Research in Autism Spectrum Disorders 7 (2013) 12821290

Contents lists available at ScienceDirect

Research in Autism Spectrum Disorders

Journal homepage: http://ees.elsevier.com/RASD/default.asp

Efcacy of cognitive behavior therapy-based social skills intervention for

school-aged boys with autism spectrum disorders
Cyndie Koning a,*, Joyce Magill-Evans a, Joanne Volden a, Bruce Dick b
a
b

Rehabilitation Medicine, Corbett Hall, 8205 114 St. NW University of Alberta, Edmonton, Alberta, Canada T6G 2G4
Anesthesiology and Pain Medicine, 8-120 Clinical Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2G3

A R T I C L E I N F O

A B S T R A C T

Article history:
Received 9 July 2011
Accepted 13 July 2011

School-aged children with Autism Spectrum Disorders (ASD) experience signicant

difculty with peer interaction. Research to identify the most effective strategies to
address this difculty has increased but more evidence is needed. Cognitive behavior
therapy (CBT), which focuses on changing how a person thinks about social situations as
well as how he behaves, is a promising approach. This study evaluated the efcacy of a 15
week CBT-based social skills intervention for boys aged 1012 years diagnosed with an
ASD. Boys with average or better IQ and receptive language skills were randomly assigned
to either a control (n = 8) or intervention condition (n = 7). During intervention, boys
attended weekly 2 h long group sessions focusing on self-monitoring skills, social
perception and affective knowledge, conversation skills, social problem-solving, and
friendship management skills. Comparison of the outcomes using repeated measures
analyses indicated that boys receiving the intervention scored signicantly better on
measures of social perception, peer interaction, and social knowledge than boys who had
not received intervention. There were no differences on general measures of socialization.
The manualized intervention used in this study shows promise but replication with larger
samples is needed.
Crown Copyright 2011 Published by Elsevier Ltd. All rights reserved.

Keywords:
Cognitive behavior therapy
Social skills
Intervention
ASD
Autism

1. Introduction
Difculties engaging in social interaction are a primary concern for children with high-functioning autism (HFA) or
Aspergers Syndrome (AS). While they are considered high-functioning by virtue of IQs in the average or above average range,
they have social decits which are primarily centered around social reciprocity, social cognition, and pragmatic language
(e.g., Adams, Green, Gilchrist, & Cox, 2002; Church, Alisanski, & Amanullah, 2000; Downs & Smith, 2004). Social difculties
become more evident as they begin school and move towards adolescence when the nuances of social interaction are more
demanding. These children initiate fewer social interactions with peers (Orsmond, Krauss, & Seltzer, 2004) and are less
socially responsive (Volkmar, 1987). They have difculty inferring others emotions and responding appropriately (Koning &
Magill-Evans, 2001a; Ozonoff, Pennington, & Rogers, 1990), taking others perspectives (Rehfeldt, Dillen, Ziomek, &
Kowalchuk, 2007), and understanding social rules and conventions of interaction (Church et al., 2000).

* Corresponding author. Current address: Occupational Therapy, Glenrose Rehabilitation Hospital, 10230-111 Avenue, Edmonton, Alberta, Canada T5G
0B7. Tel.: +01 780 735 7999x15239; fax: +01 780 735 6022.
E-mail addresses: Cyndie.Koning@albertahealthservices.ca (C. Koning), Joyce.magill-Evans@ualberta.ca (J. Magill-Evans), Joanne.volden@ualberta.ca
(J. Volden), Bruce.Dick@ualberta.ca (B. Dick).
1750-9467/$ see front matter . Crown Copyright 2011 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.rasd.2011.07.011

C. Koning et al. / Research in Autism Spectrum Disorders 7 (2013) 12821290

1283

Given the range of social difculties these children experience, it is important to determine the most pressing areas to
target for intervention and to determine which intervention methods are most appropriate for specic children. Bauminger
(2007a) suggested that the two main capabilities predicting social competence in middle childhood were conversational
skills and cooperative prosocial skills such as sharing feelings and experiences, collaborating, and comforting. Furthermore,
she felt that interventions needed to help these children develop multi-dimensional social competence, integrating
behavioral (e.g., social interaction), cognitive (e.g., accurate processing of information, perspective taking, problem-solving)
and affective skills (e.g., emotion knowledge) to adapt exibly to diverse social contexts and demands. (p. 1594).
Theoretical models help explain the impact of specic difculties and identify targets for intervention. A model of social
information-processing developed by Crick and Dodge (1994) for typically developing children and used with children with
other disabilities (Walz, Yeates, Wade, & Mark, 2010) may be useful for children with an Autism Spectrum Disorder (ASD)
(Embregts & van Nieuwenhuijzen, 2009). The model describes the elements of social interaction as well as the inuence of
social context and motivation. In the model, an interaction is composed of the following: social messages, conveyed both
verbally and nonverbally, are received and interpreted; social goals are claried; responses are generated; one or more is
selected and then enacted. All of these elements are inuenced by acquired social rules, social schemas, and social
knowledge. Studies of social competence interventions for school-aged children with HFA have examined one or more
outcomes that are congruent with the Crick and Dodge model (Bauminger, 2007a, 2007b; LeGoff & Sherman, 2006; SemrudClikeman, 2007; Tse, Strulovitch, Tagalakis, Meng, & Fombonne, 2007) as did the study reported here. Evidence related to the
models utility is gradually emerging.
Although current research has primarily focused on identifying decits in enacting social behavior and measuring effects
of intervention based on behavioral principles, cognitive behavioral therapy (CBT) is a promising treatment modality that
focuses on multiple dimensions of social competence and is congruent with Crick and Dodges model. CBT is broadly dened
as brief, structured therapy focused on context-driven problem-solving by linking thoughts, feelings and behaviors to
develop effective behaviors (Friedburg & McClure, 2002). It has a primary focus on monitoring and modifying thoughts and
beliefs that are unhelpful, negatively affect ones function, or result in inappropriate or unwanted behavior. It also stresses
the importance of understanding context and meaning assignment, the interplay between cognitive systems and emotional
and behavioral responses, and the role of schemas (beliefs, rules, assumptions about self and others) (Alford & Beck, 1997).
In contrast to behavioral techniques which rely on external factors for behavior change and which are often criticized for
failing to generalize gains, cognitive behavioral techniques have a greater focus on internal factors. Generalization through
homework is emphasized in CBT. It usually takes the form of a contract between the therapist and participants to try
specic strategies in natural settings, with the expectation that success apart from the therapy context is both internally
reinforcing and more likely to produce generalization. CBT focuses on the impact of thoughts on affect and how these factors
inuence individuals perceptions of the social world. This treatment method focuses on how a persons thoughts, feelings
and behaviors inuence each other to produce social responses. CBT has been adapted for children with ASD by increasing
the structure and predictability, including visual supports and verbal labeling, explicitly drawing attention to important
social cues and greater parent involvement (Beebe & Risi, 2003).
CBT-based social skills interventions have been evaluated for children with ASD (Bauminger, 2002, 2006, 2007a, 2007b;
Crooke, Hendrix, & Rachman, 2008; Laugeson, Frankel, Mogil, & Dillon, 2009; Lopata, Thomeer, Volker, & Nida, 2006; Wood
et al., 2009). Baumingers (2002) intervention addressed the inuence on behavior of teaching emotion recognition/
interpretation and social problem-solving; and the relationship of social problem-solving to later social adjustment. In her
research, children with HFA provided more relevant social solutions in problem-solving tasks post intervention. Solomon,
Goodlin-Jones, and Anders (2004) implemented a social adjustment curriculum for 8 to 12 year old boys with HFA which
also focused on affective education and a cognitive problem-solving approach and demonstrated clear improvements,
particularly on measures of emotion recognition and problem-solving. This focus on social problem-solving and
understanding emotions is a consistent theme in subsequent studies using a CBT approach with children with an ASD. Recent
studies (White et al., 2010; Wood et al., 2009) have applied principles of CBT to intervention for anxiety and social
interaction. These studies reported signicant improvements, usually based on parent- or self-report measures, satisfaction
surveys, social knowledge questions, or the assessment of emotion recognition.
To summarize, CBT seems particularly well suited to treatment of social skills in school-aged children with HFA or AS
because it addresses generalizability (Beebe & Risi, 2003); uses a problem-solving, coping approach that reects how
children in general learn to make social responses (Crick & Dodge, 1994); and uses cognitive strategies to build on the relative
strength of children who have IQs in the average or above range. The present pilot study examined the efcacy of a 12 week
group CBT-based social skills intervention. It was hypothesized that boys receiving intervention would demonstrate
signicant improvements compared to those who did not receive intervention on measures of social perception, social
problem-solving and knowledge, peer interaction, and general measures of socialization.
2. Methods
2.1. Participants
Participants were recruited through the local autism society newsletter and a local autism clinic database. Inclusion
criteria consisted of being male, aged 10 to 12 years, diagnosed with ASD using DSM-IV-TR criteria (American Psychiatric

1284

C. Koning et al. / Research in Autism Spectrum Disorders 7 (2013) 12821290

Association, 2000) by a child psychiatrist or developmental pediatrician, and having receptive language and nonverbal IQ
within normal limits (80 or above 1 standard error of measurement). Diagnosis was conrmed using the Autism Diagnostic
Observation Schedule (ADOS; Lord, Rutter, DiLavore, & Risi, 1999).
Average receptive language and IQ were considered necessary prerequisites because of the language and cognitive
requirements for the CBT-based intervention (Reynolds, Girling, Coker, & Eastwood, 2006). Boys were excluded if their
parents reported signicant behavioral difculties that would disrupt the group or if their parents had difculty with English
that would interfere with their ability to complete questionnaires and weekly homework assignments with their sons. All
participants except one attended regular educational settings and none were involved in any other behavioral intervention
during participation in this study.
Eligible boys and their parents attended an initial meeting to review the study objectives, complete the consent process,
and undergo assessment using the ADOS (Lord et al., 1999). Boys who scored above the ASD cut-off (ADOS total score greater
than 7) were then scheduled for an IQ and language assessment as described below. Parents and participants provided
consent and assent. Ethical approval was obtained from the local Ethics Board.
2.2. Measures
2.2.1. Inclusion criteria measures
IQ scores were obtained using the Wechsler Abbreviated Scale of Intelligence (WASI; Wechsler, 1999), except for one boy
recently assessed using the Test of Nonverbal Intelligence-3 (Brown, Sherbenou, & Johnsen, 1997). The WASI is an
individually administered, reliable and valid measure for individuals aged 689 years. The Full Scale IQ score was used in this
study [Mean (M) standard score = 100; Standard Deviation (SD) = 15]. It was administered by a trained research assistant,
supervised by an experienced psychologist.
The norm-referenced Clinical Evaluation of Language Fundamentals Fourth Edition (CELF-4; Semel, Wiig, & Secord,
2003) measures language and communication skills of individuals aged 521 years. The CELF-4 is reliable and valid. The full
measure was administered but only the receptive language score is reported (M = 100; SD = 15). The CELF-4 was
administered by an experienced speech language pathologist.
The ADOS (Lord et al., 1999) is a semi-structured standardized assessment considered one of the gold standard measures
of ASD with good reliability and validity. When used along with a developmental history by an experienced clinician, it
provides an accurate diagnosis (Ozonoff, Goodlin-Jones, & Solomon, 2005). It was administered by a research trained ADOS
assessor.
2.2.2. Outcome measures
As recommended by Williams White, Koenig, and Scahill (2006) and Koenig, De Los Reyes, Cicchetti, Scahill, and Klin
(2009), outcomes were measured by both observational and standardized measures and via multiple informants. All
observational outcome assessments were completed by qualied assessors blind to group.
Measures of socialization included the socialization scale of the Vineland Adaptive Behavior Scales Second Edition
(Vineland-II; Sparrow, Cicchetti, & Balla, 2005), and the Social Responsiveness Scale (SRS; Constantino & Gruber, 2005). The
Vineland-II is a standardized, norm-referenced, and widely used measure of adaptive behavior. The socialization domain of
the parent/caregiver rating form was used in the analysis (M = 100; SD = 15). Higher scores represent better adaptive
function. Internal consistency (ages 1012) ranged from .89 to .92 and testretest reliability was .93. The socialization score
has been used to measure intervention outcome for children with ASD (Laugeson et al., 2009; LeGoff & Sherman, 2006; Salt
et al., 2002).
The SRS (Constantino & Gruber, 2005) examines interpersonal behavior and communication including social awareness,
social cognition, social communication, social motivation, and social mannerisms. Ratings are on a 4-point Likert scale.
Higher scores reect greater impairment. The total score (M = 50; SD = 10), which reects the severity of social decits, was
used in this study. The authors suggest that the SRS may be used as an outcome measure for social skills intervention. Tse
et al. (2007) reported signicant differences with moderate effect sizes after a 12 week social skills intervention, suggesting
that it is sensitive to behavioral change over this period. The manual reports internal consistency (alpha) for the total score
for boys as .93, testretest reliability as .85 with a 17 month period between testing, and inter-rater reliability ranging from
.75 between father and teacher to .91 between mother and father.
The Child and Adolescent Social Perception measure (CASP; Magill-Evans, Koning, Cameron-Sadava, & Manyk, 1995)
assessed the ability to infer the emotional state of others from nonverbal (e.g., facial expression, tone of voice, gestures) and
contextual cues. The Emotion Score (ES) measures emotion recognition and the Nonverbal Cues Score (NCS) measures the
use of cues to infer emotions. Using short, video scenarios depicting social situations that children often encounter, the CASP
approximates natural social situations in that multiple and even conicting cues are presented simultaneously and the
environmental context must be considered. Lower scores reect greater impairment. The authors report internal consistency
as .88.92, testretest reliability as .83.87, and inter-rater reliability from .94 to .99. Cronbachs alpha for the current study
was .75 for the ES and .87 for the NCS. In a validation study, boys with ASD (n = 32) scored signicantly lower than 29
typically developing boys matched on age and vocabulary (Koning & Magill-Evans, 2001b). Semrud-Clikeman, Walkowiak,
Wilkinson, and Minne (2010) found that children with AS scored signicantly lower on the ES and NCS compared to control
children and lower than children with Attention Decit Hyperactivity Disorder-Predominantly Inattentive on the ES. Raw ES

C. Koning et al. / Research in Autism Spectrum Disorders 7 (2013) 12821290

1285

and the NCS were used in the analyses as these scores represent the number of emotions partially or completely named and
the actual number of cues recognized.
The Peer Interaction Measure (PIM; Koning, Magill-Evans, & Volden, 2008), a measure of initiating and maintaining
conversations, was developed for this study. Scores are obtained by viewing a video-recording of responses to a structured,
contrived social situation, in which a confederate greets the participant, encourages initiation and maintenance of a
conversation, and shifts conversation topics. The peer confederate, an 11 year old boy with acting experience, was recruited
from a performing arts school and paid to consistently enact the following waiting room scenario. The confederate was
seated, playing a hand-held video game. The participant was brought to the room and asked to wait. The boys were left alone.
The confederate then provided opportunities for the participant to interact without initially engaging the participant directly
in conversation. Hidden cameras recorded the interaction. After 10 min, the examiner returned and told the participant that
she was ready to see him.
Verbal and nonverbal behaviors were scored using a coding system derived from the ADOS (Lord et al., 1999), Magill
(1987), and Barry et al. (2003). The total score from Part II was used in the analyses. It has eight behaviors which are rated on
either a three or a ve point scale: orienting to peer confederate, gestures related to activity, facial expressions directed to
peer to communicate affect, ow of interaction, overall appropriateness of interaction, ability to adjust speech to peer, sense
of shared enjoyment in interaction, and showing an interest in peers responses (awareness of his interests). Higher scores
indicated greater skill. Cronbachs alpha, measuring internal consistency, was .89. Two coders tested the coding system with
typically developing boys, rened the coding process through consensus scoring, and then independently scored three
videotaped interactions, achieving an inter-rater reliability of 80% or greater prior to data collection. Inter-rater agreement
was calculated for 10% of the videotaped interactions and was 76%. This lower inter-rater agreement may have been partly
due to greater difculty coding interactions for children with ASD than typically developing boys. This measure was intended
to be a socially valid measure of how the participant responded in a natural setting and whether social skills generalized
beyond the treatment setting.
To measure changes in Social Knowledge, a test was developed which addressed understanding of social norms and rules,
starting and maintaining conversations, affective cues, and strategies for dealing with difcult situations. Fourteen brief one
sentence scenarios were followed by several possible answers or an open-ended question (total possible score of 34). The
measure was pilot tested with typically developing children in the target age group to develop an answer key and ensure that
questions were clear and unambiguous.
2.3. Procedures
To facilitate a greater likelihood that the boys in the two conditions would be equivalent on receptive language scores, 17
boys who met criteria for participation were stratied into two groupings: boys scoring above a standard score of 100 on the
Clinical Evaluation of Language Fundamentals-4 (Semel et al., 2003), and those scoring below 100. Boys in each grouping
were then randomly assigned to either a control condition (n = 8) or intervention (n = 9) using a random numbers table. One
boy assigned to the intervention dropped out before intervention began and another boy dropped out after participating in 5
of the 15 sessions. Seven boys completed the intervention.
Participants attended 2-h weekly group intervention for 15 sessions. Each group was led by two leaders with extensive
experience providing social skills intervention. All participants attended at least 13 of the 15 sessions; only one of the 7 boys
missed two sessions. The missed material was reviewed with the parents and participant in the week following the missed
group. All participants were given an honorarium of $50.00.
Testing occurred before the sessions began and after intervention was complete. For boys in the control condition, testing
was before and after a 12 week period. Parents completed the Vineland-II and the SRS while the participants were assessed
with the outcome measures. Parent survey forms for the Vineland-II were scored with the ASSIST software. SRS forms were
manually scored and rechecked using Excel.
2.3.1. Curriculum for intervention
Goals and activities followed a standard protocol employing CBT principles to connect thoughts, feelings, and
behaviors and addressing components of the Crick and Dodge (1994) model. The rst hour of intervention focused on
teaching social motivation and initiation, social perception and appropriate social responding, skill generalization, and
social problem-solving. Bakers (2003) manual and worksheets from Garcia Winners manuals (2002, 2005) provided
weekly goals and activities. Leaders coached, prompted, praised, and highlighted conversation and problem-solving
skills during another 45 min of loosely structured natural situations with fun activities (e.g., games, building). In
addition to the 12 regular sessions, three extra sessions (every 4th week) focused entirely on cooperative group
activities.
Intervention techniques included, for example, problem-solving activities such as an assigned Lego building activity,
feedback on video recorded specic role-playing, use of Polaroid cameras for facial expressions, eye spy games to track
someone elses gaze, thinking with your eyes, guessing group leaders nonverbal communication, and a who wants to be a
social skills millionaire? game. Activities in the group focused on asking questions to start and maintain conversations, even
when the question they were asking was not particularly interesting. Unstructured time provided out loud self talk
opportunities for participants to use cognitive rehearsal. At the end of each group session, participants were given individual

1286

C. Koning et al. / Research in Autism Spectrum Disorders 7 (2013) 12821290

feedback through an adaptation of an activity from Kendall and Barmish (2007) called Show That I Can, where participants
rated their accomplishments in meeting group goals.
Parents received handouts each week summarizing the goals and activities and a simple activity to try at home to
facilitate generalization. At the beginning of the following group, parents rated how well they thought the child had done
related to the previous weeks topic and whether they had tried the homework activity.
Treatment delity was maintained by using manualized intervention, documenting variations both on an individual and
group basis to any of the protocols, using a weekly checklist completed by participants and parents about homework, and
consistency of group leaders. Fidelity to the intervention model was monitored by a psychologist with extensive CBT
experience evaluating two to three of the 12 sessions/group to ensure adherence to CBT principles and equivalence between
groups, based on elements from the weekly lesson plan and the content and activities in the 45 min fun time.
3. Results
Prior to intervention, there were no signicant differences between boys who received the intervention and those who
did not on age, full scale IQ, ADOS total scores, and receptive language scores as measured using a multivariate ANOVA, F(4,
10) = .58, p = .68. Table 1 presents the mean descriptive variables for each group along with results for post-hoc test
comparisons for each variable in the analysis.
Changes in outcome measure scores for the intervention group were compared to changes in scores for the control group
using a series of repeated measures analyses of variance (ANOVA) examining the interaction of group (intervention, control)
and time (prior to intervention, after intervention). Random assignment with a small sample does not ensure that groups are
initially equivalent on the constructs of interest. The interaction effect indicated which group changed the most over time,
ensuring that differences between scores prior to intervention were taken into account. Signicance was set at p < .05 and
corrections were not made for multiple comparisons because the study presents pilot, exploratory data. Effect sizes were
calculated using partial eta squared (Brown, 2007). Partial eta squared values are considered small at .01, medium at .09 and
large at .25 (Bakeman, 2005; Wuensch, 2009).
There was a signicant Group by Time interaction effect for the two Child and Adolescent Social Perception scores, F(1,
13) = 9.84, p = .003, demonstrating that the group that received the intervention improved signicantly over time compared to
the group that did not receive intervention. As shown in Table 2, two post hoc univariate ANOVAs revealed signicant
interaction effects for each CASP score. The average gain of 10 points for the intervention group in their ability to correctly
recognize and label or partially label emotions is clinically signicant. Variability in the scores for the intervention group
increased from pre- to post-intervention testing, suggesting that not all boys made similar gains. For the Nonverbal Cues Score,
the intervention groups average score increased by 16.4 points, doubling after the 15 sessions, again with more variability
within the group after the intervention. The abilities of the control group on average remained very similar over the same time
period with unchanged variability. The effect size was large for the Emotion Score and for the Nonverbal Cues Score.
There was also a signicant Group by Time interaction effect for the knowledge questionnaire. The intervention groups
average score more than doubled. Although the two groups had different scores at the time of initial assessment, the control
groups average score increased only minimally without intervention. Neither group was close to the ceiling of 34 points. The
effect size was large.
Finally, there was a signicant Group by Time interaction effect for the Peer Interaction Measure total score. This measure
was an opportunity for the boys to use their social skills in a structured but naturalistic social situation with a peer and allowed a
limited evaluation of generalization of social skills learned in the intervention to a more typical situation. The average score for
the intervention group increased by 6 points, while the control group increased minimally. The average score for the control
group started out higher than that of the intervention group. Neither group was close to the ceiling of 36 points.
There were no signicant interaction effects on either of the norm referenced parent report measures (Vineland II, Social
Responsiveness Scale) indicating that the intervention group did not improve signicantly more than the control group over
time. Both of these measures are norm-referenced and use parent report.
Table 1
Descriptive information for participants in intervention (n = 7) and control (n = 8) groups.
Mean (SD)

Range

Group

Mean (SD)

10.312.2

Int
Ctl
Int
Ctl
Int
Ctl
Int
Ctl

10.99
11.15
113.14
106.37
92.29
93.50
13.14
10.25

Difference
between groups
p

F
Age in years
WASI Full Scale

11.07 (.54)
109.53 (17.50)

78136

CELF-4 receptive language

92.93 (10.68)

79113

ADOS algorithm total

11.60 (3.52)

717

(.52)
(.58)
(18.51)
(17.17)
(10.59)
(11.46)
(3.44)
(3.20)

.30

.59

.54

.48

.05

.83

2.85

.16

Note: WASI, Wechsler Abbreviated Scale of Intelligence (M = 100, SD = 15); ADOS, Autism Diagnostic Observation Schedule, scores of 7 or greater indicate
ASD, 10 or greater indicate Autism; CELF-4, Clinical Evaluation of Language Fundamentals Fourth Edition (M = 100, SD = 15).

C. Koning et al. / Research in Autism Spectrum Disorders 7 (2013) 12821290

1287

Table 2
Differences between the intervention and control groups on outcome measures.
Measure
Peer interaction

T1
T2

CASP ES

T1
T2

CASP NCS

T1
T2

Social knowledge

T1
T2

Vineland-II socialization

T1
T2

SRS totalb

T1
T2

Intervention Mean (SD)

Range

Control Mean (SD)

Range

17.14
927
23.43
1332
14.14
028
24.86
229
16.57
041
33.00
062
8.57
016
19.00
1323
72.57
6882
78.29
6296
80.71
7590
74.85
6092

24.62
1433
25.38
1736
18.63
1029
18.25
627
19.88
933
21.38
436
16.75
1022
17.88
1225
74.00
6285
77.50
6296
85.00
7898
79.62
5990

(5.46)
(7.04)
(9.46)
(13.35)
(14.14)
(21.44)
(5.02)
(3.65)
(5.53)
(11.53)
(5.22)
(11.61)

(6.16)

F(1,13)a

Effect size

Power

4.87

.046

.29

.52

20.49

.001

.61

.99

10.71

.006

.45

.86

27.90

<.001

.68

.99

.15

.70

.01

.07

.01

.92

.00

.05

(6.21)
(6.57)
(6.27)
(9.17)
(9.64)
(3.65)
(3.90)
(8.32)
(11.36)
(6.39)
(9.53)

Note. T1, testing before intervention; T2, testing after intervention; CASP, Child and Adolescent Social Perception Measure (ES, Emotion score; NCS,
Nonverbal Cues score); SRS, Social Responsiveness Scale (M = 50, SD = 10).
a
Degrees of freedom for CASP are 2, 12-both CASP scores were entered into one repeated measure calculation.
b
Decreased scores reect improvement.

4. Discussion
This small prospective study used a CBT-based social skills intervention group for boys aged 10 to 12 years and revealed
encouraging results despite the small sample. In comparison to the control group who received no intervention, the
intervention group signicantly improved in their ability to infer emotions, use nonverbal cues effectively, interact
successfully with a peer, and correctly answer questions about how to respond to social situations. The results support CBT as
an intervention technique for enhancing social skills of children with ASD whose cognitive skills fall in the typical range,
building on prior research (e.g., Bauminger, 2002, 2006, 2007a; Laugeson et al., 2009; White et al., 2010). Teaching children
to use metacognitive techniques rather than only teaching discrete skills (Crooke et al., 2008) appears promising.
Improvements in understanding social cues suggest that the boys with ASD in our sample can learn to recognize
nonverbal social cues when taught in the context of typically occurring social scenarios and reinforced during play-based
activities. Previous research also reported that children with ASD whose IQ falls in the average to above average range can be
taught to use their cognitive skills to better understand emotions (e.g., Bauminger, 2007a, 2007b; Kasari, Chamberlain, &
Bauminger, 2001; Sofronoff, Attwood, & Hinton, 2005; Solomon, Goodlin-Jones, & Anders, 2004). Crick and Dodges (1994)
model suggests that a better understanding of emotions should contribute to better overall social functioning.
Although the mechanisms for improvements need to be better understood, other CBT-based research suggests that a
contributing factor may be verbally describing the cognitive processes linked to observing and interpreting others nonverbal
cues. For example, participants were encouraged to say out loud what they were noticing about others cues and what this
might mean, earning points in games. The improvements seen on the social knowledge questionnaire on questions requiring
participants to describe how to respond to peers, may be related to encouraging them to take anothers perspective by
considering context, what they know about the person, and nonverbal cues. Application of this self-talk technique adds to
previous research using self talk with children with ASD to teach memory strategies (Bebko & Ricciuti, 2000) and self-control
(Groden & LeVasseur, 1995). This process makes explicit the more subtle ways that typically developing children infer
emotions.
Changes were observed on the Peer Interaction Measure which rates behaviors such as gestures, facial expressions,
adjusting to the speaker, showing an interest, and sharing enjoyment. Improvements on this measure also address
generalization of skills learned during intervention. The boys may have been more motivated to engage in the interaction
due to the peer confederates play with a handheld video game, an attractive activity for many of the participants. Many of
the boys discussed video games during unstructured time in the group and thus there were many opportunities to teach
appropriate ways to engage with peers about topics such as video games. Parent report measures failed to nd
improvements in other types of social situations which may have been less motivating.

1288

C. Koning et al. / Research in Autism Spectrum Disorders 7 (2013) 12821290

Failing to nd changes on the parent report measures may also relate to the measures sensitivity to change or to factors
related to how the participants behaved socially at home. In studies including control groups, there have been mixed results
from studies using parent report measures with some reporting trends for improvement (Owens, Granader, Humphrey, &
Baron-Cohen, 2008) and some showing improvement but with much longer intervention (LeGoff & Sherman, 2006).
Interestingly, the Child and Adolescent Social Perception measure, the Peer Interaction Measure and the social knowledge
questionnaire were all more direct measures of the childs skills or knowledge but only the CASP has some normative data.
The results from this paper and prior research (Embregts & van Nieuwenhuijzen, 2009) also indicate that Crick and
Dodges (1994) social information-processing model may be a useful framework for understanding social informationprocessing difculties, providing intervention, and guiding measurement of changes in social skills for children with ASD.
The model suggests that social behavior is a result of a series of steps which includes social perception (encoding and
interpreting social cues) and response generation based on social problem-solving, two areas that were specically
addressed in both measurement and intervention in this study.
4.1. Limitations
An obvious limitation is sample size (n = 15), which affects power and limits generalization. Limited power may also be an
explanation for the results on the Social Responsiveness Scale. Tse and colleagues study (2007) with 34 participants and no
control group was able to demonstrate signicant differences between pre- and post-intervention scores with a larger
sample size. Another concern is related to whether the effects are specic to the boys whose parents volunteered them and
could support attendance and homework completion. Generalization is also limited by inclusion criteria related to age,
language, and IQ. However, the signicant differences were strong and reected clinically signicant changes, suggesting
that the changes were real and meaningful and hold promise for generalizability. The results are in keeping with a growing
body of social skills intervention research and add rigor by inclusion of a randomly assigned control group.
Most of the signicant results were on measures that were not norm-referenced. This may be related to sensitivity to
change but further development of the Peer Interaction Measure is needed to ensure its validity. Whether improvements
were sustained has not been addressed. Twelve weeks of intervention is a relatively small dosage but not uncommon in CBT
programs. Clinical experience supports the need for booster sessions, frequently used in CBT. Two evidence-based
components of intervention, use of peer tutors and an emphasis on parent involvement should be addressed in future
studies.
5. Conclusion and future directions
Further research employing larger samples, including other evidence-based aspects of intervention and measures of
generalization will help to further delineate the efcacy of social skills intervention for children with ASD. In addition, Koenig
et al. (2009) emphasize the need for social validity in measurement. Results from the Peer Interaction Measure suggest that
further development of this measure related to its reliability and validity may be valuable. If its validity is corroborated by a
strong relationship between scores and observations of the child in natural settings, the Peer Interaction Measure may be a
more efcient method of collecting data on how children interact with peers.
Systematic reviews of social skills interventions for children with ASD (Cappadocia & Weiss, 2011; Rao, Beidel, & Murray,
2008; Reichow & Volkmar, 2010; Wang & Spillane, 2009) have called attention to important gaps in research rigor. Most
reviews also call for a focus on understanding which strategies and techniques work best for specic age groups or
developmental level. Future research should test CBT methods against other methods, as opposed to testing it against no
intervention. The present manualized study contributes to a growing body of research that supports CBT as a short-term
intervention for improving social-emotional understanding and social interaction in school-aged children with ASD.
References
Adams, C., Green, J., Gilchrist, A., & Cox, A. (2002). Conversational behavior of children with Asperger syndrome and conduct disorder. Journal of Child Psychology
and Psychiatry, 43, 679690.
Alford, B. A., & Beck, A. T. (1997). The integrative power of cognitive therapy. New York, NY: Guilford Publications.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed. text revision). Washington, DC: American Psychiatric
Association.
Bakeman, R. (2005). Recommended effect size statistics for repeated measures designs. Behavior Research Methods, 37, 379384.
Baker, J. E. (2003). Social skills training for children and adolescents with Asperger syndrome and social-communication problems. Shawnee Mission, KS: Autism
Asperger Publishing Company.
Barry, T. D., Klinger, L. G., Lee, J. M., Palardy, N., Gilmore, T., & Bodin, S. D. (2003). Examining the effectiveness of an outpatient clinic-based social skills group for
high-functioning children with autism. Journal of Autism and Developmental Disorders, 33, 685701.
Bauminger, N. (2002). The facilitation of socialemotional understanding and social interaction in high-functioning children with autism: Intervention outcomes.
Journal of Autism and Developmental Disorders, 32, 283298.
Bauminger, N. (2006). Multimodal (individual and group) CBT social skills training for high functioning children with ASD. Paper presented at the Gatlinburg
Conference, San Diego, CA.
Bauminger, N. (2007a). Brief report: Individual social-multimodal intervention for HFASD. Journal of Autism and Developmental Disorders, 37, 15931604.
Bauminger, N. (2007b). Brief report: Group social-multimodal intervention for HFASD. Journal of Autism and Developmental Disorders, 37, 16051615.
Bebko, J. M., & Ricciuti, C. (2000). Executive functioning and memory strategy use in children with autism: The inuence of task constraints on spontaneous
rehearsal. Autism: An International Journal of Research and Practice, 4, 299320.

C. Koning et al. / Research in Autism Spectrum Disorders 7 (2013) 12821290

1289

Beebe, D. W., & Risi, S. (2003). Treatment of adolescents and young adults with high-functioning autism or Asperger syndrome. In M. A. Reineck, F. M. Dattilio, & A.
Freeman (Eds.), Cognitive therapy with children and adolescents: A casebook for clinical practice (2nd ed., pp. 369401). New York, NY: The Guilford Press.
Brown, J. D. (2007). Statistics corner. Questions and answers about language testing statistics: Sample size and power. Shiken: JALT., Testing, Evaluation, SIG.,
Newsletter, 11, 3135., Also retrieved from the World Wide Web at, http://jalt.org/test/bro_28.htm.
Brown, L., Sherbenou, R. J., & Johnsen, S. K. (1997). Test of nonverbal intelligence: A language-free measure of cognitive ability (3rd ed.). Austin, TX: Pro-Ed.
Cappadocia, M. C., & Weiss, J. A. (2011). Review of social skills training groups for youth with Asperger Syndrome and High Functioning Autism. Research in Autism
Spectrum Disorders, 5, 7078, doi:10.1016/j.rasd.2010.04.001.
Church, C., Alisanski, S., & Amanullah, S. (2000). The social, behavioral, and academic experiences of children with Asperger syndrome. Focus on Autism and Other
Developmental Disabilities, 15, 1220.
Constantino, J. N., & Gruber, C. P. (2005). The Social Responsiveness Scale (SRS) manual. Los Angeles, CA: Western Psychological Services.
Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social information-processing mechanisms in childrens social adjustment. Psychological Bulletin,
115, 74101.
Crooke, P. J., Hendrix, R. E., & Rachman, J. Y. (2008). Brief report: Measuring the effectiveness of teaching social thinking to children with Asperger syndrome (AS)
and high functioning autism (HFA). Journal of Autism and Developmental Disorders, 38, 581591.
Downs, A., & Smith, T. (2004). Emotional understanding, cooperation, and social behavior in high-functioning children with autism. Journal of Autism and
Developmental Disorders, 34, 625635.
Embregts, P., & van Nieuwenhuijzen, M. (2009). Social information processing in boys with autistic spectrum disorder and mild to borderline intellectual
disabilities. Journal of Intellectual Disability Research, 53, 922931.
Friedburg, R. D., & McClure, J. M. (2002). Clinical practice of cognitive therapy with children and adolescents. New York: Guilford Press.
Garcia Winner, M. (2002). Thinking about you, thinking about me. San Jose, CA: Michelle Garcia Winner.
Garcia Winner, M. (2005). Worksheets! for teaching social thinking and related skills. San Jose, CA: Michelle Garcia Winner.
Groden, J., & LeVasseur, P. (1995). Cognitive picture rehearsal: A system to teach self-control. In K. Quill (Ed.), Teaching children with autism: Strategies to enhance
communication and socialization (pp. 287305). Albany, NY: Delmar Publishing.
Kasari, C., Chamberlain, B., & Bauminger, N. (2001). Social emotions and social relationships in autism: Can children with autism compensate? In J. Burack, T.
Charman, N. Yirmiya, & P. Zelazo (Eds.), Development and autism: Perspectives from theory and research (pp. 309323). Hillsdale, NJ: Erlbaum Press.
Kendall, P. C., & Barmish, A. J. (2007). Show-That-I-Can (homework) in cognitive-behavioral therapy for anxious youth: Individualizing homework for Robert.
Cognitive and Behavioral Practice, 14, 289296.
Koenig, K., De Los Reyes, A., Cicchetti, D., Scahill, L., & Klin, A. (2009). Group intervention to promote social skills in school-age children with pervasive
developmental disorders: Reconsidering efcacy. Journal of Autism and Developmental Disorders, 39, 11631172.
Koning, C., & Magill-Evans, J. (2001a). Social and language skills in adolescent boys with Asperger syndrome. Autism, 5, 2336.
Koning, C., & Magill-Evans, J. (2001b). Validation of the child and adolescent social perception measure. Occupational Therapy Journal of Research, 21, 4967.
Koning, C., Magill-Evans, J., & Volden, J. (2008 May). Measurement of peer interaction in children with HFA. Poster presented at the International Meeting for Autism
Research, London Abstract retrieved from http://imfar.confex.com/imfar/2008/webprogam/Paper1289.html.
Laugeson, E. A., Frankel, F., Mogil, C., & Dillon, A. R. (2009). Parent-assisted social skills training to improve friendships in teens with Autism Spectrum Disorders.
Journal of Autism and Developmental Disorders, 39, 596606.
LeGoff, D. B., & Sherman, M. (2006). Long-term outcome of social skills intervention based on interactive LEGO play. Autism, 10, 317329.
Lopata, C., Thomeer, M. L., Volker, M. A., & Nida, R. E. (2006). Effectiveness of a cognitive-behavioral treatment on the social behaviors of children with Asperger
disorder. Focus on Autism and Other Developmental Disabilities, 21, 237244.
Lord, C., Rutter, M., DiLavore, P., & Risi, S. (1999). Autism diagnostic observation schedule manual. Los Angeles, CA: Western Psychological Services.
Magill, J. (1987). The nature of social decits in children with autism . Unpublished doctoral dissertation. University of Alberta, Edmonton, AB.
Magill-Evans, J., Koning, C., Cameron-Sadava, A., & Manyk, K. (1995). The child and adolescent social perception measure. Journal of Nonverbal Behavior, 19,
151169.
Orsmond, G. I., Krauss, M. W., & Seltzer, M. M. (2004). Peer relationships and social and recreational activities among adolescents and adults with autism. Journal of
Autism and Developmental Disorders, 34, 245256.
Owens, G., Granader, Y., Humphrey, A., & Baron-Cohen, S. (2008). LEGO1 therapy and the social use of language programme: An evaluation of two social skills
interventions for children with high functioning autism and Asperger syndrome. Journal of Autism and Developmental Disorders, 38, 19441957.
Ozonoff, S., Goodlin-Jones, B. L., & Solomon, M. (2005). Evidence-based assessment of autism spectrum disorders in children and adolescents. Journal of Clinical
Child and Adolescent Psychology, 34, 523540.
Ozonoff, S., Pennington, B. F., & Rogers, S. J. (1990). Are there emotion perception decits in young autistic children? Journal of Child Psychology and Psychiatry, and
Allied Disciplines, 31, 343361.
Rao, P. A., Beidel, D. C., & Murray, M. J. (2008). Social skills interventions for children with Aspergers syndrome or high-functioning autism: A review and
recommendations. Journal of Autism and Developmental Disorders, 38, 353361.
Rehfeldt, R. A., Dillen, J. E., Ziomek, M. M., & Kowalchuk, R. E. (2007). Assessing relational learning decits in perspective-taking in children with high functioning
autism spectrum disorder. Psychological Record, 57, 2347.
Reichow, B., & Volkmar, F. R. (2010). Social skills interventions for individuals with autism: Evaluation for evidence-based practices within a best evidence
synthesis framework. Journal of Autism and Developmental Disorders, 40, 149166.
Reynolds, S., Girling, E., Coker, S., & Eastwood, L. (2006). The effect of mental health problems on childrens ability to discriminate amongst thoughts, feelings and
behaviours. Cognitive Therapy and Research, 30, 599607.
Salt, J., Shemilt, J., Sellars, V., Boyd, S., Coulson, T., & McCool, S. (2002). The Scottish Centre for autism preschool treatment programme II: The results of a controlled
treatment outcome study. Autism, 6, 3446.
Semel, E. M., Wiig, E. H., & Secord, W. (2003). Clinical evaluation of language fundamentals fourth edition (CELF-4). San Antonio, TX: Psychological Corporation.
Semrud-Clikeman, M. (2007). Social competence in children. NewYork: Springer.
Semrud-Clikeman, M., Walkowiak, J., Wilkinson, A., & Minne, E. P. (2010). Direct and indirect measures of social perception, behavior, and emotional functioning
in children with Aspergers disorder, nonverbal learning disability, or ADHD. Journal of Abnormal Child Psychology, 38, 509519.
Sofronoff, K., Attwood, T., & Hinton, S. (2005). A randomised controlled trial of a CBT intervention for anxiety in children with Asperger syndrome. Journal of Child
Psychology and Psychiatry, 46, 11521160.
Solomon, M., Goodlin-Jones, B. L., & Anders, T. F. (2004). A social adjustment enhancement intervention for high functioning autism, Aspergers syndrome, and
pervasive developmental disorder NOS. Journal of Autism and Developmental Disorders, 34, 649668.
Sparrow, S. S., Cicchetti, D. V., & Balla, D. A. (2005). Vineland adaptive behavior scales: second edition (Vineland II), survey interview form/caregiver rating form. Livonia,
MN: Pearson Assessments.
Tse, J., Strulovitch, J., Tagalakis, V., Meng, L., & Fombonne, E. (2007). Social skills training for adolescents with Asperger syndrome and high-functioning autism.
Journal of Autism and Developmental Disorders, 37, 19601968.
Volkmar, F. R. (1987). Social development. In D. J. Cohen, A. M. Donnellan, & R. Paul (Eds.), Handbook of autism and pervasive developmental disorders (pp. 4160).
New York, NY: Wiley.
Walz, N. C., Yeates, K. O., Wade, S. L., & Mark, E. (2010). Social information processing skills in adolescents with traumatic brain injury: Relationship with social
competence and behavior problems. Journal of Pediatric Rehabilitation Medicine, 2, 285296.
Wang, P., & Spillane, A. (2009). Evidence-based social skills interventions for children with autism: A meta-analysis. Education and Training in Developmental
Disabilities, 44, 318342.
Wechsler, D. (1999). Wechsler Abbreviated Scale of Intelligence (WASI). San Antonio, TX: Harcourt Assessment.

1290

C. Koning et al. / Research in Autism Spectrum Disorders 7 (2013) 12821290

White, S. W., Albano, A. M., Johnson, C. R., Kasari, C., Ollendick, T., Klin, A., & Scahill, L. (2010). Development of a cognitive-behavioral intervention program to treat
anxiety and social decits in teens with high-functioning autism. Clinical Child and Family Psychology Review, 13, 7790.
Williams White, S., Koenig, K., & Scahill, L. (2006). Social skills development in children with autism spectrum disorders: A review of the intervention research.
Journal of Autism and Developmental Disorders, 37, 18581868.
Wood, J. J., Drahota, A., Sze, K., Har, K., Chiu, A., & Langer, D. A. (2009). Cognitive behavioral therapy for anxiety in children with autism spectrum disorders: A
randomized, controlled trial. Journal of Child Psychology and Psychiatry, 50, 224234.
Wuensch, K. L. (2009). Standardized effect size estimation: Why and how? Retrieved from http://core.ecu.edu/psyc/wuenschk/StatsLessons.htm.