INT J LANG COMMUN DISORD, JANUARY–FEBRUARY 2014,

VOL. 49, NO. 1, 30–48

Research Report
Improving comprehension in adolescents with severe receptive language
impairments: a randomized control trial of intervention for coordinating
conjunctions
Susan H. Ebbels†‡, Nataša Marić†, Aoife Murphy∗ † and Gail Turner†
†Moor House School, Oxted, UK
‡Division of Psychology and Language Sciences, University College London, London, UK
(Received October 2012; accepted June 2013)

Abstract
Background: Little evidence exists for the effectiveness of therapy for children with receptive language difficulties,
particularly those whose difficulties are severe and persistent.
Aims: To establish the effectiveness of explicit speech and language therapy with visual support for secondary
school-aged children with language impairments focusing on comprehension of coordinating conjunctions in a
randomized control trial with an assessor blind to group status.
Methods & Procedures: Fourteen participants (aged 11;3–16;1) with severe RELI (mean standard scores: CELF4
ELS = 48, CELF4 RLS = 53 and TROG-2 = 57), but higher non-verbal (Matrices = 83) and visual perceptual
skills (Test of Visual Perceptual Skills (TVPS) = 86) were randomly assigned to two groups: therapy versus waiting
controls. In Phase 1, the therapy group received eight 30-min individual sessions of explicit teaching with visual
support (Shape Coding) with their usual SLT. In Phase 2, the waiting controls received the same therapy. The
participants’ comprehension was tested pre-, post-Phase 1 and post-Phase 2 therapy on (1) a specific test of the
targeted conjunctions, (2) the TROG-2 and (3) a test of passives.
Outcomes & Results: After Phase 1, the therapy group showed significantly more progress than the waiting controls
on the targeted conjunctions (d = 1.6) and overall TROG-2 standard score (d = 1.4). The two groups did not
differ on the passives test. After Phase 2, the waiting controls made similar progress to those in the original therapy
group, who maintained their previous progress. Neither group showed progress on passives. When the two groups
were combined, significant progress was found on the specific conjunctions (d = 1.3) and TROG-2 raw (d = 1.1)
and standard scores (d = 0.9). Correlations showed no measures taken (including Matrices and TVPS) correlated
significantly with progress on the targeted conjunctions or the TROG-2.
Conclusions & Implications: Four hours of Shape Coding therapy led to significant gains on comprehension of
coordinating conjunctions which were maintained after 4 months. Given the significant progress at a group level and
the lack of reliable predictors of progress, this approach could be offered to other children with similar difficulties
to the participants. However, the intervention was delivered one-to-one by speech and language therapists, thus
the effectiveness of this therapy method with other methods of delivery remains to be evaluated.

Keywords: intervention, randomized control trial, language impairment, language disorder, receptive language, Shape
Coding, adolescents.

What this paper adds?

What is already known on this subject?
Many children supported by speech and language therapy services have receptive language impairments. However,
there is little evidence of the effectiveness of speech and language therapy for receptive language difficulties, particularly
in older children and those with severe receptive language difficulties.

Address correspondence to: Susan Ebbels, Moor House School, Mill Lane, Hurst Green, Oxted RH8 9AQ, UK; e-mail:
ebbelss@moorhouseschool.co.uk.
∗
Present address: University of Limerick, Limerick, Ireland.
International Journal of Language & Communication Disorders
ISSN 1368-2822 print/ISSN 1460-6984 online  C 2013 Royal College of Speech and Language Therapists

DOI: 10.1111/1460-6984.12047
Improving comprehension in adolescents with severe receptive language impairments 31

What this study adds?

This randomized control trial with blind assessment shows that explicit speech and language therapy with visual
support can lead to significant gains on a specific test of the targeted structures and can also generalize to a standardized
test. Progress was maintained for 4 months. Therefore, a small amount of speech and language therapy can lead to
significant gains in comprehension, even for adolescents with severe receptive language difficulties.

abilities were correlated with procedural memory in typ-

Introduction ically developing children, but with declarative memory
Language impairments affect approximately 7% of chil- in children with SLI, indicating that children with SLI
dren (Tomblin et al. 1997) and for some their language may be using their largely intact declarative memory
impairment persists into adolescence, affecting their system to learn grammar instead of their impaired pro-
educational achievements (Conti-Ramsden 2008), and cedural memory system.
into adulthood, affecting their employment prospects Linguistic theories of SLI predict that providing
(e.g. Law et al. 2009). Language impairments persist enhanced but essentially normal language stimulation
more in children with both receptive and expressive lan- should have little effect on grammatical abilities and
guage impairments (RELI). Indeed, Clark et al. (2007) intervention should teach compensatory strategies us-
found that of 58 children with severe RELI at school ing the children’s relative strengths to help them learn
entry, only two had language scores in the normal range language. The PDH implies that intervention should
6 years later. They concluded that ‘receptive SLI [spe- use their largely intact declarative memory system and
cific language impairment] rarely resolves and trials of teach grammar explicitly to children with SLI. Also, if
therapy are urgently needed’ (p. 614). working and declarative memory are normal for visuo-
Despite the long-term, severe difficulties of chil- spatial information, then intervention could capitalize
dren with RELI, receptive language impairments have on their relative visual strengths. Thus, the PDH pre-
received relatively little research attention. This is partic- dicts the most effective interventions should explicitly
ularly the case for adolescents, who also receive limited teach the rules of language using visual support. In con-
professional services (Dockrell et al. 2006). trast, processing theories would predict that any inter-
vention which focuses on explicit teaching of linguistic
rules should exacerbate their difficulties, as this would
involve additional processing.
Theories of SLI and implications for intervention
Several theories of SLI have been proposed. Linguistic Intervention studies including a focus on improving
theories claim that grammatical difficulties are the pri- receptive language in school-aged children with
mary deficits. Processing theories, in contrast, propose
receptive language difficulties
that processing problems cause language difficulties, ei-
ther directly or because certain aspects of language are Reviews of the effectiveness of speech and language ther-
particularly vulnerable. apy for children with language impairments (e.g. Law
Ullman and Pierpont (2005) put forward a differ- et al. 1998, 2003) have found little reliable evidence
ent proposal: the Procedural Deficit Hypothesis (PDH), that speech and language therapy for receptive language
which hypothesizes that many children with SLI have difficulties can be effective and have concluded that ‘the
a deficit in procedural memory. This is involved in im- most substantial single gap in the literature [ . . . ] is
plicit acquisition, storage and use of knowledge and is the lack of good-quality literature about intervention
hypothesized to be used in implicitly learning the rule- for children with severe receptive language difficulties’
governed aspects of grammar. However, they proposed (Law et al. 2004: 935) and that there is ‘an overall lack of
that another memory system, declarative memory (which evidence for approaches to effective treatment for chil-
learns explicit information), is spared in SLI and can be dren with RELI’ (Boyle et al. 2010: 997). Of the few
used partially to compensate for the procedural memory studies that have investigated receptive language inter-
deficits, by learning rules explicitly. vention in school-aged children, several have failed to
The PDH is supported by several studies, but most find any significant effects (Bishop et al. 2006, Boyle
recently by Lum et al. (2012) who found that while et al. 2009, Gillam et al. 2008, McCartney et al. 2011).
children with SLI were impaired at procedural memory, Others, however, have found significant improvements
visuo-spatial working memory was intact, as was declar- in receptive vocabulary (Parsons et al. 2005, Throneburg
ative memory once working memory and language et al. 2000), specific grammatical structures (Ebbels and
deficits were controlled for. They found grammatical van der Lely 2001, Ebbels 2007), and general text and
32 Susan H. Ebbels et al.
sentence comprehension (Joffe et al. 2007, Tallal et al. tions and passives (Ebbels and van der Lely 2001) and
1996). The studies investigating receptive language in- dative sentences (Ebbels 2007). In contrast, a study
tervention at sentence level are discussed in more detail using implicit computerized training aiming to im-
below. prove comprehension of specific grammatical structures
(Bishop et al. 2006) found no difference between chil-
dren trained either with or without modified speech
Intervention for general sentence-level comprehension and untrained children who received only their ‘stan-
dard’ speech and language therapy package. These con-
Several studies of the Fast ForWord (FFW) programme
flicting results may be due to different approaches (ex-
(Scientific Learning Corporation 1998) (which con-
plicit versus implicit) or method of delivery (SLT versus
sists of highly intensive listening exercises recorded with
computer).
acoustically modified speech) have focused on improv-
ing receptive language. Early studies reported signifi-
cant improvement in children’s language comprehen- Intervention for adolescents with language impairments
sion (Tallal et al. 1996). However, recent independent
Of the studies discussed above, very few have included
large-scale randomized control trials (RCTs) (Cohen
adolescents with language impairments. The exceptions
et al. 2005, Fey et al. 2010, Gillam et al. 2008) found
are the studies by Ebbels and colleagues, where all the
control groups showed equal progress to those receiving
children were over 11 years and Bishop et al.’s (2006)
FFW.
study, which included children aged 8–13 years. In ad-
Studies by Boyle and McCartney and colleagues
dition, we know of two published studies of speech and
of commonly used interventions delivered in main-
language therapy for word finding difficulties which in-
stream schools (Boyle et al. 2009, McCartney et al.
cluded adolescents (Ebbels et al. 2012, Wright et al.
2011) found that on the Clinical Evaluation of Lan-
1993) and one on production of verb argument struc-
guage Fundamentals—3rd Edition (CELF-3) Receptive
ture (Ebbels et al. 2007).
Language Scale, children receiving speech and language
therapy showed no greater progress than controls. How-
ever, this could be due to the fact that very few of Summary
the therapy activities, as described by McCartney et al.
(2004), appear to include work on receptive language Receptive language impairments, particularly in adoles-
(with the exception of vocabulary, which is unlikely to cents, are neglected in the literature, despite receptive
have much impact on the CELF Receptive Language language difficulties affecting young people’s long-term
Scale). More concerning is the finding that the children life chances. Studies have shown that intervention aimed
with RELI made no progress with expressive language at improving the basic auditory processing of language
either. This was in contrast to those with Expressive and hence receptive language is probably not effective.
Language Impairment (ELI) who made progress with Generalized language intervention also does not seem
expressive language when they received therapy from a to improve receptive language (or expressive language
speech and language therapist (SLT) or SLT assistant in children with RELI). However, specific intervention
employed by the research project (Boyle et al. 2009), that explicitly teaches the rules of grammar and uses
but not when it was delivered using the ‘consultative visual support appears to be effective.
model’ (McCartney et al. 2011). Nevertheless, another
study (Ebbels et al. 2007) showed explicit therapy im- Aims of the current study
proved the production of verb argument structure in
children with RELI. Thus, it seems that school-aged This study aims to investigate the effectiveness of a
children with RELI can make progress in expressive lan- speech and language therapy method (Shape Coding;
guage with therapy when the intervention is explicit and Ebbels 2007) which explicitly teaches the rules of En-
targeted and the outcome measures are closely linked to glish using visual patterns of shapes as a support (thus
the intervention. fitting well with the PDH) using an RCT design. It
focuses on a specific area of comprehension and uses
measures which are closely related to the target and
Intervention for comprehension of specific grammatical also investigates generalization. For our specific targets,
we chose comprehension of coordinating conjunctions,
structures
as these occur frequently in classroom instructions and
Ebbels and colleagues showed that explicit speech and general conversation. Indeed, use of and and but is usu-
language therapy using ‘Shape Coding’ (which uses vi- ally introduced in Year 2 in English schools (albeit with
sual templates and patterns to explicitly teach the rules a focus on written English) and use of connectives in-
of grammar) improved comprehension of ‘wh’ ques- cluding and and but is needed to achieve the National
Improving comprehension in adolescents with severe receptive language impairments 33
Curriculum Level 2 in writing (the level expected of they would not meet the strict exclusionary criteria for
7 year olds). Written use of connectives is unlikely to SLI (listed in appendices A and B).
be accurate unless the children can understand them Selecting participants on the basis of comprehen-
and use them in speech. We targeted those coordinating sion difficulties, regardless of their other characteristics,
conjunctions which appear in the Test of Reception of allowed us to see which (if any) participant character-
Grammar (TROG-2) (Bishop 2003): but not, neither istics affected response to intervention. Given the na-
nor, not only but also, as we could then use responses to ture of the therapy, visual processing skills and gen-
this test as part of our outcome measures. eral non-verbal reasoning were of particular interest.
Thus, the primary research question was: For young We therefore collected information on their most recent
people with RELI of secondary school age, will 4 h scores on the TVPS (unavailable for two participants).
of one-to-one explicit teaching of coordinating con- Unfortunately, we had not carried out a non-verbal
junctions with visual support (Shape Coding) with test at the beginning of the study, but many partici-
their usual SLT improve their comprehension of tar- pants had been assessed on non-verbal tests previously.
geted coordinating conjunctions? The secondary re- Among these non-verbal tests, Matrices tests were most
search questions were: Will this have any effect on scores commonly included, although using a variety of tests
on a standardized test of comprehension of grammar (Kaufman Brief Intelligence Test (KBIT), Kaufman and
(TROG-2) or generalize to other grammatical structures Kaufman 2004; Wechsler Abbreviated Scale of Intel-
(passives)? ligence (WASI), Wechsler 1999; Wechsler Intelligence
Scale for Children (WISC-IV), Wechsler 2004). We
recorded these scores and for participants with no score
Method or scores more than a year old, we administered the Ma-
trices test of the Wide Range Intelligence Test (WRIT)
Participants (Glutting et al. 2000) after completion of the therapy.
This study was carried out at a specialist residential To enable easy comparison across tests, we converted
school for pupils with language impairments in the all Matrices scores to standard scores (mean of 100 and
UK. All pupils are tested at specific time points (at 11, standard deviation (SD) of 15). The actual test used for
14 and 16 years) on the Clinical Evaluation of Lan- each participant is listed in appendices A and B.
guage Fundamentals—4th Edition (CELF-4) (Semel While several participants had standard scores on
2006) and British Picture Vocabulary Scale—2nd Edi- Matrices < 85, most of them showed a discrepancy be-
tion (BPVS-2) (Dunn et al. 1997) to measure general tween their Matrices and CELF scores. Indeed an anal-
language progress during their time in the school. They ysis of variance (ANOVA) (with a Greenhouse–Geisser
are also assessed on the Test of Visual Perceptual Skills correction) comparing CELF Receptive and Expressive
(TVPS) (Gardner 1988) as part of their pre-admission Language, Matrices and TVPS found a significant differ-
assessment and at the same intervals as the language tests ence, F(1.3) = 45.2, p < 0.001, ηp2 = 0.80. Post-hoc
if they are on the occupational therapy caseload. For this Bonferroni corrected t-tests showed that the Matrices
study, all pupils in years 6–9 (aged 10–14 years) in the and TVPS standard scores did not differ significantly
academic year before the study intervention started were from each other, p = 0.08 and the CELF Receptive
tested on the TROG-2. Those with a standard score and Expressive Language did not differ from each other,
< 85, who also made at least three errors on the blocks p = 0.57, but both CELF measures differed significantly
of the TROG-2 testing comprehension of but not (Block from both the Matrices and TVPS, p < 0.001. There-
P), neither nor (Block O), and not only but also (Block fore, as a group their language scores were significantly
H) were included in the study. An additional pupil in lower than their non-verbal and visual processing scores.
the year above was added, as receptive grammar was felt The therapy method (Shape Coding) was already
to be a priority area for her and she matched the partic- being used in the school, but the participants’ knowledge
ipant criteria. A flow diagram of progress through the of the system varied. Because this could affect response
study is shown in figure 1. to the intervention, we asked each participant’s SLT to
Fourteen participants were identified; a summary rate the participant’s prior knowledge of Shape Coding
of their most recent scores on the standardized tests on a scale of 0–4.
at the beginning of the study is shown in table 1 and
their individual data in appendices A and B. Table 1
Study design
shows all the participants had severe RELI, including
severe difficulties with comprehension of grammar, as This study is a single-blind cross-over trial with random
measured by TROG-2. In addition to their RELI, some allocation to groups with an assessor blind to group
participants had difficulties or diagnoses which mean status. The 14 participants were randomly assigned to
34 Susan H. Ebbels et al.

Year 6-9 before

study started
49

CELF4 RLI <85 CELF4 RLI ≥ 85

39 10

TROG SS <85 TROG SS ≥ 85 TROG SS<85 TROG SS ≥ 85

19 20 3 7

poor ok poor ok
conjuncƟons conjuncƟons conjuncƟons conjuncƟons
13 6 0 3

randomised
13 +1 older pupil

Therapy WaiƟng
group controls
7 7

completed therapy did not complete completed therapy did not complete
as planned therapy as planned as planned therapy as planned
6 1 6 1

Figure 1. Flow of participants through the trial.

Table 1. Pre-therapy standard scores on standardized tests

Therapy group Waiting controls Overall

Boys:girls 4:3 6:1 10:4
Mean (SD) Range Mean (SD) Range Mean (SD)
Age 13;4 (1;6) 11;3–15;8 13;8 (1;7) 11;5–16;1 13;6 (1;6)
Test of visual perceptual skills 86 (12) 64–97 86 (21) 63–114 86 (16)
Matricesa 81 (9) 68–94 81 (18) 57–102 81 (14)
BPVS–II 66 (10) 47–78 64 (14) 47–85 65 (24)
TROG-2 57 (5) 55–69 61 (9) 55–78 59 (7)
CELF-4 Receptive language 53 (5) 45–58 53 (6) 45–61 53 (5)
CELF-4 Expressive language 48 (3) 45–53 53 (6) 45–59 50 (5)
Note: a The standard score was computed from the z-score to aid in the comparison with other tests.

two groups (Therapy versus Waiting Control) by the first age, but grammatical comprehension was not targeted.
author using the random number function in Excel to Therefore both groups received their normal amount of
sort the participants into a random sequence which was speech and language therapy at all times, thus avoiding
then split into two groups. The therapy group received a potential Hawthorne effect. The study was approved
the study therapy in the Spring Term (Phase 1) and by the Moor House School Ethics Forum.
the waiting controls in the Summer Term (Phase 2). Analyses of the pre-therapy scores for the two groups
All participants were assessed pre-therapy (in January), showed no differences between the two groups in gender
immediately after Phase 1 (in April) and immediately distribution, χ 2 (1) = 1.4, p = 0.24, age, t(12) = 0.4,
after Phase 2 (in July). All testing was carried out by an p = 0.67, d = 0.19, TROG-2 raw score, t(12) = 0.36,
SLT assistant blind to group status, but trained in using p = 0.28, d = 0.73, standard scores on the CELF-4
the tests. Expressive Language,1 U = 14.5, n1 = 7, n2 = 7, p =
When the participants were not receiving the study 0.21, CELF-4 Receptive Language, t(12) < 0.001, p =
therapy (Phase 1 for waiting controls (WC) and Phase 1.0, d = 0.0, TROG-2, U = 15.0, n1 = 7, n2 = 7,
2 for the original therapy group (T)), they continued p = 0.30, Matrices, t(12) = 0.02, p = 0.99, d = 0.01,
with their normal speech and language therapy pack- or TVPS, t(10) < 0.001, p = 1.0, d = 0.0 or prior
Improving comprehension in adolescents with severe receptive language impairments 35
knowledge of Shape Coding, U = 12.5, n1 = 7, n2 = r Active (e.g. the car hits the lorry).
7, p = 0.10. r Adjectival passive (e.g. the teddy is mended).
r Long passive (e.g. the girl is washed by the boy).
r Short passive (e.g. the fish is being eaten).
Measurements
Test of Reception of Grammar—2 (TROG-2) (Bishop We analysed only the proportion of passives cor-
2003) rect (adjectival, long and short combined) as the active
sentences were at ceiling pre-therapy.
All participants were tested on the TROG-2 (a picture-
pointing comprehension test) at all testing points (pre-
therapy, post-Phase 1 therapy and post-Phase 2 therapy). Therapy method
The stimuli are presented in blocks of four targeting the This study used the Shape Coding therapy approach
same construction. Three blocks target the conjunc- (Ebbels 2007) to teach the participants in an explicit
tions receiving intervention in this study. The number way the targeted coordinating conjunctions and their
of correct answers in each of these blocks was recorded meaning in a variety of sentence structures, using visual
separately and combined with the specific test described templates and cues.
below. We also recorded the raw score (in terms of num- We aimed for the participants’ usual SLT to deliver
ber of blocks passed; where all answers in the block were the therapy for 30 min, once a week for 8 weeks (4 h
correct) and standard score. in total), on an individual basis in their usual therapy
setting. Six different SLTs were involved, each seeing be-
Specific test of comprehension of conjunctions tween one and five participants. The SLT who worked
with each participant is shown in appendices A and B.
We constructed a specific picture-pointing test to assess The therapy programme was devised by the first au-
the participants’ comprehension of the targeted coordi- thor and consisted of a sequence of 14 steps, which
nating conjunctions further. The items were presented were provided to the SLTs (reproduced in appendix C).
in a random order, mixed in with the control items The SLTs also received a pack of materials (blank Shape
(passives; discussed below). The position on the page of Coding templates, target conjunctions with visual cues
the correct picture was also randomized. Each targeted underneath, black and white line drawings for colour-
conjunction (but not, neither nor and not only but also) ing activities and small animal objects). The first author
had four items, with different grammatical construc- was available to answer queries throughout the study.
tions (four items using and were also included but not The SLTs worked through the therapy sequence at the
analysed as they were at ceiling pre-therapy): participants’ own pace, and stopped after eight sessions,
regardless of the step reached.
r Noun phrase + conjunction + noun phrase + In the first session, the participants were introduced
adjective (e.g. the ladder and the snake are long). to the approach. The conjunctions were then intro-
r Noun phrase + conjunction + noun phrase + duced, one at a time, in a variety of sentence con-
verb (e.g. the horse but not the sheep is eating). structions. Each step built on the previous steps and
r Noun phrase + adjective + conjunction + adjec- contrasted new conjunctions with those previously in-
tive (e.g. the pencil is neither long nor blue). troduced. The SLTs did not move onto the next step
r Noun phrase + verb + conjunction + verb (e.g. until they judged the participant had understood the
the lady is not only standing, but also waving). current step, to ensure good comprehension of each
taught conjunction and a solid foundation for the next
step.
Specific test of control construction (passives)
Each conjunction was introduced with a spoken ex-
We tested the participants’ understanding of passive sen- planation of its meaning and a corresponding visual
tences to act as a within-subject control and to ascertain cue (see appendix C). Some SLTs also reinforced each
whether any improvement in the comprehension of con- question word with its Makaton sign and head nod-
junctions generalized to another area of grammar. We ding/shaking, but this was reduced during the therapy.
selected items from the Test of Active and Passive Sen- But not was introduced first, then neither nor, then
tences (van der Lely 1996b), which tests three different not only but also. Therapy activities consisted of the
forms of the passive and active constructions using a SLTs taking turns with the participants to generate and
picture pointing task. We tested four items for each act out sentences using the shape templates as an aid
construction to ensure similarity with the specific con- (these sentences did not match exactly those used in the
junctions test. The four constructions were (for a full specific test or the TROG-2). When the SLT judged
description of these, see van der Lely 1996a): that a participant’s understanding of a conjunction was
36 Susan H. Ebbels et al.
accurate, the templates were removed and the activities two phases of the study, we carried out t-tests compar-
repeated. ing the amount of individual, group and total speech
and language therapy received by the two groups in
each phase. None of these analyses showed any signifi-
Attendance and treatment fidelity
cant differences between the groups (Phase 1: individual
Attendance and treatment fidelity were assessed indi- therapy, t(11) = 0.5, p = 0.60, d = 0.34; group therapy,
rectly by the first author after each phase by sending t(11) = 0.7, p = 0.52, d = 0.42; total therapy t(11) =
questionnaires to the SLTs and checking their case notes. 0.8, p = 0.43, d = 0.53; Phase 2: individual therapy,
Two participants did not receive the full 4 h of therapy. t(11) = 2.1, p = 0.06, d = 1.42; group therapy, t(11) =
One (WC4) withdrew from the study during Phase 2 0.5, p = 0.65, d = 0.30; total therapy t(11) = 1.7, p =
(his therapy phase). The other (T2) attended eight ther- 0.12, d = 1.12). The difference between the two groups
apy sessions, but due to listening and attention diffi- almost reached significance and showed a large effect size
culties had shorter sessions than the others, so in total during Phase 2 for individual therapy. This is because
received 2 h and 40 min of therapy. His results were three of the study authors completed the study therapy
excluded from all analyses, but the results of WC4 were with the absent SLT’s caseload, but the other children
included during Phase 1, when he was acting as a con- on her caseload from the original therapy group did not
trol, but excluded from Phase 2 (his therapy phases). receive cover for their individual therapy. The group
Thus, the results analyse whether those who received therapy they received was unaffected.
the full amount of therapy made progress.
The therapy was usually provided by the partici-
pants’ own SLT as part of their normal therapy package Results
(see appendices A and B for which SLTs saw which par-
ticipants). The exceptions to this were three participants We analysed the study in two phases. Phase 1, which
(WC2, WC3 and WC7) whose SLT was absent at the takes the form of a typical RCT with a therapy group and
end of Phase 2, so their final two or three sessions were a control group, was analysed separately from Phase 2 to
carried out by one or two of the study authors (see enable comparison with other RCTs. In Phase 2, we then
appendix B). provided the same therapy to the waiting controls. The
Because the therapy programme was delivered at Phase 2 analyses consider whether the waiting controls
the participants’ own pace, not all participants com- made a similar amount of progress to the original therapy
pleted the programme and some were not introduced to group with therapy and whether the original therapy
all the targeted conjunctions. Six participants completed group maintained their progress. Complete data were
the whole therapy programme (T1, T6, T7, WC1, available for 12 participants, six in each group.
WC5, WC7), four completed but not and neither nor
(T4, WC2, WC3, WC6), one only completed but not
(T3), and two did not complete any conjunction (T2, Phase 1 results
T5), but T2 was excluded from the analyses (see above).
Targeted coordinating conjunctions
The study therapy was provided as part of the par-
ticipants’ normal speech and language therapy package, To increase power, we combined the results from the
which includes joint planning and teaching of English specific test with the responses to those blocks on the
lessons, at least two speech and language therapy groups TROG-2 which test the targeted conjunctions. This
per week and individual speech and language therapy gave a score out of eight for each targeted conjunction
sessions. The focus of the groups attended by the par- (four items from the relevant TROG-2 block and four
ticipants during the study period included: vocabulary, items from the specific test) and an overall score of 24.
narrative, reading comprehension, use of signing, social Table 2 shows mean scores on the targeted conjunc-
skills and life skills. Individual therapy while in the con- tions at each time point for the two groups (individual
trol phase of the study focused on: articulation, vocab- scores are shown in appendices A and B). Table 2 shows
ulary, past tense, phonological awareness, conversation that overall, from pre-therapy to post-Phase 1 therapy,
ability, signing, narrative and voice, but not compre- the therapy group improved, while the waiting con-
hension of grammar. Some participants also received trols got worse. A comparison of the change in targeted
other individual speech and language therapy during conjunctions over Phase 1 (post-Phase 1 therapy minus
their study therapy phase in addition to the study ther- pre-therapy) between the two groups using a two-tailed
apy; this focused on articulation, vocabulary, inferenc- t-test revealed a significant difference between the two
ing, narrative, idioms and use of signing. groups, t(11) = 2.7, p = 0.02, d = 1.64, where the
To test whether the amount of speech and language therapy group showed more progress than the waiting
therapy differed between the two groups in either of the controls.
Improving comprehension in adolescents with severe receptive language impairments 37
Table 2. Means (standard deviations) at each testing point

Scores at each testing point

Change in scores
Post-Phase 1 Post-Phase 2
Pre-therapya therapya therapyb Over Phase 1a Over Phase 2b
Therapy group Therapy period Maintenance
Conjunctions (/24) 10.3 (3.6) 13.7 (2.1) 14.2 (3.7) 3.3 (3.8) 0.5 (3.8)
TROG-2 Raw Score 6.3 (2.3) 8.5 (2.1) 9.2 (2.9) 2.2 (1.7) 0.7 (1.0)
TROG-2 Standard Score 57.3 (5.7) 62.2 (7.8) 65.2 (9.0) 4.8 (7.5) 3.0 (3.7)
Passives (/12) 7.8 (2.5) 8.7 (2.0) 8.0 (1.3) 0.8 (1.7) –0.7 (0.8)
Waiting controls Baseline Therapy period
Conjunctions (/24) 10.1 (4.1) 8.4 (4.9) 15.5 (6.7) −1.7 (2.9) 6.7 (3.1)
TROG-2 Raw Score 7.7 (3.0) 7.3 (2.3) 10.3 (4.2) −0.4 (2.6) 2.5 (2.6)
TROG-2 Standard Score 61.4 (9.0) 58.0 (5.5) 70.5 (14.3) −3.4 (4.5) 12.0 (10.7)
Passives (/12) 7.0 (3.7) 6.3 (2.9) 5.3 (3.3) −0.7 (1.6) −1.3 (0.8)
Notes: a Includes WC4.
b
Excludes WC4.

Table 3. Pearson’s r-values for correlations between progress and all other measures taken

Overall Overall
conjunctions TROG-2 conjunctions TROG-2 SS Prior knowledge
progress SS progress pre-therapy pre-therapy Matrices TVPS of Shape Coding
TROG-2 SS progress 0.51
Overall conjunctions pre-therapy −0.29 0.27
TROG-2 SS pre-therapy 0.34 0.17 0.35
Matrices 0.35 −0.36 −0.53 0.23
TVPS 0.20 −0.14 −0.11 0.47 0.89∗∗
Prior knowledge of Shape Codinga 0.48 0.24 −0.32 −0.12 −0.36 −0.32
Step reached with therapya 0.39 0.79∗∗ 0.07 0.11 0.07 0.09 0.13
a
Note: Spearman’s r.
p-values (two-tailed): p < 0.05∗ , p < 0.01∗∗ and p < 0.001∗∗∗ .

To test whether progress made by either group was not analyse the progress of the waiting controls as their
significantly better than zero, we analysed their change scores decreased.
in score over Phase 1. One-tailed,2 one-sample t-tests
showed that overall the therapy group made progress
which was significantly greater than zero, t(5) = 2.1, Control structure (passives)
p = 0.04, d = 0.86. We did not analyse the change for Table 2 also shows the scores out of 12 on the specific
the waiting controls as this was negative. passives test. This shows a small positive change in the
therapy group and a small negative change in the waiting
controls over Phase 1. A two-tailed t-test revealed no
significant difference between the progress of the two
Generalization of progress (TROG) groups over Phase 1, t(11) = 1.68, p = 0.12, d =
Table 2 also shows the mean scores at each testing point 1.01, despite a large effect size. One-tailed, one-sample
on the TROG-2 (Raw and Standard Scores). This shows t-tests showed that the progress made by the therapy
that over Phase 1, the therapy group improved on the group was not significantly greater than zero, t(5) =
TROG-2, while the waiting control group did not. A 1.2, p = 0.14, d = 0.49. The waiting controls’ scores
two-tailed t-test showed the progress of the two groups decreased.
over Phase 1 differed significantly on standard scores,
t(11) = 2.5, p = 0.03, d = 1.48. The difference on
Summary of Phase 1 results
raw score just failed to reach significance despite a large
effect size, t(11) = 2.1, p = 0.06, d = 1.24. The therapy group made significantly more progress
One-tailed, one-sample t-tests on the change in raw than the waiting controls on the targeted conjunctions
and standard scores showed that the therapy group made and TROG-2 standard score. The therapy group also
progress which was significantly greater than zero on made progress which was significantly greater than zero
the raw, t(5) = 3.1, p = 0.01, d = 1.26, but not the on the targeted conjunctions and TROG-2 raw score.
standard score, t(5) = 1.6, p = 0.09, d = 0.64. We did We found no obvious generalization to passives.
38 Susan H. Ebbels et al.
Phase 2 results ing Phase 2. Therefore, no statistical analyses of progress
were carried out.
Targeted coordinating conjunctions
Table 2 also shows the post-Phase 2 therapy scores and Summary of Phase 2 results
the change in scores over Phase 2. To analyse whether
The waiting controls also made progress on the targeted
the waiting controls made similar progress with therapy
conjunctions when they received therapy. The scores
to the original therapy group, we compared the changes
for the original therapy group remained stable; thus
during the specific period when each group received
progress was maintained. When the two groups were
therapy (over Phase 1 for the original therapy group and
combined, they showed significant progress over their
over Phase 2 for the waiting controls) and found no
therapy period.
significant difference, t(10) = 1.6, p = 0.13, d = 1.04.
On the TROG-2, the waiting controls made similar
A one-tailed t-test confirmed that the waiting controls
progress to the original therapy group, who maintained
also made progress with therapy which was significantly
their previous progress. When the groups were com-
greater than zero overall, t(5) = 5.2, p = 0.002, d = 2.1.
bined, both the raw and standard score of the TROG-2
To get an overall effect size for progress made by
showed significant progress with therapy.
all participants for just the period when enrolled in the
On the passives, the scores for both groups declined.
therapy (progress over Phase 1 for the original therapy
Thus, the waiting controls did not replicate the small,
group and over Phase 2 for the original waiting con-
non-significant progress made by the original therapy
trols), we combined the two groups and compared their
group in Phase 1 and the original therapy group did not
progress to zero and found a significant effect, t(11) =
appear to maintain their original progress.
4.6, p < 0.001, d = 1.33.
To analyse whether the original therapy group main-
Possible factors influencing progress
tained the progress they had made in Phase 1 during
Phase 2, we compared their change in score over Phase 2 To investigate the factors which could have influenced
to zero using a two-tailed t-test.3 This showed that their progress, we carried out two-tailed correlations between
scores remained stable, t(5) = 0.3, p = 0.76, d = 0.13. the progress scores with therapy (on overall conjunctions
and TROG-2 standard scores) and the other measures
Generalization of progress (TROG) taken. The correlations are shown in table 4. Progress
Table 2 also shows the post-Phase 2 therapy scores and was not significantly correlated with any test measure.
changes on the TROG-2 during Phase 2. To analyse We also considered variables more directly related to
whether the waiting controls made similar progress with the therapy itself: the SLT’s rating of the participants’
therapy to the original therapy group, we compared prior knowledge of Shape Coding and step of the pro-
change during the specific period when each group re- gramme reached by the end of therapy, using Spearman’s
ceived therapy and found no significant difference on r. We found a significant correlation between progress on
either raw, t(10) = 0.3, p = 0.80, d = 0.17, or stan- the TROG-2 SS and the step of therapy reached, but not
dard scores, t(10) = 1.3, p = 0.21, d = 0.85. One- between either measure of progress and the SLTs’ ratings
tailed t-tests confirmed that the waiting controls made of the participants’ prior knowledge of Shape Coding.
progress with therapy (over Phase 2) which was signif- The two non-language tests (Matrices and TVPS) were
icantly greater than zero on both the raw, t(5) = 2.4, significantly correlated with each other, but not with
p = 0.03, d = 0.97, and standard scores, t(5) = 2.8, any other measure.
p = 0.02, d = 1.13. When we combined the two groups We included participants with a wide range of dif-
(thus increasing power) and compared their progress ficulties in addition to their RELI. Thus, the majority
over the period when enrolled in the therapy to zero, we do not fit the strict exclusionary criteria for SLI. Taking
found a significant effect for both the raw, t(11) = 3.8, the strict criteria for SLI of non-verbal IQ needing to
p = 0.001, d = 1.11 and standard scores, t(11) = 3.1, be equal to or greater than 85 and excluding partici-
p = 0.005, d = 0.88. pants with any other diagnosis, only two participants
A two-tailed t-test showed that the change in score met these strict criteria (one of whom also has a diag-
over Phase 2 for the original therapy group was not nosis of dyslexia). A statistical comparison of the overall
significant for either the raw, t(5) = 1.6, p = 0.18, d = progress on the conjunctions made by these participants
0.65, or standard scores, t(5) = 2.0, p = 0.10, d = 0.82, and the others was not possible due to small numbers,
i.e. they maintained their progress. however their progress on targeted conjunctions with
therapy (7 and 5) was very similar to the rest of the par-
ticipants (mean = 5). If the non-verbal exclusionary cri-
Control structure (passives)
terion is relaxed to 70, five participants meet the criteria
Table 2 shows change in scores on passives over Phase 2. for SLI. The progress of this group was not significantly
This shows that the scores declined for both groups dur- different from those not meeting criteria, t(10) = 0.45,
Improving comprehension in adolescents with severe receptive language impairments 39
p = 0.66, d = 0.31. Four participants had diagnosed are not key indicators of the ability to make progress with
medical syndromes, but did not differ from the others this kind of therapy, even though the therapy involves
in their progress, t(10) = 0.31, p = 0.76, d = 0.22 and meta-linguistic explanations and uses visual cues.
neither did the three with hearing impairments, t(10) = Analyses of whether participants meeting certain di-
0.88, p = 0.40, d = 0.64, although the numbers are too agnostic criteria (e.g. those meeting strict criteria for
small to draw any strong conclusions. SLI) differed from the others in their response to in-
tervention revealed no significant differences. Therefore
Discussion we cannot, from our data, make any predictions regard-
ing which children make the most progress with this
Four hours of explicit therapy with visual support, fo-
therapy.
cused on comprehension of coordinating conjunctions,
The significant correlation between progress on the
one-to-one with an SLT significantly improved compre-
TROG-2 and the step of therapy reached indicates that
hension of the targets among adolescents with severe
if the participants who did not complete the therapy in
RELI. The waiting controls made little progress dur-
the eight weeks had been able to continue and finish
ing their baseline period, even getting slightly worse
the programme, they might have made more progress.
(Phase 1) but made progress when they too received
Thus, a future project could measure the effectiveness
the therapy (after Phase 2). At the end of Phase 1, af-
of completion of the therapy programme, regardless of
ter only one group had received the study therapy, the
the time taken, rather than the model used here of a
progress of the two groups differed significantly (d =
restricted amount of time dictating when the therapy
1.64). Progress could not have been due to a placebo or
should stop.
Hawthorne effect as both groups received their normal
therapy package with their usual SLT at all times, but
during their study therapy phase, this included therapy Implications for theories of SLI
targeted at comprehension of coordinating conjunctions
The therapy approach in this study used explicit teach-
for 30 min per week. Given this and the fact that the
ing with visual cues. This could be argued to increase the
participants were randomly assigned to the two groups,
processing demands on the participants. If the main rea-
it is most likely that the content of the therapy was re-
son for failing comprehension tasks was processing lim-
sponsible for progress, rather than any other factors. The
itations, then it is unlikely that this therapy approach
effect size for progress with therapy of the two groups
would improve their performance on these tests. The
combined (relative to zero change) was d = 1.33.
success of the therapy approach therefore makes it less
The original therapy group maintained their
likely that impaired processing was the cause of their
progress for 4 months after their therapy had ceased.
difficulties.
Progress generalized to the TROG-2 raw and standard
More likely, however, is that the therapy worked by
scores such that on average, over the whole study, the
allowing the participants to use their relative strengths
participants closed the gap with their typically devel-
to compensate for their weaknesses. Within the frame-
oping peers by eight standard score points. This could
work of the PDH, it could be that the explicit therapy
be partly due to progress on the blocks containing the
enabled them to use their better declarative memory
targeted conjunctions, however, the progress on these
system to compensate for their more impaired procedu-
blocks was often insufficient for a block to be passed
ral memory system. The visual support could also have
(e.g. progress from one to three out of four in a block
enabled them to enlist their better visuo-spatial skills to
would not be registered in the TROG-2 raw score).
compensate for their weaker verbal working memory.
Also, several participants made progress on blocks other
However, this study was not specifically designed to ex-
than those containing the targeted conjunctions. How-
amine the underlying processes, thus these hypotheses
ever, the data do not indicate that the effects of therapy
remain speculative.
generalized to comprehension of passives.

Factors influencing progress Implications for clinical practice

The analyses showed that progress was not significantly Clinicians are under increasing pressure to base their
correlated with any of the standardized tests, although intervention on evidence. Unfortunately, we have little
several of the effects were medium (r > 0.3) or large or weak evidence for many of the approaches which we
(r > 0.5). Performance on a Matrices test and the TVPS, commonly use (particularly for receptive language). The
while significantly correlated with each other, were not publication of recent UK government-funded research
significantly related to any other measure, including (Law et al. 2012) investigating the current evidence base
progress measures. Thus, it seems that visual perceptual is very welcome, as are initiatives to make the evidence
skills and performance on a non-verbal test (Matrices) more easily accessible to clinicians, such as the What
40 Susan H. Ebbels et al.
Works website being set up by the Communication The precise aspects of the therapy which are respon-
Trust.4 sible for progress could also be investigated, as could
This study provides evidence of effectiveness of the amount of therapy which is needed to make signifi-
speech and language therapy for older children with cant or optimal gains. Future work could investigate the
RELI, as, on average, our participants made progress effects of completing the whole therapy protocol regard-
with targeted therapy which was maintained. Combined less of the length of time taken to do so and also take
with the findings of other studies, it seems that the Shape other measures of pre-therapy abilities (e.g. executive
Coding therapy approach can be effective for improv- functions) to try to find reliable predictors of progress.
ing a range of areas of grammatical comprehension and This study focused on a small area of comprehen-
production in older children with RELI. Its effectiveness sion. While other less rigorous studies indicate the effec-
for younger children or children with other profiles of tiveness of the Shape Coding approach for some other
difficulties remains to be established. areas of comprehension (passives, ‘wh’ questions and da-
We were not able to find factors which predicted tives, Ebbels and van der Lely 2001, Ebbels 2007), the
which participants would make the most progress, as effectiveness of this therapy approach for a wider range
diagnostic criteria seemed to make no reliable difference of areas of comprehension remains to be established. In-
and pre-therapy language levels (within the small range deed, studies of its effectiveness for expressive language
in our study), visual perceptual skills and non-verbal are also limited. An RCT has shown it is effective for
reasoning abilities as measured on Matrices also seemed production of verb argument structure, but only smaller
to have little effect. It is difficult to draw strong conclu- less robust studies have investigated its effectiveness for
sions given the small size of this study, but it seems that production of past tense (Ebbels 2007, Kulkarni et al.
based on our evidence, this approach could be tried with 2013), passives and ‘wh’ questions (Ebbels and van der
any adolescent with a severe receptive language impair- Lely 2001). Thus, more studies are required to establish
ment affecting comprehension of grammar. However, whether this approach can be effective for a wider range
we do not know whether young people, like partici- of targets, with a wider range of children (particularly in
pant T2, with listening and attention difficulties will terms of age) and a wider range of adults delivering the
benefit when given the full amount of therapy (albeit therapy in a variety of settings.
in smaller chunks) or whether such children would be
able to complete the full therapy programme and make
Conclusions
progress given sufficient time.
Adolescents with severe RELI receiving 4 h of individual
Limitations and future directions therapy with an SLT focused on the comprehension
of specific coordinating conjunctions made significant
The RCT presented here was small in scale and based progress. Waiting controls made no progress until they
in one school. This may have affected the results in also received the therapy when they made similar gains.
several ways and we need to be careful not to either Progress was maintained for 4 months.
under- or over-interpret the data. The small sample size This study contributes further to existing evidence
means that we had limited power in our analyses. Some that speech and language therapy for adolescents with
of our non-significant results showed large effect sizes. language impairments can be effective. Thus, this age
Thus, at times the data appear to indicate a difference group should not be neglected in the provision of ther-
between groups, or that progress differed from zero, or apy services. It also provides the first strong evidence,
that progress correlated with possible predictors, but using a (small-scale) RCT, that speech and language
the sample size was too small to evaluate whether these therapy for school-aged children with receptive language
were statistically significant. Thus, larger scale trials with impairments can improve their comprehension of tar-
more power are required to investigate these results fur- geted structures.
ther as we may be under-interpreting our findings.
However, our small sample size and the fact that Acknowledgements
the participants all came from the same school means
We would like to thank Sarah Bayliss for carrying out the ‘blind’
that it cannot be assumed that our results will general- assessment and the following Speech and Language Therapists for
ize to other groups of young people with language im- helping carry out the therapy: Mary Jennings, Niamh Elleson, Sarah
pairments. Future work could aim to establish whether O’Sullivan and Lisa Morel. This study was funded by Moor House
similar results to our study are obtained for the same School where all authors were employed at the time of the study.
therapy method but with different age groups. Different Declaration of interest: The authors report no major conflicts of
interest. However, Susan Ebbels runs courses on Shape Coding and
methods of delivery could also be investigated, includ- Moor House School sells CDs of Shape Coding resources. The
ing whether this therapy method can be delivered by authors alone are responsible for the content and writing of the
non-SLTs or to pairs or groups of children. paper.
Improving comprehension in adolescents with severe receptive language impairments 41
Appendix A. Individual data for participants in the original therapy group (receiving therapy in Phase 1)

Participant code T1 T2 T3 T4 T5 T6 T7
Gender Male Male Male Female Female Male Female
Age at the start of the study 14;11 11;3 12;11 12;2 13;1 13;4 15;8
Medical diagnoses/exclusionary criteriaa NV < 85 ASD, NV < 70 none synd, NV < 85 NV < 85 synd NV < 85
Matrices test used WRIT WASI WASI KBIT WISC WISC WRIT
SLT(s) delivering therapy A L M E M S N
Pre-Phase 1 Test of Visual 95 86 97 n.a. 64 90 85
therapy Perceptual Skills SS
Matrices SS 81 69 94 84 70 88 83
CELF-4 UK receptive 45 58 58 58 50 55 50
language SS
CELF-4 UK expressive 47 47 53 49 47 45 47
language SS
BPVS SS 47 73 78 70 61 64 68
TROG-2 RS 6 3 10 5 4 5 8
TROG-2 SS 55 55 69 55 55 55 55
Prior knowledge of 1 1 1 2 1 1 1
Shape Coding
Targeted conjunctions 4 10 11 11 14 13 9
(/24)
Passives (/12) 4 5 9 9 8 6 11
Post-Phase 1 TROG-2 RS 9 4 10 7 5 10 10
therapy TROG-2 SS 58 55 69 55 55 74 62
Targeted conjunctions 14 9 16 14 13 15 10
(/24)
Passives (/12) 7 7 7 10 8 8 12
Post-Phase 2 TROG-2 RS 10 2 11 6 5 11 12
therapy TROG-2 SS 62 55 74 55 55 74 71
Targeted conjunctions 8 8 16 17 18 14 12
(/24)
Passives (/12) 7 3 7 9 7 8 10
Step of the therapy programme reached 13 3 7 11 4 14 14
Note: a Exlusionary criteria: synd = diagnosed medical syndrome; and NV = non-verbal skills as measured on Matrices.
42 Susan H. Ebbels et al.
Appendix B. Individual data for participants in original waiting control group (receiving therapy in Phase 2)

Participant code WC1 WC2 WC3 WC4 WC5 WC6 WC7

Gender Male Male Male Male Female Male Male
Age at the start of the study 14;8 13;2 13;6 16;1 14;6 11;5 12;5
Medical diagnoses/exclusionary criteriaa HI synd, HI, epi, CI epi, LKS, synd, ADHD, (dyslexia)
NV<70 NV < 85 NV < 70 NV < 70
Matrices test used WRIT WRIT WRIT KBIT WRIT WASI WRIT
SLT(s) delivering therapy A M+N+S M+A A A L M+S
Pre-Phase 1 Test of Visual n.a. 73 114 90 71 63 106
therapy Perceptual Skills SS
Matrices SS 102 66 99 80 57 68.5 97
CELF-4 UK receptive 58 58 50 45 50 52 61
language SS
CELF-4 UK expressive 55 52 45 45 53 59 59
language SS
BPVS SS 55 62 55 47 62 80 85
TROG-2 RS 11 9 4 6 5 7 12
TROG-2 SS 67 65 55 55 55 55 78
Prior knowledge of 4 2 2 0 3 2 1
Shape Coding
Targeted conjunctions 7 7 11 6 11 11 18
(/24)
Passives (/12) 8 3 2 5 10 10 11
Post-Phase 1 TROG-2 RS 10 5 7 4 8 7 10
therapy TROG-2 SS 62 55 55 55 55 55 69
Targeted conjunctions 7 2 5 6 13 10 16
(/24)
Passives (/12) 6 5 2 4 9 10 8
Post-Phase 2 TROG-2 RS 16 5 6 n.a. 12 10 13
therapy TROG-2 SS 90 55 55 n.a. 71 69 83
Targeted conjunctions 19 7 8 n.a. 21 15 23
(/24)
Passives (/12) 5 2 1 0 8 9 7
Step of the therapy programme reached 14 12 9 7 14 10 14
Note: a Exlusionary criteria: synd = diagnosed medical syndrome; HI = hearing impairment; epi = epilepsy; CI = cochlear implant; LKS = Landau–Kleffner Syndrome; and NV =
non-verbal skills as measured on Matrices.
Improving comprehension in adolescents with severe receptive language impairments 43
Appendix C: Therapy programme with template showing coordination of VPs, e.g.,
1. Introduction to Shape Coding The cow is jumping (over the fence) and running (round the field)

Introduce the following (but only for those pupils (b) Relate the coordinated VP to the question word
who do not already know them) “What doing” – discuss how ‘and’ means the subject
is doing both verbs (or Verb Phrases).
(c) Take turns to produce sentences using ‘and’ while
(a) ‘Who’ subject the other one acts out the sentence
(b) aux ‘is’ and ‘are’ (but gloss over the plural/singular (d) Introduce template with ‘but not’
The cow is jumping (over the fence) but not running (round the field)
distinction)
(e) Relate coordinated VP to the question word “What
(c) Verb phrase (only use single verbs here)
doing” – discuss how ‘but not’ means subject does
only the first, not the second verb phrase (shown by
(d) Adjective phrase the cross).
The last shape (Adj P) could be introduced just (f ) Take turns to produce sentences using ‘but not’
before step 4 or here (before step 2), at the discretion of while the other one acts out the sentence
the SLT. (g) take turns to create a sentence matching one of the
two templates (‘and’ or ‘but not’) and the other one
2. ‘And’ vs ‘but not’ (Subject NP + Verb) acts out the sentence
(h) when accurate, remove templates, bring back to
(a) Introduce the templates showing coordi- check responses
nation of NPs in subject position, e.g., (i) take turns to create sentences matching one of four
templates from sections 2 and 3, other acts out
The cow and the dog are jumping sentence
(j) when accurate, remove templates, bring back to
(b) Relate the coordinated subject to the question word check responses
“Who” – discuss how ‘and’ means both NPs are
carrying out the action. (k) Make combinations of the templates using coordi-
nated subjects and/or verbs e.g.,
(c) Take turns to produce sentences using ‘and’ while
the other one acts out the sentence r the cow and the cat are jumping but not running
(d) Introduce template with ‘but not’ r the cow but not the cat is lying down and sliding
r the cow and the cat are standing and jumping
The cow but not the dog is jumping r the cow but not the cat is lying down but not
sliding
(e) Relate the coordinated subject to the question word
“Who” – discuss how ‘but not’ means only the first,
(l) take turns to create sentences matching these com-
not the second NP (shown by the cross) is carrying
bined templates, other one acts out sentence
out the action.
(m) when accurate, remove templates, bring back to
(f ) Take turns to produce sentences using ‘but not’
check responses
while the other one acts out the sentence
(g) take turns to create a sentence matching one of the
two templates (‘and’ or ‘but not’) and the other acts 4. ‘And’ vs ‘but not’ (Subject NP + Adjective)
out, using template as a guide
(h) when accurate, remove templates, bring back to (a) Revise the templates showing coordination of
check responses NPs in subject position with verb. Show sim-
ilarity with template showing coordination of
3. ‘And’ vs ‘but not’ (Verb Phrase) NPs in subject position with adjectives, e.g.,

(a) Revise the templates showing coordination of NPs The ball and the hat are red
in subject position with verb. Show similarity
44 Susan H. Ebbels et al.
(b) Relate the coordinated subject to the question word (i) take turns to create sentences matching one of
“Who” – discuss how ‘and’ means both NPs have four templates from sections 4 and 5, other
the feature of the adjective. draws/colours in
(c) Take turns to produce sentences using ‘and’ while (j) when accurate, remove templates, bring back to
the other one draws/colours in the sentence check responses
(d) Introduce template with ‘but not’ (k) Make combinations of the templates using coordi-
nated subjects and/or adjectives e.g.,
The ball but not the hat is red r the hat and the ball are big but not blue
r the hat but not the ball is yellow and stripy
r the hat and the ball are small and black
(e) Relate the coordinated subject to the question word r the hat but not the ball is red but not spotty
“Who” – discuss how ‘but not’ means only the first,
not the second NP (shown by the cross) has the
feature of the adjective. (l) take turns to create sentences matching these com-
bined templates, other one draws/colours in
(f ) Take turns to produce sentences using ‘but not’
while the other one draws/colours in the sentence (m) when accurate, remove templates, bring back to
check responses
(g) take turns to create a sentence matching one of the
two templates (‘and’ or ‘but not’) and the other 6. Neither nor (Subject NP + Verb)
draws/colours in, using template as a guide
(h) when accurate, remove templates, bring back to (a) Revise the templates showing coordination of
check responses NPs in subject position with VP. Use ‘neither
nor’ as coordinator and discuss how ‘neither
nor’ means that not the first and not the sec-
5. ‘And’ vs ‘but not’ (Adjective Phrase)
ond NP are doing the action (shown by crosses)

(a) Revise the templates showing coordination of NPs Neither the cow nor the cat is jumping (around)
in subject position with adjective. Show similarity
with template showing coordination of APs, e.g., (b) Take turns to produce sentences using ‘neither nor’
while the other one acts out the sentence
The cow is big and black
(c) take turns to create a sentence matching one of the
three templates (‘neither nor’, ‘and’ or ‘but not’) and
(b) Relate the coordinated AP to the question word the other acts out, using template as a guide
“What like” – discuss how ‘and’ means the subject (d) when accurate, remove templates, bring back to
has the feature of both adjectives. check responses
(c) Take turns to produce sentences using ‘and’ while
the other one draws/colours in the sentence 7. Neither nor (VP)
(d) Introduce template with ‘but not’
(a) Introduce template with ‘neither nor’
The cow is big but not black
The cow is neither jumping nor running

(e) Relate coordinated AP to the question word “What (b) Relate coordinated VP to the question word “What
like” – discuss how ‘but not’ means subject only has doing” – discuss how ‘neither nor’ means subject
features of the first, not the second adjective (shown does not do the first, and not the second verb (shown
by the cross). by the crosses).
(f ) Take turns to produce sentences using ‘but not’ (c) Take turns to produce sentences using ‘neither nor’
while the other one one acts out the sentence while the other one acts out the sentence
(g) take turns to create a sentence matching one of the (d) take turns to create a sentence matching one of the
two templates (‘and’ or ‘but not’) and the other one three templates (‘neither nor’, ‘and’ or ‘but not’) and
draws/colours in the other one acts out the sentence
(h) when accurate, remove templates, bring back to (e) when accurate, remove templates, bring back to
check responses check responses
Improving comprehension in adolescents with severe receptive language impairments 45
(f ) take turns to create sentences matching one of six (f ) when accurate, remove templates, bring back to
templates from sections 2, 3, 6 and 7 other acts out check responses
sentence (g) Make combinations of the templates using coordi-
(g) when accurate, remove templates, bring back to nated subjects and/or adjectives e.g.,
check responses r Neither the hat nor the ball is big and blue
(h) Make combinations of the templates using coordi- r the hat but not the ball is neither yellow nor stripy
nated subjects and/or verbs e.g., r the hat and the ball are neither small nor black
r Neither the cow nor the cat is jumping and run- (h) take turns to create sentences matching these com-
ning bined templates, other one draws/colours in
r the cow and the cat are neither standing nor jump-
(i) when accurate, remove templates, bring back to
ing
r the cow but not the cat is neither lying down nor check responses
sliding
10. Not only, but also (Subject NP + Verb)
(i) take turns to create sentences matching these com-
bined templates, other one acts out sentence (a) Revise the templates showing coordination of NPs
in subject position with VP. Use ‘not only, but
(j) when accurate, remove templates, bring back to also’ as coordinator and discuss how ‘not only,
check responses but also’ means that both the first and the second
NP are doing the action (no crosses, like ‘and’)
8. Neither nor (Subject NP + Adjective)
Not only the cow but also the cat is jumping
(a) Introduce template with ‘neither nor’

Neither the cow nor the cat is black (b) Take turns to produce sentences using ‘not only, but
also’ while the other one acts out the sentence
(c) take turns to create a sentence matching one of
(b) Take turns to produce sentences using ‘neither nor’ the four templates (‘not only, but also’, ‘and’, ‘but
while the other one colours in/draws not’, or ‘neither nor’) and the other acts out, using
(c) take turns to create a sentence matching one of the template as a guide
three templates (‘neither nor’, ‘and’ or ‘but not’) and (d) when accurate, remove templates, bring back to
the other one colours in/draws check responses
(d) when accurate, remove templates, bring back to
check responses 11. Not only, but also (VP)

9. Neither nor (Adjective Phrase) (a) Introduce template with ‘not only, but also’
The cow is not only jumping but also running
(a) Introduce template with ‘neither nor’
(b) Relate coordinated VP to the question word “What
The cow is neither big nor black doing” – discuss how ‘not only, but also’ means sub-
ject does both the first and the second verb (shown
by no crosses).
(b) Take turns to produce sentences using ‘neither nor’ (c) Take turns to produce sentences using ‘not only, but
while the other one acts out the sentence also’ while the other one acts out the sentence
(c) take turns to create a sentence matching one of the (d) take turns to create a sentence matching one of the
three templates (‘neither nor’, ‘and’ or ‘but not’) and four templates (‘not only, but also’, ‘and’, ‘but not’,
the other one draws/colours in ‘neither nor’) and the other one acts out the sentence
(d) when accurate, remove templates, bring back to (e) when accurate, remove templates, bring back to
check responses check responses
(e) take turns to create sentences matching one of (f ) take turns to create sentences matching one of eight
six templates from sections 4, 5, and 8, other templates from sections 2, 3, 6, 7, 10 & 11, other
draws/colours in acts out sentence
46 Susan H. Ebbels et al.
(g) when accurate, remove templates, bring back to r Not only the hat but also the ball is neither big
check responses nor blue
(h) Make combinations of the templates using coordi-
r the hat but not the ball is not only yellow but also
nated subjects and/or verbs e.g., stripy
r the hat and the ball are not only small but also
r Not only the cow but also the cat is jumping and black
running
r the cow and the cat are not only standing but also (h) take turns to create sentences matching these com-
jumping bined templates, other one draws/colours in
r the cow but not the cat is not only lying down but
(i) when accurate, remove templates, bring back to
also sliding check responses
(i) take turns to create sentences matching these com-
bined templates, other one acts out sentence 14. Everything together
(j) when accurate, remove templates, bring back to
check responses (a) take turns to create sentences using any of the coor-
dinators in any of the positions introduced in any
12. Not only, but also (Subject NP + Adjective) combination, other one act out
(b) use templates to check any disagreements
(a) Introduce template with ‘not only, but also’

Not only the cow but also the cat is black

Notes
1. Mann–Whitney tests were used for non-normally distributed
data: CELF Expressive Language Scale and TROG Standard
(b) Take turns to produce sentences using ‘not only, but Score were negatively skewed for the therapy group and prior
also’ while the other one colours in/draws knowledge of Shape Coding resulted in six of seven participants
in the therapy group receiving the same rating.
(c) take turns to create a sentence matching one of the 2. A one-tailed test was used to analyse progress as we were only
four templates (‘not only, but also’, ‘and’, ‘but not’, interested in change in one direction. Where change was negative,
‘neither nor’) and the other one colours in/draws statistical tests were not carried out.
3. A two-tailed t-test was used as we had no particular predic-
(d) when accurate, remove templates, bring back to tion about the direction of any change during the maintenance
check responses period.
4. See http://www.thecommunicationtrust.org.uk/schools/what-
13. Not only, but also (Adjective Phrase) works-database.aspx/.

(a) Introduce template with ‘not only, but also’ References

BISHOP, D. V. M., 2003, The Test for Reception of Grammar. TROG
The cow is not only big but also black 2 (London: Psychological Corporation).
BISHOP, D. V. M., ADAMS, C. V. and ROSEN, S., 2006, Resistance
of grammatical impairment to computerized comprehension
(b) Take turns to produce sentences using ‘not only, but training in children specific and non-specific language impair-
also’ while the other one draws/colours in ments. International Journal of Language and Communication
Disorders, 41(1), 19–40.
(c) take turns to create a sentence matching one of the BOYLE, J. M., MCCARTNEY, E., O’HARE, A. and FORBES, J., 2009,
four templates (‘not only, but also’, ‘and’, ‘but not’, Direct versus indirect and individual versus group modes of
‘neither nor’) and the other one acts out the sentence language therapy for children with primary language impair-
ment: principal outcomes from a randomized controlled trial
(d) when accurate, remove templates, bring back to and economic evaluation. International Journal of Language
check responses and Communication Disorders, 44(6), 826–846.
(e) take turns to create sentences matching one of six BOYLE, J., MCCARTNEY, E., O’HARE, A. and LAW, J., 2010, Interven-
templates from sections 4, 5, and 8, 9, 12 & 13 tion for mixed receptive–expressive language impairment: a
review. Developmental Medicine and Child Neurology, 52(11),
other draws/colours in 994–999.
(f ) when accurate, remove templates, bring back to CLARK, A., O’HARE, A., WATSON, J., COHEN, W., COWIE, H., EL-
check responses TON, R., NASIR, J. and SECKL, J., 2007, Severe receptive
language disorder in childhood—familial aspects and long-
(g) Make combinations of the templates using coordi- term outcomes: results from a Scottish study. Archives of Dis-
nated subjects and/or adjectives e.g., ease in Childhood, 92(7), 614–619.
Improving comprehension in adolescents with severe receptive language impairments 47
COHEN, W., HODSON, A., O’HARE, A., BOYLE, J., DURRANI, T., LAW, J., BOYLE, J., HARRIS, F., HARKNESS, A. and NYE, C., 1998,
MCCARTNEY, E., MATTEY, M., NAFTALIN, L. and WATSON, Child health surveillance: screening for speech and language
J., 2005, Effects of computer-based intervention using acous- delay: a systematic review of the literature. Health Technology
tically modified speech (Fast ForWord-Language) in severe Assessment, 2, 1–184.
mixed receptive–expressive language impairment: outcomes LAW, J., GARRETT, Z. and NYE, C., 2004, The efficacy of treat-
from a randomized control trial. Journal of Speech, Language ment for children with developmental speech and language
and Hearing Research, 48(3), 715–729. delay/disorder: a meta-analysis. Journal of Speech, Language
CONTI-RAMSDEN, G., 2008, Heterogeneity of specific language im- and Hearing Research, 47(4), 924–943.
pairment (SLI): outcomes in adolescence. In C. F. Norbury, LAW, J., GARRETT, Z. and NYE, N., 2003, Speech and language
B. Tomblin and D. V. M. Bishop (eds), Understanding De- therapy interventions for children with primary speech and
velopmental Language Disorders (Hove: Psychology Press), pp. language delay or disorder (Cochrane Review). Cochrane
117–130. Database of Systematic Reviews, Issue 3, Art. No: CD004110.
DOCKRELL, J. E., LINDSAY, G., LETCHFORD, B. and MACKIE, C., DOI: 10.1002/14651858.
2006, Educational provision for children with specific speech LAW, J., LEE, W., ROULSTONE, S., WREN, Y., ZENG, B. and LINDSAY,
and language difficulties: perspectives of speech and language G., 2012, What Works: Interventions for Children with Speech,
therapist managers. International Journal of Language and Language and Communication Needs (Nottingham: Depart-
Communication Disorders, 41, 423–440. ment for Education (DfE)).
DUNN, L. M., DUNN, L. M., WHETTON, C. and BURLEY, J., 1997, LAW, J., RUSH, R., SCHOON, I. and PARSONS, S., 2009, Modeling
The British Picture Vocabulary Scale, Second Edition. BPVS II developmental language difficulties from school entry into
(Windsor: NFER-Nelson). adulthood: literacy, mental health, and employment out-
EBBELS, S. H., 2007, Teaching grammar to school-aged children comes. Journal of Speech, Language and Hearing Research,
with specific language impairment using Shape Coding. Child 52(6), 1401–1416.
Language Teaching and Therapy, 23(1), 67–93. LUM, J., CONTI-RAMSDEN, G., PAGE, D. and ULLMAN, M., 2012,
EBBELS, S. H., NICOLL, H., CLARK, B., EACHUS, B., GALLAGHER, Working, declarative and procedural memory in specific lan-
A. L., HORNIMAN, K., JENNINGS, M., MCEVOY, K., NIMMO, guage impairment. Cortex, 48, 1091–1250.
L. and TURNER, G., 2012, Effectiveness of semantic therapy MCCARTNEY, E., BOYLE, J., BANNATYNE, S., JESSIMAN, E.,
for word-finding difficulties in pupils with persistent lan- CAMPBELL, C., KELSEY, C., SMITH, J. and O’HARE, A.,
guage impairments: a randomized control trial. International 2004, Becoming a manual occupation? The construc-
Journal of Language and Communication Disorders, 47(1), tion of a therapy manual for use with language im-
35–51. paired children in mainstream primary schools. Interna-
EBBELS, S. and VAN DER LELY, H., 2001, Meta-syntactic therapy tional Journal of Language and Communication Disorders, 39,
using visual coding for children with severe persistent SLI. 135–148.
International Journal of Language and Communication Disor- MCCARTNEY, E., BOYLE, J., ELLIS, S., BANNATYNE, S. and TURNBULL,
ders, 36(Suppl.), 345–350. M., 2011, Indirect language therapy for children with per-
EBBELS, S. H., VAN DER LELY, H. K. J. and DOCKRELL, J. E., 2007, sistent language impairment in mainstream primary schools:
Intervention for verb argument structure in children with outcomes from a cohort intervention. International Journal of
persistent SLI: a randomized control trial. Journal of Speech, Language and Communication Disorders, 46(1), 74–82.
Language and Hearing Research, 50, 1330–1349. PARSONS, S., LAW, J. and GASCOIGNE, M., 2005, Teaching receptive
FEY, M. E., FINESTACK, L. H., GAJEWSKI, B. J., POPESCU, M. vocabulary to children with specific language impairment:
and LEWINE, J. D., 2010, A preliminary evaluation of Fast a curriculum-based approach. Child Language Teaching and
ForWord-Language as an adjuvant treatment in language in- Therapy, 21(1), 39–59.
tervention. Journal of Speech, Language and Hearing Research, Scientific Learning Corporation, 1998, Fast ForWord-Language
53(2), 430–449. (Berkley, CA: Scientific Learning Corporation).
GARDNER, M. F., 1988, Test of Visual–Perceptual Skills (Non-Motor) SEMEL, E., WIIG, E. H. and SECORD, W. A., 2006, Clinical Evaluation
(Belford: Ann Arbor Publ.). of Language Fundamentals: Fourth edition: CELF 4 (London:
GILLAM, R. B., LOEB, D. F., HOFFMAN, L. M., BOHMAN, T., Psychological Corporation).
CHAMPLIN, C. A., THIBODEAU, L., WIDEN, J., BRANDEL, TALLAL, P., MILLER, S. L., BEDI, G., BYMA, G., WANG, X. Q.,
J. and FRIEL-PATTI, S., 2008, The efficacy of Fast For- NAGARAJAN, S. S., SCHREINER, C., JENKINS, W. M. and
Word Language intervention in school-age children with MERZENICH, M. M., 1996, Language comprehension in
language impairment: a Randomized controlled trial. Jour- language-learning impaired children improved with acous-
nal of Speech, Language and Hearing Research, 51(1), 97– tically modified speech. Science, 271(5245), 81–84.
119. THRONEBURG, R. N., CALVERT, L. K., STURM, J. J., PARAMBOUKAS,
GLUTTING, J., ADAMS, W. and SHESLOW, D., 2000, Wide Range A. A. and PAUL, P. J., 2000, A comparison of service delivery
Intelligence Test (Wilmington, DE: Wide Range). models: effects on curricular vocabulary skills in the school
JOFFE, V., CAIN, K. and MARIC, N., 2007, Comprehension problems setting. American Journal of Speech–Language Pathology, 9,
in children with specific language impairment: does mental 10–20.
imagery training help? International Journal of Language and TOMBLIN, J. B., RECORDS, N. L., BUCKWALTER, P., ZHANG, X.,
Communication Disorders, 42(6), 648–664. SMITH, E. and O’BRIEN, M., 1997, Prevalence of spe-
KAUFMAN, A. S. and KAUFMAN, N. L., 2004, Kaufman Brief Intelli- cific language impairment in kindergarten children. Jour-
gence Test, Second Edition (Bloomington, IN: Pearson). nal of Speech, Language and Hearing Research, 40, 1245–
KULKARNI, A., PRING, T. and EBBELS, S., forthcoming 2013, Eval- 1260.
uating the effectiveness of Shape Coding therapy to develop ULLMAN, M. T. and PIERPONT, E. I., 2005, Specific language im-
the use of regular past tense morphemes in two children with pairment is not specific to language: the procedural deficit
language impairments. hypothesis. Cortex, 41(3), 399–433.
48 Susan H. Ebbels et al.
VAN DER LELY, H. K. J., 1996a, Specifically language impaired and WECHSLER, D., 2004, Wechsler Intelligence Scale for Children—Fourth
normally developing children: verbal passive vs adjectival pas- Edition (London: Pearson Assessment).
sive sentence interpretation. Lingua, 98(4), 243–272. WRIGHT, S. H., GORRIE, B., HAYNES, C. and SHIPMAN,
VAN DER LELY, H. K. J., 1996b, Test of Active and Passive Sentences A., 1993, What’s in a name? Comparative therapy for
(TAPS) (available at: http://www.dldcn.com). word-finding difficulties using semantic and phonologi-
WECHSLER, D., 1999, Wechsler Abbreviated Scale of Intelligence (New cal approaches. Child Language Teaching and Therapy, 9,
York, NY: Psychological Corporation/Harcourt Brace). 214–229.