assessment was included
in the study to investigate maintenance of these results over time. Behrmann and
Lieberthal (1989) argued that improvements observed post-treatment were the result of
therapy and not spontaneous recovery, as the participant demonstrated no significant
fluctuations in comprehension ability on pre-treatment assessments, which were conducted using spoken and written stimuli, over a three-week period. Although the study
showed promising findings, limited information was provided regarding feedback utilised
for correct or incorrect responses from the participant during the treatment, making it
difficult to replicate this study.
Bastiaanse et al. (1993) described another word-level therapy approach to target comprehension impairment, the Auditory Language Comprehension Program (ALCP). According
to the authors, the ALCP is a structured programme consisting of semantic, phonological and
complement (combination of semantic and phonological tasks) phases, with each phase
containing a number of blocks that progressively challenge the patient. The participant is
required to select a picture in response to a verbal label, from a selection of four black and
white pictures. During pre-treatment assessments, Bastiaanse et al. (1993) determined that
their participant, a 37-year-old male who was approximately three months post-onset,
presented with Wernicke’s aphasia with semantic impairments being the most salient feature;
therefore, only the semantic phase of the ALCP was completed.
The results of their study revealed significant improvement in the participant’s comprehension of single words, with generalisation of improved comprehension to control
words and sentence level tasks. One unique aspect of this study, when compared to other
comprehension studies, was the improvement noted in the participant’s spoken language,
specifically, his naming ability. Unfortunately, no measures of maintenance were included
to determine the long-term effect of therapy. Additionally, Bastiaanse et al. (1993) did not
complete the pre-treatment assessments necessary to discount spontaneous recovery
influencing therapy outcomes, a particularly important factor given that the participant
was three months post-onset and may still have been demonstrating spontaneous recovery
(Pulvermüller et al., 2001; Meinzer et al., 2004).
Finally, a more recent study (Morris & Franklin, 2012) investigated treatment of
comprehension deficits in a 69-year-old male who was 14 months post-onset of stroke
(participant one), and a 59-year-old male who was 16 months post-onset of stroke
(participant two). On pre-treatment assessment, both participants demonstrated phonological processing difficulties and significant impairments in accessing semantic information. The initial baseline assessments were administered to eliminate spontaneous recovery
as a factor for any change observed during the treatment study; however, details pertaining to the timing of these assessments are limited. Both participants received the same
total number of therapy hours, although the intensity and scheduling of therapy sessions
differed across the two participants. During therapy, the participants were presented with
images of 100 treatment items and provided with a verbal label, either the target or a
semantically related name, and were asked to either accept or reject the label based on the

4

P. MUNRO AND S. SIYAMBALAPITIYA

image presented. Participants were provided with immediate feedback regarding the
accuracy of their response.
Following therapy, participant 2 demonstrated improved comprehension on both treated
and untreated items. While participant two demonstrated generalisation to untreated items in
the control set, no significant improvements were observed on additional similar posttreatment assessments (i.e. picture verification task), indicating a lack of generalisation to
other tasks. Improved performance on treated and untreated items within the therapy tasks
was maintained on assessment four months post-treatment. In contrast, participant one
showed no statistically significant improvement on any therapy tasks. Morris and Franklin
(2012) suggested that this significant difference in performance could possibly be the result of
differing levels of impairment between the two participants wherein one participant had an
impairment that was primarily semantic in nature (i.e. implicating meaning) whereas the
other had an impairment that was lexical-semantic (i.e. implicating word form).
Previous research (Behrmann & Liberthal, 1989; Bastiaanse et al., 1993; Francis et al., 2001;
Morris & Franklin, 2012) has shown that appropriate therapy can lead to improvement in
semantic comprehension in individuals with aphasia; however, further research is needed due to
the limited number of existing studies, lack of methodological inclusions in some studies (e.g.
thorough pre-treatment baseline assessments, follow-up assessment to investigate maintenance
of treatment effects), along with the diversity of aphasia types targeted across treatment studies.
Further studies targeting semantically based comprehension deficits in people with aphasia are
needed to strengthen and extend the findings of preceding studies. A treatment approach that
merits further exploration is the use of SFA to treat comprehension deficits.
Numerous naming studies have employed a SFA or modified SFA approach with
significant improvements observed post-treatment for treated items and with improvements maintained following the withdrawal or completion of therapy (Boyle & Coelho,
1995; Drew & Thompson, 1999; Conley & Coelho, 2003; Boyle, 2004; Coelho, Davis, 2005;
Rider et al., 2008; Rose & Douglas, 2008). In her review of SFA treatment studies conducted
with aphasic individuals with anomia, Boyle (2010) notes that SFA has been used to refer to
a number of treatments that deviate from the original treatment developed by Ylvisaker and
colleagues (1985, as cited in Boyle, 2010). The original SFA treatment required participants
to analyse a concept and generate semantic features, whereas some later studies employing
the label SFA have not required features to be generated. Instead, participants have been
required to employ a different method of processing semantic features such as selecting
features from amongst distractors or verifying that a semantic feature is related to a concept.
Long-lasting positive results observed within naming studies employing a semanticbased therapy approach, such as SFA, have been attributed to the active role individuals
are required to take during therapy tasks for semantic feature exploration and production (Hickin et al., 2002; Boyle, 2004; Leonard et al., 2008; Boyle, 2010). Results for
generalisation of improved naming to control items following SFA and SFA-based
treatments remain variable within the literature (Boyle & Coelho, 1995; Rider et al.,
2008; Rose & Douglas, 2008; Boyle 2010). As very few participants in the treatment
studies have presented with severe aphasia, Boyle (2010) has also highlighted that more
studies are needed employing SFA with individuals with more severe diagnoses. Given
the centrality of semantics to both language production and comprehension, the
mechanism that underlies SFA in treatment of naming deficits may also produce
positive effects in the treatment of comprehension impairment. Some evidence of this

CLINICAL LINGUISTICS & PHONETICS

5

was found in the study by Behrmann and Lieberthal (1989), in which positive results
were demonstrated when implementing an approach involving analysis of semantic
features to target comprehension deficits. A limitation of the existing literature for
comprehension treatment approaches is that only a single study (Francis et al., 2001)
included measures investigating maintenance of improved comprehension ability following withdrawal of therapy.
The aim of the present study, therefore, was to further investigate the use of semantically based therapy to remediate comprehension deficits at word level using a modified
SFA approach in an individual with Global aphasia. We have employed the term ‘modified
SFA’ to highlight the different task employed in comparison to traditional SFA studies. In
the present study, participants were required to select the correct semantic feature from a
choice of two, rather than generating the semantic feature themselves. The modifications
to the original SFA procedure were necessary to adapt the task to target comprehension
and to accommodate the lack of nonverbal output that individuals with severe aphasia
may present with. To strengthen the design of the present study, in comparison to past
studies investigating remediation of semantically based comprehension deficits, several
factors were taken into account. The present study aimed to investigate remediation of
chronic aphasia, to allow observation of treatment effects in isolation of any possible
effects of spontaneous recovery. Unlike individuals with aphasia in the acute phase, such
as participants in the studies by Behrmann and Lieberthal (1989) and Bastiaanse et al.
(1993), individuals with chronic aphasic (greater than 12 months post-onset of stroke;
Meinzer et al., 2004; Barthel et al. 2007) are unlikely to improve spontaneously as a result
of neural recovery; however, they have previously demonstrated significant improvement
following provision of naming therapy (Pulvermüller et al., 2001; Meinzer et al., 2005;
Moss & Nicholas, 2006; Barthel et al., 2007). In addition, as previous studies have failed to
provide conclusive evidence regarding maintenance of improvements following the withdrawal of therapy, this study also incorporated follow-up assessments at 12 weeks posttreatment in order to determine the longer term benefits, if any, of semantically based
therapy for comprehension deficits. Finally, therapy was delivered at a relatively high
intensity (10 one-hour sessions over two weeks), since several studies have found that
increased intensity of treatment may enhance therapy success for individuals with chronic
aphasia (Hinckley & Craig, 1998; Pulvermüller et al., 2001; Bhogal, Teasell, & Speechley,
2003; Meinzer, Djundja, Barthel, Elbert, & Rockstroh, 2005; Raymer, Kohen, & Saffell,
2006; Barthel, Meinzer, Djundja, & Rockstroh, 2007; Cherney, Patterson, Raymer,
Rymark, & Schooling, 2008). It was anticipated, based on previous findings from both
comprehension and naming studies, that an individual with chronic Global aphasia with a
semantically based comprehension deficit would demonstrate improved single word
comprehension following modified SFA therapy.

Method
Ethical approval
This single case treatment study was completed in accordance with ethical approval from
the Human Research Ethics Committees of the relevant University and local Health
Service District.

6

P. MUNRO AND S. SIYAMBALAPITIYA

Participant details
Case history
The participant, MB, was a 69-year-old male who was a native English speaker. He was 1
year, 8 months post-onset of a left hemisphere ischaemic stroke when he participated in
the present treatment study. Neuroimaging records were not available to indicate the exact
location and extent of the lesion; however, following the stroke, MB presented with a
right-sided hemiplegia and speech pathology records indicated a diagnosis of Global
aphasia and severe apraxia of speech. At the time of participating in the present study,
MB remained largely nonverbal, consistent with his original diagnosis. MB’s handedness
was assessed using Bryden’s (1982) hand preference questionnaire which indicated that he
was right-handed for all tasks. MB had completed 7 years of formal education and had
owned his own earthmoving business up until his stroke. MB reported corrected-tonormal vision when wearing prescription glasses and wore hearing aids to correct an
industrial hearing loss. As a result of his hearing loss, the participant reported difficulty
hearing with background noise; therefore, all assessments and therapy sessions were
completed in a quiet clinical room with only the clinician, participant and participant’s
wife present.
Language profile
Prior to commencing treatment, a battery of assessments was administered to develop an
understanding of MB’s current language abilities. MB achieved an Aphasia Quotient of
14.4 on the Western Aphasia Battery Revised (WAB-R; Kertesz, 2006), with severe
impairment of both auditory comprehension and verbal expression, consistent with a
diagnosis of severe Global aphasia (see Table 1 for a summary of all language assessment
results). MB demonstrated greater success for single word comprehension when using real
objects compared to pictured objects (which consisted of black and white line drawings)
on the WAB-R; however, all tasks proved difficult for the participant. MB produced very
low scores on verbal expression tasks such as naming, sentence completion (e.g. The grass
is ____) and responsive speech (e.g. How many days are there in a week?). In the naming
subtest, tactile, phonological and semantic cues did not appear to assist with word
retrieval.
The picture–picture version of the Pyramids and Palm Trees test (PPT; Howard and
Patterson, 1992) and subtest 47 (Spoken Word-Picture Matching) of the Psycholinguistic
Assessment of Language Processing in Aphasia (PALPA; Kay, Lesser & Coltheart, 1992)
were also administered to provide more detailed information about the nature of MB’s
comprehension deficit. Results on pre-treatment standardised assessments indicated that
MB was demonstrating a semantically based comprehension deficit.

Selection of treatment stimuli
A spoken word–picture matching task was devised to identify appropriate items to target
in the treatment programme and to monitor for changes in comprehension ability across
the course of the study. Prior to commencing treatment, a total of 80 picture sets were
developed for use in the spoken word–picture matching task. Each picture set consisted of
four separate images: the target image and three distractors. Photographs or coloured

CLINICAL LINGUISTICS & PHONETICS

7

Table 1. Summary of MB’s language assessment results from pre- and post-treatment.
Test
Western Aphasia Battery Revised
Spontaneous Speech
Auditory Verbal Comprehension
Repetition
Naming and Word Finding
Aphasia Quotient
Pyramids and Palm Trees
PALPA subtest 47: Spoken Word-Picture Matching
Items correct
Close semantic errors
Distant semantic errors
Visual errors
Unrelated errors
Spoken Word–Picture Matching task
Items correct
Semantic errors
Phonological errors
Unrelated errors
a

Total score possible

Pre-treatment score

Post-treatment score

20
10
10
10
100
55

4 (20%)
2.5 (25%)
0.4 (4%)
0.3 (3%)
14.4 (14.4%)
38 (69%)

4 (20%)
2.85 (28.5%)
1 (10%)
0.7 (7%)
17.1 (17.1%)
28 (50.9%)

40

20 (50%)
11 (27.5%)
3 (7.5%)
3 (7.5%)
3 (7.5%)

23 (57.5%)
9 (22.5%)
6 (15%)
2 (5%)
0 (0%)

26

12a (46%)
8 (30.7%)
3 (11.5%)
3 (11.5%)

24b (92.3%)
2 (7.7%)
0 (0%)
0 (0%)

Average number of correct responses over three pre-treatment assessments.
Average number of correct responses over three post-treatment assessments.

b

illustrations of real objects (8 × 8 cm) were used as picture stimuli to try to make the
treatment material more aphasia friendly, since some people with aphasia have previously
shown a preference for photographic illustrations (Rose, Worrall, Hickson & Hoffman,
2012). Target images were selected using four different categories from published category
norms (Van Overschelde, Rawson, & Dunlosky, 2004): food, animals, furniture and
clothing. Distractors included a semantic distractor (an item from the same semantic
category), a phonological distractor (an item which shared at least an initial phoneme with
the target) and a familiarity distractor (an unrelated item which was a target from another
set). For example, for the target word dolphin, the distractors were pig (semantic), doll
(phonological) and mushroom (familiarity).
To complete the spoken word–picture matching task, the participant was asked to
identify the target picture in these picture sets in response to a verbal request e.g. “Show
me the ______”. Items were presented in random order and no feedback was provided to
the participant regarding his choice. The spoken word–picture matching task was completed on three separate occasions over a period of a week; the targets that were incorrectly
identified on two or more occasions were selected to be used within the treatment
programme.
A total of 26 items were identified as problematic for the participant and were then divided
into two sets labelled ‘treatment’ and ‘control’, with 13 items in each set. Items in the treatment
and control groups were statistically matched (Unpaired Wilcoxon signed rank test: Category
z = 0.000, p = 0.540, Familiarity z = −0.722, p = 0.242, Imageability z = −0.880, p = 0.196,
Kucera-Francis written frequency, z = −0.245 p = 0.410, Syllable length, z = −0.144, p = 0.500),
using category norms (Van Overschelde et al., 2004) and normative data for familiarity,
imageability, Kucera–Francis word frequency and word length from the MRC
Psycholinguistic Database (Wilson, 1988). Once the 26 items were allocated to treatment
and control sets, the spoken word–picture matching was again conducted for the 26 problematic targets to ensure the participant had comparable comprehension for treatment and
control sets pre-treatment. Targets and distractors were presented in random order and

8

P. MUNRO AND S. SIYAMBALAPITIYA

configuration, on three separate but consecutive occasions over a period of one week. Fisher’s
Exact Test indicated no significant difference in the participant’s comprehension ability of
treatment versus control items across the three pre-treatment assessments (pre-treatment 1:
Fisher’s Exact, χ2 = 3.166, p = 0.371; pre-treatment 2: Fisher’s Exact, χ2 = 0.982, p = 1.000; pretreatment 3: Fisher’s Exact, χ2 = 1.463, p = 0.861).
Following the identification and allocation of items to the treatment or control condition,
modified semantic feature analysis sets were developed for each of the treatment items.
Semantic feature analysis sets consisted of a either a large photograph or coloured illustration of the treatment item (14 × 11 cm) in the centre of a white A4 (29.7 × 21.0 cm) page,
surrounded by four pairs of printed words representing semantic features. Within each pair
of semantic features, one item was consistent with the treatment target and the other was a
distractor (see Figure 1 for a schematic example). Semantic features were presented in pairs
to allow the participant to actively engage in semantic processing, by making a conscious
decision regarding which feature was correct, without having to make a verbal response.
Written words were utilised to maximise the modalities the participant could access in order
to select a semantic feature during therapy, given his expressive difficulties (e.g. the
participant was able to point to, or attempt to verbalise, the chosen semantic feature).
Semantic feature categories were identified from categories outlined in previous SFA studies
(Boyle and Coelho, 1995; Coelho et al., 2000; Boyle, 2004). While the original SFA design
utilises six categories, including group, use, action, properties, location and associations for
each item, the semantic features explored within this study were limited to four options to
minimise the cognitive load and maximise the participant’s ability to participate despite his
comprehension deficits. Semantic features included the group (group or category the item
belonged to), the location (where the item could be found), the use (how or what the item
was used for) and properties (a property iconic for the item).

desert
forest

gills
fur

drink
Photographic image of a
BEAR

blue
brown

Figure 1. Example of semantic feature analysis treatment stimulus.

animal

CLINICAL LINGUISTICS & PHONETICS

9

Treatment procedure
Pre-treatment
Following the administration of standardised assessments to establish the nature of MB’s
language disorder, and the selection of treatment stimuli using a spoken word–picture
matching task; multiple pre-treatment assessments were then conducted to determine stability
of comprehension ability across treatment and control tasks (Franklin, 1997; Tate et al., 2008)
and to establish the presence or absence of spontaneous recovery. The spoken word–picture
matching task (consisting of all 26 treatment and control items) was administered, as
previously described, on three consecutive but separate occasions, over the period of a
week. Items were presented in random order for each pre-treatment assessment and no
feedback was provided regarding whether the participant’s response was correct. No statistical
difference was observed across the three pre-treatment assessments, suggesting that the
participant’s comprehension was stable and that he was beyond the period of spontaneous
recovery (pre-treatment 1 vs. pre-treatment 2: Fisher’s Exact, χ2 = 9.632, p = 0.342; pretreatment 1 vs. pre-treatment 3: Fisher’s Exact, χ2 = 9.393, p = 0.567; pre-treatment 2 vs. pretreatment 3: Fisher’s Exact, χ2 = 10.541, p = 0.201). This strengthens the notion that any
changes in comprehension ability observed post-treatment would be attributable to the
treatment programme as opposed to natural recovery of function or normal variation.
Treatment
Over the following two weeks, a total of ten therapy sessions of one hour were completed (see
Appendix for treatment schedule). In each session, therapy was provided for all 13 treatment
items using the modified semantic feature analysis sets. For each item, the participant was
asked to indicate which semantic feature from each of the four pairs was consistent with the
pictured target item (e.g. “Does a bear have fur?” or “Does a bear have gills?”; with the therapist
pointing to each written semantic feature as it was spoken). The participant’s response was
acknowledged as either correct or incorrect and the correct response was either restated or
explained (e.g. “Yes, the bear has fur” or “No, the bear has fur”).
Mid-treatment probes were conducted at the beginning of every third session to
monitor for improved comprehension and/or generalisation of improvement to control
items (see Appendix). For these probes, the participant completed the spoken word–
picture matching task for all treatment and control items. No feedback or cues were
provided during mid-treatment probes.
Post-treatment
Following completion of the treatment phase, post-treatment assessment was conducted
by administering the spoken word–picture matching task with all treatment and control
items. Post-treatment assessments were conducted on three separate occasions during a
period of one week to determine stability of results. Follow-up assessment was administered 12 weeks after the completion of the treatment phase to determine whether changes
in comprehension ability were maintained in the long-term, following withdrawal of
therapy. This follow-up assessment included the re-administration of standardised assessments (WAB-R, PPT, and PALPA subtest 47) and administration of the spoken word–
picture matching task.

10

P. MUNRO AND S. SIYAMBALAPITIYA

Results
Treatment items
Results from mid-treatment probes, completed at regular intervals throughout the treatment programme, indicated a progressive increase in correct responses for treatment
items (see Figure 2). Comparison of correct responses from the spoken word–picture
matching task at pre- (M = 6) and post-treatment (M = 12) indicated a significant change
in comprehension of treatment words following therapy (items correct at pre-treatment
vs. items correct at post-treatment: Wilcoxon signed rank test, exact, z = −2.705, p = 0.003;
see Table 2). Error patterns were also noted to change from semantic, phonological and
unrelated errors pre-treatment to solely semantic errors at post-treatment.

Control items (generalisation)
Analysis of control items at pre (M = 6) and post-treatment (M = 12.333) indicated that
there was also a statistically significant increase of correct responses observed for control
items (items correct at pre-treatment vs. items correct at post-treatment: Wilcoxon signed
rank test, exact, z = −2.980, p = 0.0001; see Table 2).

Maintenance

14
12
10
8
6

Treatment
Control

4
2

w

-u

p

3
Fo
llo

Po

st

-T
x

2
Po

st

-T
x

1
st
Po

id
M

-T
x

-T
x3

2
id
M

Tx

1
id
M

Tx

3
Pr
e

-T
x

2
-T
x
Pr
e

-T
x

1

0
Pr
e

No. of correct responses (N=13)

Results from the follow-up assessment, administered 12 weeks after the completion and
withdrawal of therapy, indicated that improvements observed at post-treatment assessment were maintained, with only one less correct response for treatment items and two
less for control items. No statistically significant difference was observed between the
final post-treatment assessment and follow-up assessment indicating comprehension
ability was maintained over the 12 weeks (treatment items: Wilcoxon signed rank test,
exact, z = −.577, p = 0.500; control items: Wilcoxon signed rank test, exact, z = −1.342,
p = 0.250) (see Table 2).

Session

Figure 2. Accuracy of comprehension for treatment and control items across treatment schedule.

CLINICAL LINGUISTICS & PHONETICS

11

Table 2. Accuracy of responses across treatment schedule.

Group
Treatment
items (N =
13)
Control items
(N = 13)

Correct
Correct
responses
responses
at postPrePrePrePostPostPostat pre- vs treatment
treatment treatment treatment treatment treatment treatment Followpost3 versus
1
2
3
1
2
3
up
treatmenta follow-upa
5
6
7
12
12
12
11
p = .003*
p = .500
7

5

6

12

12

13

11

p = .0001*

p = .250

a
Wilcoxon signed rank test.
*p ≤ 0.05 statistically significant.

Post-treatment language profile
Post-treatment results on all standardised assessment, except the PPT, indicated an
improvement in scores, including an improvement in the WAB-R AQ from 14.4 to
17.1. Following treatment, the score on the spontaneous speech subtest of the WABR demonstrated no change, while small improvements were observed on auditory
comprehension (3.5%), repetition (6%), and naming (0.4%) subtests (see Table 1).
An overall improvement of 7.5% was observed for PALPA subtest 47 (Spoken WordPicture Matching). Error response patterns also changed from pre- to post-treatment. At pretreatment, assessment errors included close semantic (27.5%), distant semantic (7.5%), visual
(7.5%) and unrelated errors (7.5%). At post-treatment, errors consisted of close semantic
(22.5%), distant semantic (15%) and visual errors (5%) only (see Table 1). The participant’s
performance on the PPT demonstrated a reduction in his overall score to 18.1% (see Table 1).

Discussion
The present study provides some support that using a modified SFA therapy approach was
effective in improving auditory comprehension of single words in a person with Global
aphasia. Following therapy, the participant demonstrated improved auditory comprehension of treatment items, with generalisation to control items within the same semantic
categories (i.e. food, animals, clothes, and furniture). Furthermore, follow-up assessment
completed 12 weeks after the completion of therapy indicated that improved comprehension of both treatment and control items was maintained following the withdrawal of
therapy. The findings of the present study provide preliminary evidence that a modified
SFA treatment can be used to improve comprehension deficits in individuals with aphasia,
extending the findings of previous studies that have employed this approach to treat
semantically based naming deficits (see Boyle & Coelho, 1995; Coelho et al., 2000; Kiran &
Thompson, 2003; Boyle, 2004; Rider et al., 2008).
As the participant demonstrated stable comprehension across multiple pre-treatment
assessments, it seems likely that the improvements observed in the participant’s comprehension post-treatment are attributable to the treatment programme itself and not spontaneous recovery. Additionally, post-treatment results on standardised assessments (i.e.
PPT, PALPA subtest 47, WAB-R) provided further evidence to negate the possibility of

12

P. MUNRO AND S. SIYAMBALAPITIYA

improvement being due to spontaneous recovery. For example, the participant demonstrated only a small degree of improvement on tasks not targeted in therapy such as
repetition or naming. If spontaneous recovery is occurring, all aspects of language should
improve together (Franklin, 1997), therefore, the disproportionate change in ability across
tasks suggests that the participant was not experiencing spontaneous or natural recovery
of function.
Mechanism of improvement in comprehension
The participant’s improvement in auditory word comprehension following therapy may
be explained using the semantic processing concept of automatic spreading activation
(Collins & Loftus, 1975; Neely, 1991). According to this concept, word meanings are
stored as nodes which are interconnected to other nodes based on a semantic relationship
or association. As a word is accessed, activation spreads to other words which are
semantically related to the retrieved word. This activation can facilitate the retrieval of
related words by reducing the threshold which must be reached in order to access the
related words.
A semantically based deficit occurs when these semantic representations or connections
between words are damaged, reducing an individual’s ability to access words either
directly or through spreading activation (Boyle & Coelho, 1995; Coelho et al., 2000;
Boyle, 2004). SFA therefore aims to remediate semantic deficits by strengthening semantic
connections and networks (Boyle & Coelho, 1995; Coelho et al., 2000; Boyle, 2004). The
findings of the present study further support results of previous naming studies that have
used SFA or modified-SFA approaches, which suggest that by providing the participant
with opportunities to repetitively access and activate semantic networks, it is possible that
connections and networks targeted will be reorganised and strengthened. Strengthening
semantic networks will in turn allow for enhanced semantic processing, through improved
access to the semantic system and enriched semantic representation (Boyle & Coelho,
1995; Boyle, 2004).
In the present study, encouraging the participant to explore semantic features that were
closely related to the target word may have improved the retrieval of concepts via the
activated semantic network. Activating a number of semantic features related to the target
provides it with spreading activation from several related nodes, thus allowing threshold
to be reached and facilitating word retrieval. Nodes and connections that were previously
degraded due to damage caused by the stroke may have been strengthened via the therapy
task. Additionally, the repetitive nature of therapy is likely to have contributed to
strengthening semantic representations over time, allowing threshold for activation to
be reached more easily for target items.
Generalisation and maintenance of treatment effects
Generalisation of improved comprehension to untreated items following semantically
based therapy approaches has been inconsistent across previous comprehension treatment
studies. In the current study, generalisation to control items may have been due to the
treatment and control items being from the same semantic categories (i.e. food, animals,
clothes and furniture). Boyle (2004) suggests that an SFA approach to therapy enables

CLINICAL LINGUISTICS & PHONETICS

13

category-specific generalisation through the repetitive stimulation of a category, consequently improving the participant’s ability to access words within that category. According
to the concept of automatic spreading activation, by accessing a word or concept such as a
category name (i.e. food, animals, clothes, and furniture), words that share a semantic
connection with this word also receive activation (Collins & Loftus, 1975; Neely, 1991). In
the present study, the participant was required to identify the semantic category to which
the target belonged and by doing so was presumably activating the surrounding semantic
network, causing activation to spread to other items within that category.
In terms of maintenance of treatment effects, previous studies suggest that better
maintenance in semantically based treatment approaches can be explained by active
participant involvement during treatment (e.g. producing a definition, identifying correct
semantic features, deciding whether semantic information is correct or not) (Francis et al.,
2001; Hickin et al., 2002; Leonard et al., 2008). Findings from the present study also
highlight the possible role of participant involvement as a beneficial factor in establishing
better maintenance of results. By presenting semantic features in pairs, one correct and
one incorrect, the participant was required to actively engage in the treatment process.
This active involvement may have led the participant to engage in semantic processing
and possibly assisted in internalising the therapy process.
Changes in error response patterns
In addition to improved and maintained comprehension of treatment and control items, a
qualitative change in the error response pattern demonstrated on post-treatment assessment may also support a change in the participant’s central semantic ability. Prior to
therapy, error responses on the spoken word–picture matching task developed for the
treatment programme included semantic, phonological and unrelated errors. Following
therapy, errors were purely semantic. This qualitative change may suggest that the treatment improved the participant’s access to semantic information relating to the target
word. As previously discussed, this change to the semantic system may have been achieved
through a general strengthening of semantic representations and processes, improving the
participant’s ability to effectively access and select semantically related items, resulting in
less phonological or unrelated errors. Given the small amount of error data available, these
hypotheses are preliminary until confirmed by replication of the present results.
Overall changes to language ability
Overall results on PALPA subtest 47 (Spoken Word–Picture Matching) showed an
increase in the number of correct responses, suggesting a positive change in the participant’s general comprehension of single words. The smaller degree of improvement
observed on PALPA subtest 47, in comparison to the spoken word–picture matching
task used in the treatment programme, could be explained by the broader diversity of
semantic categories on PALPA subtest 47. Reduced levels of improvement on PALPA
subtest 47 may also be explained by the different number of distractors presented for each
target (i.e. PALPA subtest 47 had four distractors while the spoken word–picture matching task from the treatment programme had three distractors). Findings from a study by
Pierce, Jareci and Cannito (1990) concluded that increasing the number of stimuli

14

P. MUNRO AND S. SIYAMBALAPITIYA

presented simultaneously increases the cognitive load, which in turn reduces the participant’s ability to make a decision. These conclusions are based on the resource allocation
model which suggests that by increasing the number of stimuli presented, the individual is
required to scan, process and retain more information while engaging in semantic
processing, therefore increasing cognitive demands and reducing cognitive resources
available to complete the task (Pierce et al., 1990; Howland & Pierce, 2004).
Comparison between pre- and post-treatment results on PALPA subtest 47 also
indicated a change in error responses. Initially, errors consisted of close semantic, distant
semantic, visual and unrelated errors. Following treatment, errors consisted primarily of
close semantic (53%; 9/17 errors) and distant semantic errors (35%; 6/17 errors). This
pattern is consistent with that seen on the spoken word–picture matching task developed
for the treatment programme and provides further evidence that the participant showed
improved semantic processing post-therapy, allowing him to activate semantic networks
closer to, or related to the target, as opposed to selecting or accessing semantically
unrelated information (i.e. phonological, visual or unrelated distractors).
The participant’s superior performance on the spoken word–picture matching task
from the treatment programme, relative to his performance on PALPA subtest 47 (Spoken
Word–Picture Matching) and the PPT, may be due to the visual appearance of the
assessment stimuli. Black and white line drawings are used as stimuli for both PALPA
subtest 47 and the PPT, while stimuli for the current study consisted of coloured photographs of real objects which may have been more representative of the target items and
facilitated semantic processing and comprehension to a greater degree. The less representative line drawing stimuli featured in the PPT may provide a possible explanation for
the unexpected finding seen on the PPT at post-treatment (i.e. scores on the PPT
decreased in contrast to improved scores on the other language assessment tasks).
The participant’s results on the WAB-R, a general language assessment, are also consistent
with a change in central semantic ability post-treatment, with minor improvements observed
on all subtests except for spontaneous speech (i.e. auditory comprehension, repetition,
naming). While these improvements were not clinically significant, they further suggest
that the participant is demonstrating improved semantic access and representation, since
semantic processing is central to both auditory comprehension and verbal expression.
Expressive ability was not a focus of this study since the participant presented with Global
aphasia and co-occurring apraxia of speech, therefore it is not surprising that expressive
ability did not show any change. Given the difficulties associated with treating language
deficits in individuals with Global aphasia, due to the limited initial spontaneous recovery
post stroke and slow overall progress with speech therapy (Alexander & Loverso, 1992), the
results of this study are particularly promising for this population.
Limitations and future directions
The findings outlined within this study remain preliminary and require further research
and replication to support the outcomes of this treatment approach. To further clarify and
consolidate the results observed within the present study, it would have been beneficial to
include an auditory comprehension task (i.e. spoken word picture matching task) that
included items from the same category that were not used as probes, as well as items from
other semantic categories.

CLINICAL LINGUISTICS & PHONETICS

15

While single case studies have been identified as a valid and appropriate means of
researching the efficacy of aphasia treatments, it is recognised that variables such as an
individual’s aphasia diagnosis or severity, and time post-onset, can influence response to
treatment (Holland et al., 1996; Franklin, 1997; Thompson, 2006). It would therefore be
advantageous to trial this treatment with a wider range of individuals with comprehension
deficits, who present with differing language profiles. Further single case experimental
studies or case series would strengthen the findings presented here.
Additionally, only a small number of items were identified as problematic for this
participant. Future studies should ideally aim to include a larger number of treatment and
control items to increase the power of results observed. The small number of stimuli did,
however, enable therapy to be provided in a comprehensive manner during the daily, one
hour sessions. Where there are a greater number of stimuli, the length of sessions would
need to be considered to ensure adequate time is available to implement the therapy
properly. Although treatment intensity was not a focus of this study, it is possible that
intensity contributed to the positive results observed.

Conclusions
The current study demonstrated that use of a modified SFA therapy approach may have
been beneficial in improving single word comprehension of an individual with Global
aphasia, who presented with a semantically based comprehension deficit. Results indicated
that significant improvements in single word comprehension were achieved for both
treatment and control items. Furthermore, results appeared to be maintained for 12
weeks following the withdrawal of therapy. These findings expand on conclusions made
by previous naming studies that use SFA, by indicating that this approach can also be used
successfully to address semantically based comprehension deficits. While the present study
provides positive evidence for the use of a modified SFA approach for individuals who
present with a semantically based comprehension deficit, replication of this study is
required to strengthen and expand the findings.

Acknowledgements
We wish to thank the participant MB for his dedicated participation in the therapy programme.
Portions of this work were presented at the 15th International Aphasia Rehabilitation Conference,
Melbourne, 2012.

Declaration of interest
The authors declare no conflicts of interest.

References
Albert, M. L. (1998). Treatment of aphasia. Archives of Neurology, 55, 1417–1419.
Alexander, M. P., & Loverso, F. (1992). A specific treatment for global aphasia. Clinical Aphasiology,
21, 277–289.

16

P. MUNRO AND S. SIYAMBALAPITIYA

Barthel, G., Meinzer, M., Djundja, D., & Rockstroh, B. (2007). Intensive language therapy in chronic
aphasia: Which aspects contribute most? Aphasiology, 22(4), 408–421.
Bastiaanse, R., Nijboer, S., & Taconis, M. (1993). The auditory language comprehension programme: A description and case study. European Journal of Disorders of Communication, 28,
415–433.
Behrmann, M., & Lieberthal, T. (1989). Category-specific treatment of a lexical-semantic deficit: A
single case study of global aphasia. International Journal of Language and Communication
Disorders, 24(3), 281–299.
Best, W., & Nickels, L. (2000). From theory to therapy in aphasia: Where are we now and where to
next? Neuropsychological Rehabilitation, 10(3), 231–247.
Bhogal, S. K., Teasell, T., & Speechley, M. (2003). Intensity of aphasia therapy, impact and recovery.
Stroke, 34, 000–000.
Boyle, M. (2004). Semantic feature analysis treatment for anomia in two fluent aphasia syndromes.
American Journal of Speech-Language Pathology, 13(3), 236–249.
Boyle, M. (2010). Semantic feature analysis treatment for aphasic word retrieval impairments:
What’s in a name? Topics in Stroke Rehabilitation, 17(6), 411–422.
Boyle, M., & Coelho, C. (1995). Application of semantic feature analysis as a treatment for aphasic
dysnomia. American Journal of Speech-Language Pathology, 4, (94–98).
Bryden, M. P. (1982). Laterality: Functional asymmetry in the intact brain. New York: Academic
Press.
Cherney, L. R., Patterson, J. P., Raymer, A., Rymark, T., & Schooling, T. (2008). Evidence-based
systematic review: Effects of intensity of treatment and constraint-induced language therapy for
individuals with stroke-induced aphasia. Journal of Speech, Language, and Hearing Research, 51
(5), 1282–1299.
Coelho, C. A., McHugh, R. E., & Boyle, M. (2000). Semantic feature analysis as a treatment for
aphasic dysnomia: A replication. Aphasiology, 14(2), 133–142.
Collins, A. M., & Loftus, E. F. (1975). A spreading activation theory of semantic processing.
Psychological Review, 52, 407–428.
Conley, A., & Coelho, C. (2003). Treatment of word retrieval impairment in chronic Broca’s
aphasia. Aphasiology, 17(3), 203–211.
Davis, L. A. (2005). Semantic feature analysis as a functional therapy tool. Contemporary Issues in
Communication Science and Disorders, 32, 85–92.
Drew, R. L., & Thompson, C. K. (1999). Model-based semantic treatment for naming deficits in
aphasia. Journal of Speech, Language, and Hearing Research, 42(2), 972–989.
Francis, D. R., Riddoch, M. J., & Humphreys, G. W. (2001). Cognitive rehabilitation of word
meaning deafness. Aphasiology, 15(8), 749.
Franklin, S. (1997). Designing single case treatment studies for aphasic patients. Neuropsychological
Rehabilitation, 7(4), 401–418.
Hall, D. A., & Riddoch, M. J. (1997). Word meaning deafness: Spelling words that are not understood. Cognitive Neuropsychology, 14(8), 1121–1164.
Hickin, J., Best, W., Herbert, R., Howard, D., & Osborne, F. (2002). Phonological therapy for wordfinding difficulties: A re-evaluation. Aphasiology, 16, 981–999.
Hinckley, J. J., & Craig, H. K. (1998). Influence of rate of treatment on the naming abilities of adults
with chronic aphasia. Aphasiology, 12(11), 989–1006.
Holland, A. L., Fromm, D. S., DeRuyter, F., & Stein, M. (1996). Treatment efficacy: Aphasia. Journal
of Speech and Hearing Research, 39, S27–S36.
Howard, D., & Patterson, K. (1992). The Pyramids and palm trees test. London: Hardcourt
Assessment.
Howland, J., & Pierce, R. (2004). Influence of semantic relatedness and array size on single-word
reading comprehension in aphasia. Aphasiology, 18(11), 1005–1013.
Kay, J., Lesser, R. & Coltheart, M. (1992). PALPA: Psycholinguistic assessments of language processing in aphasia. East Sussex, England: Lawrence Erlbaum.
Kertesz, A. (2006). Western Aphasia Battery Revised. San Antonio: Pearson Publishing.

CLINICAL LINGUISTICS & PHONETICS

17

Kiran, S., & Thompson, C. K. (2003). The role semantic complexity in treatment of naming deficits:
Training semantic categories in fluent aphasia by controlling exemplar typicality. Journal of
Speech, Language, and Hearing Research, 46(6), 773–787.
Leonard, C., Rochon, E., & Laird, L. (2008). Treating naming impairments in aphasia: Findings
from a phonological components analysis treatment. Aphasiology, 22(9),923–947.
Meinzer, M., Djundja, D., Barthel, G., Elbert, T., & Rockstroh, B. (2005). Long-term stability of
improved language functions in chronic aphasia after constraint-induced aphasia therapy, Stroke,
36, 1462–1466.
Meinzer, M., Elbert, T., Wienbruch, C., Djundja, D., Barthel, G., & Rockstroh, B. (2004). Intensive
language training enhances brain plasticity in chronic aphasia. BMC Biology, 2, 20.
Morris, J., Franklin, S., Ellis, A. W., Turner, J. E., & Bailey, P. J. (1996). Remediating a speech
perception deficit in an aphasic patient. Aphasiology, 10(2), 137–458.
Morris, J., & Franklin, S. (2012). Investigating the effect of a semantic therapy on comprehension in
aphasia. Aphasiology, 26(12), 1461–1480.
Moss, A., & Nicholas, M. (2006). Language rehabilitation in chronic aphasia and time postonset: A
review of single-subject data. Stroke, 37, 3043–3061.
Neely, J. H. (1991). Semantic priming effects in visual word recognition: A selective review of
current findings and theories. In D. Besner & G. Humphreys (Eds.), Basic processes in reading:
Visual word recognition (pp. 264–336). Hillsdale, NJ: Erlbaum.
Pierce, R., Jareci, J., & Cannito, M. (1990). Single word comprehension in aphasia: Influence of
array size, picture relatedness, and situational context. Aphasiology, 4, 155–165.
Pulvermüller, F., Neininger, B., Elbert, T., Mohr, B., Rockstroch, B., Koebbel, P., & Taub, E. (2001).
Constraint-induced therapy of chronic aphasia after stroke. Stroke, 32, 1621–1626.
Raymer, A. M., Kohen, F. P., & D, Saffell. (2006). Computerised training for impairments of word
comprehension and retrieval in aphasia, 20(2), 257–268.
Rider, J. D., Wright, H. H., Marshall, R. C., & Page, J. P. (2008). Using semantic feature analysis to
improve contextual discourse in adults with aphasia. American Journal of Speech-Language
Pathology, 17(2), 161–172.
Rose, M., & Douglas, J. (2008). Treating a semantic word production deficit in aphasia with verbal
and gesture methods. Aphasiology, 22(1), 20–41.
Rose, T. A., Worrall, L.E., Hickson, L.M., & Hoffmann, T.C. (2012). Guiding principles for printed
education materials: Design preferences of people with aphasia. International Journal of SpeechLanguage Pathology, 14(1), 11–23.
Sarno, M. T. (1998). Acquired aphasia. (3rd ed.). San Diego: Academic Press
Tate, R., McDonald, S., Perdices, M., Togher, L., Schultz, R., & Savage, S. (2008). Rating the
methodological quailty of single-subject and n-of-1 trails: Introducing the Single-Case
Experimental Design (SCED) scale. Neuropsychological Rehabilitation, 18(4), 385–401.
Tessier, C., Weill-Chounlamountry, A., Michelot, N., & Pradat-Diehl, P. (2007). Rehabilitation of
word deafness due to auditory analysis disorder. Brain Injury, 21(11), 1165–1174.
Thompson, C. K. (2006). Single subject controlled experiments in aphasia: The science and the state
of the science. Journal of Communication Disorders, 39(4), 266–291.
Van Overschelde, J. P., Rawson, K. A., & Dunlosky, J. (2004). Category norms: An updated and
expanded version of the Battig and Montague (1969) norms. Journal of Memory and Language,
50, 289–335.
Whitworth, A., Webster, J & Howard, D. (2005). A cognitive neuropsychological approach to
assessment and intervention in aphasia: A clinician’s guide. New York: Psychology Press.
Wilson, M. D. (1988). The MRC Psycholinguistic Database: Machine Readable Dictionary, Version
2. Behavioural Research Methods, Instruments and Computers, 20(1), 6–11.

18

P. MUNRO AND S. SIYAMBALAPITIYA

Appendix
Treatment schedule
Week 1: Pre-treatment assessment and stimuli selection
Session 1: Case history, consent forms, SWPM task with all 80 picture sets (stimuli selection)
Session 2: WAB-R, SWPM with all 80 picture sets (stimuli selection)
Session 3: PALPA subtest 47, PPT, SWPM with all 80 picture sets (stimuli selection)
Week 2: Pre-treatment assessment
Session 1: SWPM task with treatment and control items
Session 2: SWPM task with treatment and control items
Session 3: SWPM task with treatment and control items
Week 3: Treatment programme
Session 1: SFA task with treatment items
Session 2: SFA task with treatment items
Session 3: SFA task with treatment items, mid-treatment probea
Session 4: SFA task with treatment items
Session 5: SFA task with treatment items
Week 4: Treatment programme
Session 6: SFA task with treatment items, mid-treatment probea
Session 7: SFA task with treatment items
Session 8: SFA task with treatment items
Session 9: SFA task with treatment items, mid-treatment probea
Session 10: SFA task with treatment items
Week 5: Post-treatment assessment
Session 1: SWPM task with treatment and control items
Session 2: SWPM task with treatment and control items
Session 3: SWPM task with treatment and control items
Week 17: Post-treatment (Follow-up assessment)
Session 1: PALPA subtest 47, PPT
Session 2: WAB-R, SWPM with all 80 picture sets
Note. SWPM = Spoken Word Picture Matching; WAB-R = Western Aphasia Battery-Revised; PALPA = Psycholinguistic
Assessment of Language Processing in Aphasia; PPT = Pyramids and Palm Trees; SFA = semantic feature analysis.
Mid-treatment probe = SWPM task with treated and control items.

a