Research in Autism Spectrum Disorders 7 (2013) 411–417

Contents lists available at SciVerse ScienceDirect

Research in Autism Spectrum Disorders

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

Using a behavioral skills training package to teach conversation skills to

young adults with autism spectrum disorders
Jodi E. Nuernberger, Joel E. Ringdahl *, Kristina K. Vargo, Anna C. Crumpecker,
Karl F. Gunnarsson
Southern Illinois University – Carbondale, United States

A R T I C L E I N F O A B S T R A C T

Article history: A behavioral skills training package was used to teach vocal and non-vocal conversation
Received 13 July 2012 skills to young adults with autism spectrum disorders. A task analysis was created and
Received in revised form 13 September 2012 verified that included both vocal conversation skills such as making comments related to
Accepted 14 September 2012 the conversation topic, and non-vocal conversation skills such as maintaining appropriate
proximity. The behavioral skills training package included behavioral skills training, in situ
Keywords: training, and reinforcer delivery. Behavioral skills training involved delivering instruc-
Autism tions, modeling of an appropriate conversation, and role-playing with the participants in a
Social skills
private training room. In situ training followed each successful role-play. Access to a
Conversation
preferred item/activity was provided for performing at or above the previous session. The
Behavioral skills training
Task analysis
training package was shown to be effective and effects maintained during four to eight
week follow-up observations.
ß 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Although social skills deficits are a central feature of autism spectrum disorder (ASD), few children receive adequate
social skills programming (Bellini, Peters, Benner, & Hopf, 2007). Individuals with ASDs have difficulty appropriately
interacting with others, and, in particular, have difficulty initiating and maintaining conversations (Painter, 2006). One goal
for individuals with ASDs should be to develop more ‘‘typical’’ conversation skills, including maintaining a conversation topic
for more than one exchange and engaging in appropriate non-vocal conversation behavior. This goal becomes increasingly
important as the individual ages and pursues more independent social situations such as employment, higher education, and
more independent living settings. Roessler, Brolin, and Johnson (1990) found students who had higher social skills based on
teacher ratings were more likely to have a higher quality of life (i.e., independent living) and be engaged in post-school
employment.
One limitation of past social skills interventions has been the lack of demonstration of the social skills in the natural
environment. Gresham, Sugai, and Horner (2001) suggested social skills intervention often take place in resource rooms or
other pullout settings. The lack of demonstration of social skills in the natural environment might be attributed to the
contrived contexts in which social skills are often taught (Gresham et al., 2001).
Some social skills interventions have been shown to be effective in separate training contexts, but not in the natural
environment. For example, Lalli, Pinter-Lalli, Mace, and Murphy (1991) used a role-play training package to teach social skills
(i.e., compliments, social interactions, politeness, criticism, social confrontation, and questions/answers) to adults with

* Corresponding author at: Rehabilitation Institute, Southern Illinois University – Carbondale, 1025 Lincoln Drive – Rehn Hall, Carbondale, IL 62901,
United States. Tel.: +1 618 453 8295.
E-mail address: joelringdahl@siu.edu (J.E. Ringdahl).

1750-9467/$ – see front matter ß 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.rasd.2012.09.004
412 J.E. Nuernberger et al. / Research in Autism Spectrum Disorders 7 (2013) 411–417

developmental disabilities at a university-affiliated group home. In a group game context, one participant drew a card with a
social situation and acted out the scenario while the next participant reacted to the situation. Participants were given
feedback about their performance. During generalization probes the same social skills group gathered in a room but the
social skills game was not played, rather the group was told to relax. No other instructions or feedback were provided.
Immediate improvement in the use of the trained social skills was observed during generalization probes. Although the
participants demonstrated the use of social skills with their group members, their use of social skills with peers outside of the
group was not shown.
Dotson, Leaf, Sheldon, and Sherman (2010) used a training package to teach conservation skills to adolescents with an
ASD in a group setting. The training package consisted of describing the target behavior, why it should be used, when it
should be used, and the steps in the skill, modeling the behavior, having the learner role-play, and delivering feedback to the
participant. Four of the five participants mastered all three conversational skills (i.e., how to give positive feedback to a
speaker, and how to answer and ask open-ended questions). The fifth participant mastered two of the skills. Some skills were
demonstrated in the natural environment with a typical peer; however, no participants demonstrated generalization of all
skills in the natural environment.
Other limitations of past social skills interventions with individuals with ASDs include the complexity of the intervention
and the intensity of human resources required to maintain the treatment effects (Cimera & Cowan, 2009). Some common
procedures used to teach social skills include social stories, script fading, and video modeling (Flynn & Healy, 2012;
Schreiber, 2010), but these procedures might not be ideal for all teaching scenarios. For example, script fading might not be
ideal when teaching multi-component skills such as flexible conversational responding (Flynn & Healy, 2012). Furthermore,
social stories and video modeling require the use of additional, sometimes costly, materials (e.g., stories and video
equipment) that might not be available in all settings (Leaf et al., 2009).
Other training packages that include role-playing components, such as behavioral skills training (BST), have also been
shown to be effective in teaching social skills. These strategies might be more useful for teaching particular social skills, such
as conversation skills to adults with ASDs because they provide individuals with the opportunity to practice the behavior and
receive feedback (Leaf et al., 2009). BST is a role-play training package that includes delivering clear instructions to the
participant, modeling the target skill, providing the participant with the opportunity to practice the skill, and providing
corrective feedback. Stewart, Carr, and LeBlanc (2007) taught family members to implement BST to teach a 10-year-old child
with Asperger’s disorder and attention-deficit/hyperactivity disorder conversation skills at home. In particular, the family
targeted making appropriate eye contact, asking whether the listener was bored, asking whether the listener preferred to
change the conversation topic, and avoiding topics that the parents reported were perseverative. The child’s frequency of
appropriate conversation skills increased following BST.
Appropriately initiating and maintaining a conversation requires a complex skill set, which can be difficult to objectively
and operationally define (Minkin et al., 1976). Individuals with ASDs exhibit behavioral patterns that may inhibit their
effective use of communication, including difficulty following social mores regarding appropriate non-verbal behaviors such
as maintaining eye contact and appropriate proximity and engaging in restricted conversational interests. Although many
studies have described the teaching of important vocal conversation skills such as reciprocating one comment/question,
making one on-topic statement during an interaction, and avoiding perseverative speech (Leaf et al., 2009; Stewart et al.,
2007), fewer studies have focused on teaching non-vocal conversational skills (e.g., maintaining eye contact, distance,
posture, and tone throughout an interaction) (Dotson et al., 2010). Additionally, few studies have examined conversation
topic maintenance, although this skill is an important aspect of conversation (Leaf et al., 2009).
The purpose of this study was to examine the efficacy of using a role-play training package (i.e., BST, in situ training, and
reinforcement) to teach vocal and non-vocal conversation skills to individuals diagnosed with ASDs. Conversation skills have
most often been addressed with children or adolescents with ASDs (Dotson et al., 2010; Leaf et al., 2009; Leaf, Dotson,
Oppenheim, Sheldon, & Sherman, 2010; Minkin et al., 1976). Thus, a second purpose of this study was to implement the
procedures specifically with young adults (i.e., between the ages of 18 and 23) diagnosed with ASDs.

2. Method

2.1. Participants and setting

Participants were recruited from a comprehensive rehabilitation facility that provides vocational, transitional, and
behavioral services to individuals with developmental and/or physical disabilities. Young adults with developmental
disabilities typically participated in the program’s services for an average of six months. Three residents at the facility
participated. Missy, 19-years-old, had a diagnosis of Autism and a mild intellectual impairment as measured by the Wechsler
Adult Intelligence Scale-III (FSIQ = 69). Matthew, 23-years-old, had a diagnosis of Asperger’s Syndrome and Cerebral Palsy
and did not have an intellectual impairment as measured by the Wechsler Adult Intelligence Scale-III test (FSIQ = 127).
Rodger, 19-years-old, was diagnosed with Autism and a mild intellectual impairment as measured by the General
Intellectual Ability portion of the Woodcock–Johnson III test of cognitive abilities (score = 65). The participants were referred
for behavioral consultation by their case managers to address deficits in social communication. Each of the participants gave
his or her consent to participate; in addition Missy’s guardians consented to her participation, as she was not her own
guardian.
 J.E. Nuernberger et al. / Research in Autism Spectrum Disorders 7 (2013) 411–417 413

Other residents at the comprehensive rehabilitation facility served as peers throughout the study. Peers were not
recruited for the study, were not aware of the purpose of the study, and did not undergo any training for the study. Peers had
either a physical or developmental disability, some had mild intellectual disabilities, and all were between the ages of 18 and
25.
Baseline, in situ, maintenance, and follow-up sessions were conducted on the living unit where participants socialized,
made meals, and resided. The living unit contained a kitchen, dining room, living room with two couches and television, and
two additional television/video game rooms that contained four chairs each. In addition to two experimenters and the
participant, 16–20 other residents and 2–4 graduate assistants hired to provide supervision at the facility were present
during sessions conducted on the living unit. BST sessions were conducted in a private observation room with only the two
experimenters and participant present. The private observation room was 4 m  5 m and contained a table, four chairs, and
the written instructions for BST sessions.

2.2. Response measurement and interobserver agreement

A doctoral graduate student who was a board certified behavior analyst and two Master’s-level graduate students served
as experimenters. Each step of the conversation task analysis was objectively defined. Observers recorded the use of correct
steps completed during conversation, and reported these data as a percentage of correct steps, defined by the Task Analysis
(see Table 1). On at least 60% of sessions across participants (range, 60–69%), a second experimenter simultaneously but
independently recorded data for the purpose of interobserver agreement. Agreement was defined as both observers
recording correct or incorrect for each step of the task analysis. Mean agreement was calculated by dividing the number of
agreements by the number of agreements plus disagreements and multiplying by 100%. Mean agreement was 84% (range,
60–100%) for Matthew, 84% (range, 80–100%) for Rodger, and 88% (range, 80–100%) for Missy. Initial sessions for Matthew
yielded lower mean agreement scores because conversations were held in the kitchen and it was difficult for the observers to
hear the conversation. On subsequent sessions observers were asked to stand closer to the participants and agreement
improved.

2.3. Validation

We established a task analysis that included vocal and non-vocal steps to initiating and maintaining an appropriate
conversation with peers by referencing sources that addressed common conversational deficits exhibited by individuals
with ASDs (Painter, 2006; Winner, 2005). Skill deficits and conversation components that were common to both sources
were selected for inclusion in the task analysis. The initial task analysis included 10 steps. Undergraduate and graduate
student interns employed by the facility were observed conversing to assess the validity of the task analysis. The criterion for

Table 1
Appropriate conversation task analysis.

Step Description Definition

1 Stand/sit at least an arm’s length away with body
toward the other person
2 Look at the other person’s face
3 Say greeting
4 Ask a question related to an appropriate topic
5 Wait appropriately for other person to respond
6 Make a statement/question related to same topic
7 Wait appropriately for other person to respond
8 Make a statement/question related to same topic
9 Wait appropriately for other person to respond
10 End the conversation when
(1) there is a pause of 4 s or more
(2) the only thing they are saying is ‘‘yes,’’ or ‘‘no,’’
and you’ve asked them at least two questions
about an appropriate topic
(3) the other person ends the conversation
Participant stood or sat at least an arm’s length away with body facing
toward the other person
Participant looked at the other person’s face when initiating the conversation
Participant said greeting
Participant asked a question related to an appropriate topic. Specifically
defined for each participant according to his or her perseverative interests
(e.g., inappropriate topics include race, religion, pregnancy)
Participant faced his/her body toward the other person, at least an arm’s
length away, and did not interrupt the other person
Participant made a statement or questions related to the same topic that
differed from a previous statement or question. Participant faced his/her body
toward the other person, at least an arm’s length away, and did not interrupt
the other person
Participant faced his/her body toward the other person, at least an arm’s
length away, and did not interrupt the other person
Participant made a statement or questions related to the same topic that
differed from previous statements or questions. Participant faced his/her body
toward the other person, at least an arm’s length away, and did not interrupt
the other person
Participant faced his/her body toward the other person, at least an arm’s
length away, and did not interrupt the other person
The participant ended the conversation when a pause of 4 s elapsed, when
3 exchanges were made and the other person only responded with one word
phrases, or in response to the other person ending the conversation
414 J.E. Nuernberger et al. / Research in Autism Spectrum Disorders 7 (2013) 411–417

step inclusion was for the step to be observed across three conversations. Results of these observations suggested that six of
the steps in the initial task analysis were on-target, two of the definitions required revision, and two new steps should be
included.
The final task analysis consisted of 10 steps: (1) Stand/sit at least arm’s length away with body toward the other person,
(2) Look at the other person’s face, (3) Say greeting, (4) Ask a question related to an appropriate topic, (5) Wait appropriately
for other person to respond, (6) Make a statement/question related to same topic, (7) Wait appropriately for other person to
respond, (8) Make a statement/question related to same topic, (9) Wait appropriately for other person to respond, and (10)
End the conversation when interest is lost or when the other person ends the conversation.

2.4. Experimental design and procedure

A multiple-baseline design across participants was used to assess the effects of using the BST package to establish
conversation skills. When a participant exhibited stability in baseline conversation performance or a decreasing trend
(according to the task analysis), BST and in situ training sessions were implemented. BST sessions preceded every in situ
training session. Three sessions of BST training were conducted followed by maintenance and follow-up sessions.

2.4.1. Baseline
The experimenter asked the participant to have a conversation with a peer. The participant chose the conversation topic.
The experimenter recorded the number of steps completed correctly according to the task analysis. The participant did not
receive access to preferred items/activities following these conversations. Peers were not aware of the study, therefore, each
session participants engaged in conversation with a different peer (i.e., whoever was available during that time). That is, each
session, the participant approached a peer engaged in a leisure activity (e.g., playing cards or video games, or watching
television) and initiated a conversation. Each session was the length of one conversation composed of at least three
exchanges (approximately 5 min).

2.4.2. Behavioral skills training

Participants received BST to teach them how to initiate and maintain an appropriate conversation based on the task
analysis. The experimenter brought the participant to a private observation room. The therapist provided the participant
with instructions, modeling, rehearsal, and feedback during each session. First, a written description of the task analysis was
given to the participant and read aloud. The description described how to have an appropriate conversation. Next, two
experimenters modeled having an appropriate conversation. Then the participant rehearsed having an appropriate
conversation with one of experimenters, while the other experimenter observed. Immediately following the rehearsal, the
experimenter gave the participant feedback regarding his or her performance. If the participant did not perform 100% of the
steps correctly, rehearsal was repeated. During training, any questions about the task analysis were answered. Comments or
questions not related to having an appropriate conversation were ignored.
When the participant could complete 100% of steps correctly during the rehearsal, s/he returned to the living center for in
situ training. No participant required more than three rehearsals to reach 100% accuracy prior to each in situ training trial
(see Table 2). During the in situ training, the experimenter asked the participant to initiate a conversation with a peer. As in
baseline, each session the participant engaged in conversation with a different peer and the participant chose the
conversation topic. During in situ training several peers were present. Therefore, feedback was delivered regarding the steps
that were implemented correctly and the steps that were not completed correctly immediately following the session, but in
the private observation room to maintain the participants’ privacy. Following the first in situ trial, participants received
3 min access to a preferred activity identified via a paired-stimulus preference assessment (Fisher et al., 1992). On
subsequent trials, access to the preferred activity (e.g., internet access) was contingent on consistent or improved
performance as compared to the previous session.
BST sessions were approximately 10 min. One to five sessions were conducted consecutively per day with each
participant. Baseline and BST sessions were conducted over a 4-week period with each participant.

2.4.3. Maintenance and follow-up

One to two weeks following BST, experimenters re-implemented baseline sessions (i.e., maintenance sessions; no
feedback and no access to preferred items/activities) during which the participant was instructed to have a conversation
with a peer, and was observed having conversations with peers on the participant’s living unit. Using the same procedures,
follow-up observations were conducted 6–8 weeks following maintenance sessions.

Table 2
BST rehearsal trial percentages prior to each BST in situ session.

Participant BST rehearsals prior to in situ session 1 BST rehearsals prior to in situ session 2 BST rehearsals prior to in situ session 3

Missy 100% 70%, 100% 100%

Rodger 80%, 100% 100% 60%, 90%, 100%
Matthew 100% 100% 100%
 J.E. Nuernberger et al. / Research in Autism Spectrum Disorders 7 (2013) 411–417 415

3. Results

Fig. 1 displays the results of the BST package for all three participants. During the baseline phase, the mean percentage of
correct steps for Missy (top panel) was variable, however a decreasing trend was observed (M = 45%; range, 10–80%). When
BST was implemented an immediate change in level of responding was observed such that the percentage of correct
responses stabilized and increased to a mean of 90% (range, 80–100%). During the maintenance phase (one week following
training) Missy continued to engage in high percentages of correct responding with a mean of 93% (range, 80–100%).
Rodger engaged in stable responding during baseline (M = 49% correct steps; range, 40–60%) over 7 sessions. Following
the implementation of BST, the percentage of correct steps increased to a mean of 83% (range, 80–90%), such that no overlap
between data paths was observed. A maintenance session was conducted one week after treatment with Rodger during
which he engaged in 90% correct responding. A four-week and eight-week follow up session was also conducted with Rodger
in which he showed maintenance of responding (90% and 80%, respectively).
During the baseline phase for Matthew, he engaged in variable responding with a mean percentage of correct steps of 38%
(range, 10–60%) over 11 sessions. Upon implementation of BST, the mean percentage of correct steps stabilized and
increased to 83% (range, 70–100%). No overlap between baseline and BST data paths was observed. Performance during the
one-week maintenance sessions averaged 97% (range, 90–100%) across three sessions. In addition, a six-week follow-up
session was completed and Matthew engaged in 100% correct responding during the session.

Fig. 1. The percentage of steps on the appropriate conversation task analysis that were implemented correctly during baseline, behavioral skills training in
situ probes, maintenance, and follow-up sessions.
416 J.E. Nuernberger et al. / Research in Autism Spectrum Disorders 7 (2013) 411–417

4. Discussion

The results of the current study demonstrated that the training package including BST, in situ training, and reinforcement
was immediately effective in teaching three young adults with ASDs vocal and non-vocal conversation skills, as defined
according to a task analysis. One concern with past social skills interventions is the lack of demonstration of the skills in the
natural environment (Gresham et al., 2001). However, in the current study, we saw immediate improvement in conversation
skills in the natural environment, and maintenance of those skills, for all participants. This training package involved
providing the participants with instructions and a demonstration of an appropriate conversation in a private training room.
Subsequently, participants role-played having an appropriate conversation with the experimenter in the training room, and
then practiced with a peer in the natural environment during in situ training. Researchers have shown that BST is effective
when in situ training is included (Johnson et al., 2006). Additionally, training in the natural environment yields better
treatment outcomes (Bellini et al., 2007). The in situ training in the current study might have had an impact on the success in
the natural environment and maintenance of treatment effects, but further research is needed to identify the impact any one
component of the training package might have had on treatment effects.
Unlike past interventions that have been critiqued as being too time and resource intensive (Cimera & Cowan, 2009), our
intervention did not require costly materials. Furthermore, our intervention was conducted within four weeks and
maintenance effects were demonstrated for seven to ten weeks following the BST package. Although our intervention
package was effective, we do not know the extent to which BST would have been effective without the reinforcement
component. However, all three participants demonstrated clear improvement during the first session of BST prior to
reinforcer delivery. Future researchers might examine whether the reinforcement component was necessary, or how
additional reinforcement affects the persistence of treatment effects following removal of the intervention.
During baseline interesting results were obtained. Some variability in responding was observed for both Rodger and
Matthew, and a decreasing trend was observed for Missy. Initial high data points for Missy suggests that perhaps she had
many of the appropriate conversation skill components in her repertoire. It is unknown why a decreasing trend was
observed; however, whether she did not have the necessary skill set or was unmotivated to engage in the appropriate
behaviors, baseline data indicated the BST intervention package was warranted.
In this study the natural environment posed many distractions for the participants (e.g., video games and movies). One
limitation of the current study was the inability to observe participant’s conversations in the absence of the experimenters.
Thus, there might have been some reactivity due to the experimenters’ presence on the living unit. One alternative approach
would have been to videotape the interactions. However, videotaping was not feasible because many of the individuals on
the living unit were not participants of the study. Video recording of sessions may have been a breach of confidentiality.
Several baseline sessions were conducted which allowed the researchers to determine if initial high percentages of correct
responding were an indication that the participant did not need the training. If the participant was not in need of training,
stable high percentages of correct responding would have been observed.
Another potential limitation of this study was that follow-up data were not collected for Missy. Missy had completed her
6-month stay at the facility and was unavailable for follow-up. Likewise, only one maintenance session was conducted with
Rodger, due to his absence during much of the maintenance period. The limited follow-up data for Rodger indicated that
treatment effects had maintained. Additionally, procedural integrity data and specific data concerning how conversations
were ended and by whom they were ended were not collected. Although more data could have supplied more strength for
the effectiveness of the BST package, results showed clear improvement for all participants immediately following training.
Our BST intervention package was effective in teaching a number of vocal and non-vocal conversation components to
young adults with ASD. Interventions involving task analysis, role-play, and reinforcement may be simple interventions that
are effective in teaching social skills. Future researchers should continue to examine interventions for this underserved
population and examine response maintenance, generalization, and the social validity of such interventions.

References

Bellini, S., Peters, J. K., Benner, L., & Hopf, A. (2007). A meta-analysis of school-based social skills interventions for children with autism spectrum disorders.
Remedial and Special Education, 28, 153–162.
Cimera, R. E., & Cowan, R. J. (2009). The costs of services and employment outcomes achieved by adults with autism in the US. Autism, 13, 285–302.
Dotson, W. H., Leaf, J. B., Sheldon, J. B., & Sherman, J. A. (2010). Group teaching of conversational skills to adolescents on the autism spectrum. Research in Autism
Spectrum Disorders, 4, 199–209.
Fisher, W., Piazza, C. C., Bowman, L. G., Hagopian, L. P., Owens, J. C., & Slevin, I. (1992). A comparison of two approaches for identifying reinforcers for persons with
severe and profound disabilities. Journal of Applied Behavior Analysis, 25, 491–498.
Flynn, L., & Healy, O. (2012). A review of treatments for deficits in social skills and self-help skills in autism spectrum disorder. Research in Autism Spectrum
Disorders, 6, 431–441.
Gresham, F. M., Sugai, G., & Horner, R. H. (2001). Interpreting outcomes of social skills training for students with high-incidence disabilities. Teaching Exceptional
Children, 67, 331–344.
Johnson, B. M., Miltenberger, R. G., Knudson, P., Egemo-Helm, K., Kelso, P., Jostad, C., et al. (2006). A preliminary evaluation of two behavioral skills training
procedures for teaching abduction–prevention skills to schoolchildren. Journal of Applied Behavior Analysis, 39, 25–34.
Painter, K. K. (2006). Social skills groups for children and adolescents with Asperger’s Syndrome: A step-by-step program. London: Jessica Kingsley Publishers.
Lalli, J. S., Pinter-Lalli, E., Mace, F. C., & Murphy, D. M. (1991). Training interactional behaviors of adults with developmental disabilities: A systematic replication
and extension. Journal of Applied Behavior Analysis, 24, 167–174.
Leaf, J. B., Dotson, W., Oppenheim, M. L., Sheldon, J. B., & Sherman, J. A. (2010). The effectiveness of group teaching interactions for young children with a pervasive
developmental disorder. Research in Autism Spectrum Disorders, 4, 186–198.
 J.E. Nuernberger et al. / Research in Autism Spectrum Disorders 7 (2013) 411–417 417

Leaf, J. B., Taubman, M., Bloomfield, S., Palos-Rafuse, L., Leaf, R., McEachin, J., et al. (2009). Increasing social skills and pro-social behavior for three children
diagnosed with autism through the use of a teaching package. Research in Autism Spectrum Disorders, 3, 275–289.
Minkin, N., Braukmann, C. J., Minkin, B. L., Timbers, G. D., Timbers, B. J., Fixsen, D. L., et al. (1976). The social validation and training of conversational skills. Journal
of Applied Behavior Analysis, 9, 127–139.
Roessler, R. T., Brolin, D. E., & Johnson, J. M. (1990). Factors affecting employment success and quality: A one year follow-up of students in special education. Career
Development for Exceptional Individuals, 13, 95–107.
Schreiber, C. (2010). Social skills interventions for children with high-functioning autism spectrum disorder. Journal of Positive Behavior Interventions, 13, 49–62.
Stewart, K. K., Carr, J. E., & LeBlanc, L. A. (2007). Evaluation of family-implemented behavioral skills training for teaching social skills to a child with Asperger’s
Disorder. Clinical Case Studies, 6, 252–262.
Winner, M. G. (2005). Worksheets! For teaching social thinking and related skills. California: Think Social Publishing.