Social Tools And Rules for Teens (START) Program
The Social Tools And Rules for Teens (START) socialization program for adolescents with autism spectrum disorder (ASD) is a peer-facilitated intervention that combines experiential learning opportunities with an interactive social curriculum. It is a 20-session program consisting of weekly 90-min group meetings. Each meeting consists of an individual check-in session, a free socialization period, a social topic discussion, a structured social activity, and a checkout session. Structured and unstructured free socialization periods are embedded within each group session to provide a supportive, natural context for social experimentation, self-management, and skill building. The interactive didactic social lessons are intended to provide opportunities to promote social insight and effective social skill use. The START program was developed at the University of California, Santa Barbara.
By the nature of their diagnosis, adolescents with ASD often experience limited social success due to challenges with social skills, insight, and motivation (Orsmond et al. 2004). These limited social competencies can leave these individuals vulnerable to social rejection or exclusion (Schroeder et al. 2014), limiting their access to the peer interaction needed to actually improve their social skillset. Many adolescents with ASD experience lower perceived social status, friendship quality, and social motivation, along with greater levels of loneliness than typically developing peers (Locke et al. 2010; Mazurek 2013). Fortunately, there has been a recent increase in empirically supported social skill interventions that attempt to address these challenges and generate crucial social momentum (Miller et al. 2014).
A supportive network of peers is a crucial, yet often overlooked, component in socialization efforts. Fortunately, the use of peer-mediated interventions is growing in popularity (Watkins et al. 2015). Typically-developing peer models are valued for their ability to connect with same-aged participants and model, teach, and constructively evaluate social performance. They are also immersed in the same unique peer social culture as the target population and can provide much more ecologically appropriate instruction than a generic social skills curriculum taught by adult clinicians. Through these positive interactions, individuals with ASD also gain exposure to unconditional social acceptance – exchanges with receptive peers who are friendly, willing to converse, and forgiving of social missteps. Having positive experiences with peers allows adolescents with ASD to build social momentum that increases both confidence and competence to interact.
The initial pilot of the START program was developed to improve social competence in adolescents with ASD through the use of immersive experiential learning techniques (Vernon et al. 2016). This process of learning operates through exposure to and processing of real-world experience (Kolb 2014). A socialization intervention that incorporates experiential learning allows participants to engage with peers and experiment with social strategies firsthand, reflect on these experiences, and learn from their mistakes. Instructional social skills lessons also have value; however, many of the social nuances associated with live interactions are unable to be replicated when teaching a didactic lesson. Individuals often learn best by doing. This process is analogous to learning a sport or musical instrument. Some gains can be made through professional instruction, tutoring, and practice exercises, but one can only master a skillset through repeated live, real-world performances. By actively engaging in a naturalistic social setting with typically developing peers, individuals can benefit from the social complexities inherent in real-world interaction. Moreover, these active learning opportunities allow individuals to experience the natural incentives of positive peer interactions, such as discussion of favorite topics, responses to humor, and opportunities to jointly engage in enjoyable activities. By practicing social interactions in a safe, welcoming environment, participants can begin to associate social engagement with rewarding positive experiences and outcomes. The efficacy of the START program has been further established through the outcomes of a randomized clinical trial (Ko et al. 2018; Vernon et al. 2018).
Rationale or Underlying Theory
In order to better understand the design and conceptualized mechanism of change of the START program, it may be helpful to explore its theoretical underpinnings. The model is grounded in an experiential model of social learning. Experiential learning theory is conceptualized as a four-stage learning process (Kolb et al. 2001) in which (1) an experience serves as the stimulus for (2) reflective observation on what is functionally working or failing to work in the learning context, which (3) solidifies one’s abstract conceptualization of a given phenomenon. Based on this accrual of knowledge, the individual is able to engage in (4) active experimentation to hone one’s knowledge base and associated skill set, which is then applied to future experiences as the cycle repeats.
When applied to socialization and conversational competence, experiential learning is heavily reliant on sustained immersion in an appropriate social environment (Baker et al. 2002). Mapped onto the four-stage experiential learning process, interactions with others provide an experiential stimulus that allow individuals to reflect on their interpersonal successes and failures, improve their understanding of successful social strategies, and modify subsequent social bids to improve the likelihood of success in future engagements. For the general adolescent population, the accumulation of these interactions appears to be crucial for establishment of social competence and seems to be predicated upon two interrelated factors: (a) the willingness (or motivation) of the individual to engage with available social partners and (b) the willingness of these partners to be responsive to the social bids of that individual.
Goals and Objectives
The START program is designed to improve the social readiness skills of adolescents with ASD. Social readiness can be conceptualized as the combination of two essential components needed for optimal interpersonal functioning: social motivation and social competency. In turn, social competency can be further broken down into mastery of social insight and social skills (Vernon et al. 2016). The START socialization curriculum was intentionally designed to integrate both experiential and didactic training components within a single peer-facilitated intervention to target these components.
Social motivation is the desire to voluntarily engage in social pursuits and the anticipation that one will derive pleasure from such interactions, serving as a popular conceptualization of autism-related vulnerabilities (Chevallier et al. 2012; Dawson et al. 2005). Social motivation is a prerequisite that makes social immersion and experiential learning possible. At its foundation, the START program was intended to maximize the motivation of individuals who are interpersonally vulnerable. A safe, accepting space was fostered for practicing social skills without the fear of rejection or judgment. Group cohesion and unconditional acceptance were emphasized. By creating a supportive experiential context, the participants were granted the opportunity to become fully immersed and accepted by a peer group, which then became the ideal forum to practice and master critical social competencies. Over the course of 20 weeks, they accumulated frequent, reoccurring experiences of social success to establish both confidence and valuable social momentum.
Social insight is the understanding of underlying principles that govern successful social behavior, encompassing various concepts of social cognition (i.e., empathy, theory of mind, problem-solving capacities, and executive functioning skills; Schmidt et al. 2011). Several program components were specifically incorporated into the program to foster social insight. The check-in and checkout phases of each group provided a private forum to intentionally reflect on one’s social experiences and process perceived success and failure. Participants were also encouraged to think through role-plays and videos and describe the reasoning behind certain social decisions. Finally, during group topic discussions, the rationale underlying the use of certain social strategies was explored and explicit links were made to principles of empathy, courtesy, respect, and constructive problem solving.
Finally, social skills describe specific behaviors that are employed in the proper context at the proper time to facilitate a successful person-to-person exchange. The hands-on activities and self-management components of the START program facilitated the development and mastery of this concrete social skillset. Individual skills were modeled, practiced, and critiqued each session. Intentional practice redundancies were also used to promote eventual automaticity of skill use. Specifically, individualized target skills were practiced during check-in, self-managed during group, and assigned as homework in between group sessions. Similarly, weekly skills highlighted in the group curriculum were critically analyzed during role-play sessions and video examples, demonstrated, practiced in pairs, and assigned as weekly social homework.
The inclusion criteria for START program participation includes an age between 12 and 17 years, a diagnosis of autism spectrum disorder, a verbal IQ score above 70, and the ability to comprehend and speak in full sentences. To date, only an English language version of the START program has been evaluated. Because the program relies on the use of conversation, discussion of social issues, and self-reflection, it is likely not suitable for individuals with more limited language skills or intellectual impairment (although an adapted version of the START program for individuals with higher support needs is currently in development).
The START program consists of 20 weekly sessions that are 90 min in length. Each meeting consists of a 5-min individual check-in session, a 20-min free socialization period, a 40-min social topic discussion, a 20-min structured social activity, and a 5-min checkout session that parents are invited to join.
Prior to beginning the program, participants complete an initial 90-min intake session, which consists of obtaining consent/assent from parents and adolescents, obtaining basic demographic information, and completing all required intake measures. Trained undergraduate-level research assistants conducted all intakes and subsequent progress meetings. During the pre-intervention session, an assessment of individual social vulnerabilities was conducted. Participants and parents were provided with a list of common social skill difficulties/vulnerabilities and also had the option to write in additional concerns that were not listed (Vernon et al. 2016). Based on a rank order of specific social skill difficulties completed separately by each adolescent, their parent, and an intake clinician, consensus was reached on an individual social skill to serve as the initial focus of self-management. The primary objective was to directly target the social skill deficit that was identified as having the largest negative impact on each adolescent’s level of social success.
The primary facilitator discussed the target skill with the adolescent and they jointly practiced self-management procedures (monitoring and tracking one’s use of a particular social skill). Specifically, the skill was operationally defined for the participant, modeled by the social facilitator, and then practiced in a brief conversational exchange while the participant tracked their skill use using a small digital tally counter. After the participant could verbally describe the target social skill and successfully demonstrate accurate self-management in conversation, they were encouraged to continue self-managing use of this skill during each START group session. Individual goals were reexamined every 5 weeks and a new primary skill was introduced as participants either (a) demonstrated adequate mastery of a previous target skill or (b) exhibited a more significant challenge in another skill domain that was identified as a more notable source of social difficulty.
START Program Session Format
START program sessions: All of the adolescents began participation in the START program within 2 weeks of completing all pre-intervention measures. The 90-min weekly program consisted of the following phases: an individual therapeutic check-in session, a group unstructured socialization time, a structured group activity, a group discussion and practice of a social skill topic, and an individual checkout session with parent involvement. Participants, two to four college-aged social facilitators, and one to two high school peers attended each group.
Individual Check-in Session
Adolescents first completed an individual check-in session with a college-aged social facilitator. This time period was allotted to provide a private forum to discuss the previous week’s social challenges and successes, review social homework, practice self-management of the current individualized target behavior, and become oriented to the activities of the upcoming session. Facilitators used this time to prime individuals of session content and answer questions.
Free Socialization Phase
All participants joined the group’s free socialization time with the facilitators and high school peers. This time was allowed to unfold without a predetermined agenda and was intended to create a natural, comfortable social environment. Topics were brought up organically by the participants and often included video games, favorite foods and places to eat, school and current events, vacation and weekend plans, and memorable personal experiences. Food and refreshments were provided for each group to aid in the creation of a casual, club-like atmosphere. While conversing with one another, the participants and social facilitators discretely tracked their use of individual target skills using self-management. The high school peer models and the college-aged facilitators also participated in this self-management process to ensure that every group member was held to the same expectations and to minimize perceived differences between session attendees.
Social Topic Discussion and Practice
After the free socialization phase, the social facilitators then introduced the week’s social skill topic, which was explored for the next 40 min. After a brief introduction of the target skill, the topic was then modeled by the social facilitators in a series of two brief role-plays – one “bad” example demonstrating poor implementation of the skill and a follow-up “good” example depicting proper use of that particular skill. Popular television and movie clips were also used to highlight good and bad examples of the skill in question. Social facilitators discussed the main points of the topic and recalled relevant personal experiences. The adolescent participants were also encouraged to contribute to the topic – discussing their experiences related to that topic and providing their own suggestions regarding the successful use of a particular social skill. Finally, for the last 5 min of the group, all participants practiced the related skill with a partner. A manualized curriculum of key points and sample stories and scripts were used to structure and guide these discussions. This portion of the group was intended to increase understanding of a social skills topic and provide opportunities to both observe and practice the skill. Social skills topics covered included making introductions, maintaining a conversation, respectfully disagreeing, and group interactions.
Social Activity Phase
For the final 20 min, the group transitioned into a structured social activity. These activities varied each week, but generally resembled commonly used team-building activities and party games. Activities were also selected to be highly enjoyable and motivating to increase the engagement of the group participants. This phase was intended to foster sharing of personal information, encourage learning about peer interests, increase comfort in the group, and promote cooperation and teamwork. Other benefits included opportunities to work on effective communication, compromise, and good sportsmanship skills.
At the end of each group session, individual checkout sessions were conducted with each participant, a parent, and their assigned social facilitator. The participants discussed their experiences in the group, reviewed the group socialization topic with their parent, and set two weekly homework assignments – one based on their individual self-management target and the other based on the weekly group curriculum. Finally, a written summary of the current topic was provided to the family as a visual reference.
The efficacy of the START program has been evaluated through three peer-reviewed research articles. The first article summarized the results of pilot study with six participants (Vernon et al. 2016). The study used a combination of a multiple baseline and a pre-post design to establish preliminary evidence of efficacy by examining improvements on parent-report measures, adolescent self-report measures, and increases in coded social skill use on videotaped social conversation probes.
A randomized clinical trial was then conducted to evaluate the impact of the START model on social performance. All participants were randomly assigned to one of two groups: Those assigned to the START treatment group immediately received 20 weeks of the START program. Those assigned to the waitlist control group were permitted to continue any preexisting therapy efforts but were not exposed to the experimental socialization treatment package. Both groups were reevaluated after a 20-week time period.
The first START RCT article detailed the results of the parent and adolescent self-report social survey measures (Vernon et al. 2018). Results of the mixed MANOVA analyses revealed the presence of significant Group × Time effects for the social outcome variables (Wilks’ Lambda = 0.53; F(5,29) = 5.22, p = .002). Group × Time interaction reached statistical significance for three outcome measures: Parent SRS, F(1,33) = 7.68, p = .009, partial η2 = .189; Parent SMCS, F(1,33) = 13.29, p = .001, partial η2 = .287; and Adolescent SSIS, F(1,33) = 7.29, p = .010, partial η2 = .181. All parents endorsed social validity ratings indicating that their adolescent highly enjoyed being a part of the social skills group (mean rating of 8.14 out of 10, SD = 0.97). Additionally, they endorsed that their child’s social skills and competence improved through participation in the group (mean of 7.50 out of 10, SD = 0.82). The adolescent participants also indicated that they enjoyed their time in the groups (mean rating of 8.41 out of 10, SD = 1.83). Likewise, they endorsed that their social skills and competence improved through participation in the group (mean of 8.34 out of 10, SD = 1).
The next peer-reviewed publication examined real-world social skill use during live peer conversations (Ko et al. 2018). Three observable social behaviors were coded: questions asked, directed positive facial expressions, and mutual engagement. Results revealed a significant Group × Time interaction effect for Questions Asked, F(1, 33) = 4.86, p = .035. Participants in the treatment group significantly increased the percentage of questions asked from pre- to post-intervention, whereas the percentage of questions asked by the waitlist group remained relatively unchanged. Effect size was medium-large (partial η2 = .128). Additionally, analyses revealed the presence of a significant Group × Time effect on the percentage of Positive Facial Expressions, F(1, 33) = 7.73, p = .009. The treatment group demonstrated significant improvements in the percentage of directed positive facial expressions during social conversations compared to the waitlist group. The Effect size was large (partial η2 = .190). Group × Time interaction did not reach statistical significance for Mutual Engagement, F(1,33) = 0.32, p = .576.
As described above, the efficacy of the START program has been evaluated using standardized survey measures (parent report and adolescent self-report measures; Vernon et al. 2018) and behaviorally coded social skill use derived from live social conversations with unfamiliar peers (Ko et al. 2018).
Survey-Based Outcome Measures
Social Skills Improvement System Rating Scales (SSIS-RS): The SSIS-RS is a 75–83 item revised version of a widely used rating scale measuring several aspects of social skills, including communication, cooperation, assertion, responsibility, empathy, engagement, and self-control (Gresham and Elliott 2008). Internal consistency alpha reliability coefficients for the parent and self-report forms are reported to be in the mid to upper .90s, with moderate to high correlations to other social and behavioral scales.
Social Responsiveness Scale, Second Edition (SRS-2): The SRS-2 is a 65-item rating scale that covers various dimensions of interpersonal behavior, communication, and stereotypic behavior associated with ASD (Constantino and Gruber 2005). Internal consistency alpha reliability coefficients for the parent forms were reported to be above .90 and strong correlations (r = .52–.74) with subscales of the ADI-R. This measure was used as an indicator of ASD symptom severity, with score reductions associated with a decrease in observable symptoms.
Social Motivation & Competencies Scale (SMCS): The SMCS is an unpublished rating scale that was developed by the current researchers for use in this study. It was designed with corresponding parent and adolescent self-report versions. Items pertaining to comfort in social interaction, conversation skill use, empathy, friendships, appropriate behavior, social contact, and social interest were rated on a 5-point Likert scales. This measure was used as an indicator of social motivational factors and concrete skill competencies.
Social Validity Survey: Parents and participants were both asked to provide ratings about the acceptability of the START program. Specifically, they were asked to provide separate ratings on a 0–10 Likert scale on both (a) enjoyment of the adolescent’s time in the group and (b) the extent to which the adolescent’s social skills and competence improved as a result of participation.
Social Conversation Probe Procedures
Social conversation data were collected during the pre-intervention intake session and after program completion (i.e., after 20 weeks of the START program or waitlist participation). At these time points, participants engaged in 5-min conversations with unfamiliar peers whom they had never met. These peer conversational partners were not trained or coached prior to their interactions with the participants. They also possessed no knowledge of the purpose of the research study and did not know the diagnostic status of the participants. They were randomly assigned to participants and never completed more than one conversation with the same participant. They were provided with the following instructions: “You will be having a conversation with another person that you have never met before. You will have five minutes to get to know each other.”
Social Conversation Outcome Measures
Dependent measures were selected to provide representative data associated with improvements to the core social deficits of ASD (APA 2013). Video-recorded conversations, pre- and post-intervention, were systematically coded for each dependent measure. The three dependent measures coded were questions asked, positive facial expressions, and mutual engagement.
Questions Asked: Social inquiries have been identified as a crucial interpersonal strategy, but compared to typically developing adolescents, adolescents with ASD are known to make far fewer social initiations to their peers. A question was defined as a verbal query that is intended to elicit a response from the conversational partner. Trained coders, who were blind to the treatment status of the participants, counted each question asked in the video clips by both the participant and the conversational partner. In order to take into account the number of questions asked by the conversational partner, Questions Asked was defined as the percentage of questions asked by the participant during conversation.
Positive Facial Expressions: Individuals with ASD commonly have deficits in appropriate affective expression. Smiling and laughter have long been established as important nonverbal indicators of social attunement in conversation and are generally associated with friendliness and positive impressions. Nonverbal social engagement was measured through observed efforts to convey interest in a social partner’s conversation (positive facial expressions). A five-second partial interval coding scheme was implemented to code for the presence or absence of a positive facial expression (defined as visible smiling or laughing). A percentage was calculated to determine the percentage of time the participant was displaying positive facial expressions during each 5-min clip.
Mutual Engagement: Challenges with social reciprocity are a hallmark characteristic of ASD. To measure changes in reciprocity, mutual engagement, or the extent to which both social partners were jointly engaging in conversation, was examined to explore the balance of conversational contributions. This measure specifically examined the contributions of a participant relative to their conversational partners. Coders examined five-second intervals and focused on whether (a) the participant was the primary speaker, (b) the conversational partner was the primary speaker, (c) both partners contributed equally to the conversational volley, or (d) no one spoke. Mutual engagement was defined as the percentage of intervals in which both individuals contributed equally to the conversation.
Qualifications of Treatment Providers
Trained undergraduate research assistants served as the social facilitators for the START program. High school volunteers were also recruited from local schools to serve as peer mentors. All facilitators received an initial 10-h training on basic group facilitation techniques, covering basic group facilitation skills, methods for fostering rapport, and exposure to practice group sessions. The social facilitators also participated in weekly 1-h supervision meetings for ongoing clinical training purposes. One undergraduate was assigned as the designated primary social facilitator for each participant and was responsible for all check-in and checkout sessions, along with all progress meetings with that individual. Advanced clinical psychology doctoral students and a licensed clinical psychologist jointly conducted all training and supervision sessions.
References and Reading
- Baker, A. C., Jensen, P. J., & Kolb, D. A. (2002). Conversational learning: An experiential approach to knowledge creation. Portsmouth: Greenwood Publishing Group.Google Scholar
- Constantino, J., & Gruber, C. (2005). Social responsiveness scale. Los Angeles: Western Psychological Services.Google Scholar
- Gresham, F. M., & Elliott, S. N. (2008). Social Skills Improvement System (SSIS) rating scales. Minneapolis: Pearson.Google Scholar
- Ko, J. A., Miller, A. R., & Vernon, T. W. (2018). Social conversation skill improvements associated with the START program for adolescents with ASD: Results of a randomized controlled trial. Autism. Online First Publication.Google Scholar
- Kolb, D. A. (2014). Experiential learning: Experience as the source of learning and development (2nd ed.). Upper Saddle River: Pearson Education.Google Scholar
- Kolb, D. A., Boyatzis, R. E., & Mainemelis, C. (2001). Experiential learning theory: Previous research and new directions. In R. J. Sternberg & L. F. Zhang (Eds.), Perspectives on thinking, learning, and cognitive styles (pp. 227–247). Mahwah: Lawrence Erlbaum.Google Scholar
- Schroeder, J. H., Cappadocia, M. C., Bebko, J. M., Pepler, D. J., & Weiss, J. A. (2014). Shedding light on a pervasive problem: A review of research on bullying experiences among children with autism spectrum disorders. Journal of Autism and Developmental Disorders, 44(7), 1520–1534.CrossRefGoogle Scholar
- Vernon, T. W., Miller, A. R., Ko, J. A., Barrett, A., & McGarry, E. (2018). A randomized controlled trial of the Social Tools and Rules for Teens (START) program: An immersive socialization intervention for adolescents with ASD. Journal of Autism and Developmental Disorders, 48(3), 892–904.CrossRefGoogle Scholar
- Vernon, T. W., Miller, A. R., Ko, J. A., & Wu, V. (2016). Social Tools and Rules for Teens (The START program): Program description and preliminary outcomes of a multi-component socialization intervention for adolescents with autism spectrum disorder. Journal of Autism and Developmental Disorders, 46(5), 1806–1823.CrossRefGoogle Scholar