J Autism Dev Disord

DOI 10.1007/s10803-016-2715-7

ORIGINAL PAPER

Social Tools And Rules for Teens (The START Program):

Program Description and Preliminary Outcomes
of an Experiential Socialization Intervention for Adolescents
with Autism Spectrum Disorder
Ty W. Vernon1 • Amber R. Miller1,2 • Jordan A. Ko1 • Victoria L. Wu1,3

! Springer Science+Business Media New York 2016

Abstract Experiential learning is an essential process in Keywords Adolescents ! Autism spectrum disorders !
the development of core social competencies. Unfortu- Experiential learning ! Social skills training ! Social
nately, adolescents with autism spectrum disorders often do competence
not possess the prerequisite skillset and motivation to
sustain the level of social immersion needed to benefit from
this learning process. These persisting social vulnerabilities Introduction
can limit their long-term relational success and associated
quality of life, creating a need for comprehensive social Experiential learning is the process of learning through
programming. This paper describes a multi-component authentic experiences and subsequent reflection on these
socialization intervention that simultaneously targets encounters (Kolb 2014). This active learning immersion
motivational, conceptual, and skill deficits using a hybrid may offer unique benefits to participants, as they can make
experiential/didactic treatment approach. Evidence of discoveries and experiment with responses and strategies
social competence improvements was noted in survey and firsthand, instead of just passively benefitting from other
live conversational measures, indicating that the START individuals’ stories, experiences, and lessons. Additionally,
program may hold promise as a method for improving the individuals are exposed to the full range of dynamic vari-
social success of participating adolescents with ASD. ables and complexities that accompany a real-world
experience, most of which is inevitably lost when infor-
mation must be consolidated and summarized into a format
suitable for a didactic instructional lesson.
Experiential learning theory is conceptualized as a four
& Ty W. Vernon stage learning process (Kolb et al. 2001): (1) A concrete
[email protected] experience serves as the stimulus for (2) reflective obser-
Amber R. Miller vation on what is functionally working or failing to work in
[email protected] the learning context, which (3) solidifies one’s abstract
Jordan A. Ko conceptualization of a given phenomenon. Based on this
[email protected] accrual of knowledge, the individual is able to engage in
Victoria L. Wu (4) active experimentation to hone one’s knowledge base
[email protected] and associated skill set, which is then applied to future
1
Department of Counseling, Clinical & School Psychology,
concrete experiences.
Koegel Autism Center, University of California Santa When applied to socialization and conversational com-
Barbara, Santa Barbara, CA 93106-9490, USA petence, experiential learning is heavily reliant on sus-
2
Present Address: Children’s Hospital Los Angeles, tained immersion in an appropriate social environment
Los Angeles, CA, USA (Baker et al. 2002). Mapped onto the four stage experien-
3
Present Address: University of California San Diego Medical tial learning process, interactions with others provide an
School, San Diego, CA, USA experiential stimulus that allow individuals to reflect on

123
 J Autism Dev Disord

their interpersonal successes and failures, improve their recognition of limited social success relative to the general
understanding of successful social strategies, and modify population.
subsequent social bids to improve the likelihood of success In addition to the aforementioned motivational and
in future engagements. For the general adolescent popu- contextual socialization barriers facing the general ado-
lation, the accumulation of these interactions appears to be lescent population, individuals with ASD also experience a
crucial for establishment of social competence and seems myriad of other challenges, including limited social con-
to be predicated upon two interrelated factors: (a) the versation, theory of mind/empathy, abstraction, and agency
willingness (or motivation) of the individual to engage with detection skills (Baron-Cohen 1995; Bauminger 2002;
available social partners and (b) the willingness of these Blair et al. 2002; Starr et al. 2003). While many of these
partners to be responsive to the social bids of that vulnerabilities appear to be linked to the inherent neuro-
individual. biology of the disorder (DiCicco-Bloom et al. 2006), there
As an initial requirement, the individual must have ade- is also growing acceptance of the hypothesis that the initial
quate motivation to engage in social interaction. There must symptoms of autism contribute to more pronounced vul-
be a personally compelling rationale for social participation, nerabilities later in life due to the transactional nature of
whether it is a desire to forge new relationships, obtain development (e.g. Jones and Klin 2009). The ultimate
acceptance into a social club or group, and/or access some social outcomes of individuals with ASD appear to be a
extrinsic incentive or ulterior motive (Ojanen et al. 2010). confluence of their initial symptom profiles, the availability
Additionally, there must be confidence that the quality of the and sustainability of suitable social experiences, and the
individual’s interactions will result in a positive social quality of available intervention efforts.
exchange. Without a clear investment in the social outcome, Structured social skills intervention models have been
the individual will not have an adequate incentive to fully growing in popularity as a means to address the core area
engage with their social partners and benefit from the natural of socialization difficulties of adolescents with ASD. There
lessons inherent within the interaction. is a growing body of evidence supporting the effectiveness
Likewise, peer acceptance is another pre-requisite for of these group socialization efforts (Miller et al. 2014;
experiential social learning. Without a group willing to National Autism Center 2009; Reichow et al. 2012).
actively engage the individual, he or she is left without a However, there is currently very limited research available
proper social context to benefit from associated learning on the potential benefits of experiential social learning
opportunities (Ollendick et al. 1992). Herein lies the within a social skills intervention context.
cyclical nature of the problem: individuals with the most Many existing social skills intervention programs focus
significant social vulnerabilities are frequently ignored or primarily on didactic lessons. These approaches rely on
even actively rejected by their peers, preventing them from systematic instructional methods to address many of the
accessing the very social context needed to enhance their social vulnerabilities associated with ASD (e.g. conversa-
skillset to the minimum threshold to prevent future rejec- tion skills, humor, perspective-taking, empathy, etc.) with a
tion (Parker and Asher 1987). Thus, such individuals are controlled follow-up opportunity to practice the skill of
left without an adequate means to obtain this crucial social interest (e.g., Laugeson et al. 2012; Ozonoff and Miller
traction unless both individual and contextual factors are 1995; Webb et al. 2004; White et al. 2010). Opportunities
targeted simultaneously. This dilemma may be particularly for experiential learning remain relatively limited and are
relevant for the significant social vulnerabilities inherent commonly relegated to social homework assignments in
among adolescents with autism spectrum disorders (ASD). school or community settings (outside of the structured
Adolescents with ASD often have a skill set that is intervention context). While this instructional format has
inadequate for the increasingly complex social situations yielded very promising evidence of therapeutic effect, it
they face as young adults (Gutstein and Whitney 2002; does not currently offer embedded experiential components
Hendricks and Wehman 2009). Even high-functioning that might address other key barriers, such as limited
individuals are often at odds with this frequently used motivation or peer acceptance challenges, which may
descriptive label. As adolescents with ASD transition to prevent or limit access to practice opportunities within
adulthood, their social vulnerabilities persist and the natural social contexts. In other words, adolescents may not
majority continue to live with their parents or other care- be able to practice a newly learned social skill with a peer
givers, remain unemployed or underemployed, and rely on group that actively rejects or ignores them.
the financial support of others (Engström et al. 2003; There may be unexplored benefits of a social instruction
Howlin 2000). Additionally, this population is at signifi- model that places equal emphasis on opportunities for
cantly higher risk for developing depression or anxiety dynamic, natural interactions within a peer group (experi-
(Church et al. 2000; Ghaziuddin et al. 2002; Kim et al. ential learning) along with more commonly used didactic
2000), as greater cognitive awareness also extends to the learning approaches. This ‘‘learn by doing’’ augmentation

123
J Autism Dev Disord

may be crucial to the development of social competency, as Method

it may expose participants to sufficient opportunities to
both (a) feel included and accepted by a group of peers and This project received Institutional Review Board (IRB)
(b) use this forum to become fully immersed in a dynamic, approval and was conducted at a North American
reflective learning process (i.e., experience both success university.
and failure within a forgiving social environment, process
these experiences, further develop one’s abstract concep- Participants
tualization of social understanding, and return to this
context to experiment with more effective strategies). This A total of six adolescents participated in the initial inves-
therapeutic arrangement provides a seamless generalization tigation of the START program. Participants were ado-
opportunity while also creating unique opportunities to lescents aged 12–17 with an existing diagnosis of ASD
embed learning strategies that promote self-awareness and meeting current DSM-5 criteria. Inclusion criteria also
reflection into a social learning program. This may be included (a) use of full sentence phrases to communicate
especially important given the concerns about sustainabil- (fluent language use), (b) a verbal IQ equal to or greater
ity and generalizability in autism intervention efforts, than 70, and (c) an ASD diagnosis confirmed by meeting
which may be linked to the limited metacognitive capaci- cut-offs designated by the Autism Diagnostic Observation
ties of individuals with ASD (Grainger et al. 2014; Wil- Schedule, Second Edition (Lord et al. 2012), Autism
liams et al. 2009). Specifically, these individuals may Diagnostic Interview, Revised (Rutter et al. 2003), and the
struggle with generalized application of social principles Social Responsiveness Scale, Second Edition (Constantino
due to limited self-awareness and reflection skills. and Gruber 2005).
As a promising method for increasing self-awareness and Participants were recruited from several sources,
reflection, self-management is a relatively discrete inter- including communications with local high schools, refer-
vention technology that can be easily embedded within an rals from community organizations, word of mouth, and
experiential context. These methods have received consid- responses from online research announcements. All
erable research attention for transferring monitoring potential participants were scheduled for intake appoint-
responsibilities to the actual person engaging a desired target ments to ensure that they met all inclusion criteria and
skill (see Lee et al. 2007 for a review and meta-analysis). complete pre-intervention measures. A summary of par-
Within this context, individuals are taught to track their own ticipant demographic information is provided in Table 1.
behavior in an effort to promote awareness and internalized
skill acquisition. These methods have been used in previous Design
investigations to specifically target social skills (e.g., Dog-
gett et al. 2013; Koegel et al. 1992; Newman and Ten Eyck For this preliminary investigation, a combination of a
2005) along with self-regulation and mindfulness (Carr et al. clinical case series and a multiple baseline design across
2014). Self-management fosters a sense of intentionality in participants and behaviors was employed to understand the
one’s actions and allows for guided experimentation with a social improvement trajectories of participants as they
new set of social interaction strategies. In addition to pro- progressed through the START program. Initially, the
moting opportunities to engage in reflective practice, another participants completed all dependent measures during two
advantage of self-management is that adolescents partici- pre-intervention sessions spaced 5 weeks apart to assess for
pating in a group intervention can self-manage different the stability of social measures in the absence of any social
social skills—each reflective of a slightly different vulner- intervention efforts. Following the onset of intervention,
ability profile. In other words, it is possible to individualize they were reassessed with all measures after completion of
an aspect of the intervention package even within the context 5, 10, 15, and 20 (final) session time points during separate
of a group-based program. This promising monitoring progress meetings.
strategy is explored as an experiential technique within the
context of the current investigation. Procedure
The goal of the current study was to evaluate a hybrid
socialization intervention approach that combines experi- Social Facilitator Training
ential and didactic components into a single multi-compo-
nent treatment model. Specifically, a program description High-achieving college students were recruited from uni-
and preliminary outcomes will be provided for the Social versity undergraduate psychology courses to serve as social
Tools And Rules for Teens (START) program, a compre- facilitators for the START program. These volunteers
hensive social skills intervention for adolescents with ASD. received an initial 10-h training on basic group facilitation

123
 J Autism Dev Disord

Peer groups, Psychiatric Consults

Peer Mentor, Monthly Psychiatry
techniques, covering basic group facilitation skills, meth-

Counseling, Monthly Psychiatry

ods for fostering rapport, and exposure to practice group
sessions. The social facilitators also participated in weekly
1-h supervision meetings for ongoing clinical training

Educational Therapy
Counseling, Respite
purposes. An advanced clinical psychology doctoral stu-
Current services

dent and a licensed clinical psychologist jointly conducted

Consults all training and supervision sessions.

Consults
None

Pre-Intervention Sessions
Grade

All participants completed an initial 90-min intake session,

10
11

10
8

which consisted of obtaining consent/assent from parents

and adolescents, obtaining basic demographic information,
Regular Education Public Junior High School

and completing all required measures. After a five-week

Home School/Online Educational Program

Home School/Online Educational Program

waiting period, participants returned to re-complete these

Regular Education Public High School
Regular Education Public High School

Regular Education Public High School

measures in a second pre-intervention session. Trained

research assistants conducted all intakes and subsequent
progress meetings.

Target Skill Selection and Self-Management

During the initial pre-intervention session, an assessment

School setting

of individual social vulnerabilities was conducted. Partic-

ipants were provided with a list of 12 common social skill
difficulties/vulnerabilities with the option to write in
additional concerns that were not listed. This list was
FSIQ

generated internally by the research team and included

99
96
81

98
86
127

commonly observed social skill deficits among the clinical

population: asking relevant (on-topic) questions, making
Nonverbal
IQ/Index

relevant comments, increasing conversational bids, limit-

89
96
88

85
129
115

ing conversational contributions (over-sharing), greeting

others, ending conversations appropriately, using voice
IQ/Index

qualities (volume, rate, intonation), making eye contact,

Verbal

directing appropriate facial expressions, expressing

108

117
97
79

85
88

empathy, choosing appropriate (positive) topics of con-

ASD, OCD, Depression

versation, using humor, and complimenting others. Based

on a rank order of specific social skill difficulties com-
Clinical diagnoses

pleted separately by each adolescent, their parent, and an

ASD, ADHD

intake clinician, consensus was reached on an individual

social skill to serve as the initial focus of self-management.
Table 1 Participant demographic information

The primary objective was to directly target the social skill

ASD
ASD

ASD
ASD

deficit that was identified as having the largest negative

impact on each adolescent’s level of social success. In this
14yo Caucasian Female

13yo Caucasian Female

15yo Caucasian Female
15yo Caucasian Male

preliminary investigation, a total of four different skill

16yo Biracial Male
Age, ethnicity, sex

14yo Latino Male

targets were targeted among the six participants.

During the second pre-intervention meeting, the
assigned facilitator reintroduced the selected social skill to
the adolescent and they jointly discussed the value of that
particular skill in the process of forming and maintaining
relationships. The skill was operationally defined for the
Participant

participant, modeled by the social facilitator, and then

practiced in a brief conversational exchange. After the
participant verbally expressed a strong understanding of
1
2
3

4
5
6

123
J Autism Dev Disord

the target social skill and successfully demonstrated accu- certain strategies make them a more desirable social part-
rate self-management in conversation (i.e. using a small ner and (b) engage in active experimentation during the
digital tally counter to record using the skill appropriately), group (e.g. make different types of social inquiries or tell
they were subsequently encouraged to self-manage their someone a different personal anecdote). After completing
use of this skill during each START group session. Indi- the individual check-in sessions, all participants were
vidual goals were reevaluated every 5 weeks and new brought together to start the group phases.
skills were introduced as participants either (a) demon- Unstructured Socialization Phase. The first 20 min of
strated adequate mastery of a previous target skill or the group consisted of unstructured socialization time.
(b) exhibited a more significant challenge in another skill When mapped onto the experiential learning model, this
domain that was identified as a greater source of social time period served as the concrete social experience to
difficulty. facilitate reflective observation. This unstructured time was
allowed to unfold without a predetermined agenda and was
START Program Sessions intended to create a natural, comfortable social context in
which to engage in both reflective observation and active
All of the adolescents began participation in the START experimentation. Topics were introduced by the partici-
program immediately following completion of their second pants and often included video games, favorite foods and
set of pre-intervention measures. The 2-h weekly program places to eat, school and current events, vacation and
consisted of the following components: an individual weekend plans, and memorable personal experiences. Food
therapeutic check-in session, a group unstructured social- and refreshments were provided during each group to aid in
ization time, a structured group activity, a group discussion the creation of a casual, club-like atmosphere. While
and practice of a social skill topic, and an individual conversing with one another, the participants and social
checkout session with parent involvement. All participants facilitators simultaneously tracked their use of individual
and 3-5 social facilitators took part in each group. The target skills. Social facilitators also participated in this
structure of each session is depicted in Fig. 1. process to ensure that every group member adhered to
Check-In Session. The 5-min individual check-in session identical program expectations and stigma was minimized.
was run by an individually assigned social facilitator and It was explained that everyone can improve the intention-
took place in a small clinic room. Within an experiential ality of their social actions, and the social facilitators
learning model, these sessions provided an opportunity for modeled this assertion by tracking their own use of a
the participants to reflect on their social experiences and personally identified skill of interest (e.g. questions, com-
social homework objectives of the previous week. To pliments, positive statements, expressions of empathy,
concretize this process, the participants first rated their etc.).
perceived level of success in using their individual target Group Activity Phase. After approximately 20 min, the
skill and their general level of social comfort over the past group transitioned into a structured social activity. These
week using a 5-point Likert scale. These ratings were then activities varied each week, but generally resembled
used to facilitate private discussion and reflection between commonly used team-building activities and party games.
participants and their facilitators on their perceived social This phase was intended to foster sharing of personal
difficulties and successes. As a method to encourage par- information, encourage learning about peer interests,
ticipants to engage in abstract conceptualization of the increase comfort in the group context, and promote coop-
previous week’s social topic, participants were encouraged eration and teamwork in a highly structured context.
to focus on a particularly salient social event and connect Examples of activities included Social Bingo, Apples to
their use of specific skills to the ultimate outcome of the Apples, Bowl of Nouns, and Headbands. Activities served
interaction. The check-in sessions were also used to prime as additional experiences for reflective observation and
the participant with a brief description of the session social experimentation. They were selected to be highly
activities and practice self-management in a brief conver- enjoyable and intrinsically motivating to increase the
sation with the clinician. Finally, participants were engagement of the group participants. Collateral benefits
reminded to (a) monitor social group reactions to their use included opportunities to work on effective communica-
of their target skill to solidify their understanding of why tion, compromise, teamwork, and sportsmanship skills.

Fig. 1 START program session

Unstructured Interactive
timeline Individual Individual
Socialization Structured Social Topic Check-Out
Check-In Time w/ Self- Social Activity Discussion Session
Session (5 mins) Management (20 minutes) (20 Minutes) (5 mins)
(20 minutes)

123
 J Autism Dev Disord

Topic Discussion and Practice Phase. After the struc- and group interactions. A complete list of the topic cur-
tured activity phase, individuals were asked to share their riculum is depicted in Table 2.
reflections on the previous week’s topic that were origi- Check-Out Session. Finally, a 5-min individual check-
nally discussed during the check-in session. Common out session was conducted with each participant, one par-
themes and challenges were identified and discussed as a ent, and a social facilitator. Participants rated their level of
group to promote deeper understanding of each social comfort (a) using their identified target skill and (b) inter-
concept. The social facilitators then introduced the week’s acting with the group using separate 1-5 Likert scales.
social skill topic, which was discussed for the remaining They were then asked to use these ratings as a means to
time of the group session. After a brief introduction of the reflect back on their experience and form hypotheses about
target skill, the topic was then modeled by the social why they felt that the group went positively or negatively.
facilitators in a series of two brief contrasting role-plays— This process was intended to help the participants review
one ‘‘bad’’ example demonstrating poor implementation of their interactions in an intentional, systematic way to
the skill and a follow-up ‘‘good’’ example depicting proper ensure that they could fully benefit from the experiential
use of that particular skill. Social facilitators then discussed learning process. Each participant was then given person-
pre-determined topic summary items and illustrated them alized feedback about aspects of their group participation
by describing relevant personal experiences. The adoles- that the facilitator felt was exemplary, along with areas that
cent participants were then encouraged to contribute to the may have warranted improvement. This process was
topic—discussing their experiences related to that topic intended to model the abstract conceptualization process by
and providing relevant suggestions regarding the successful connecting specific participant behaviors to reactions from
use of a particular social skill. Finally, for the last 5 min of their peer group. This discussion was intended to establish
the group, all participants practiced the related skill with a general guiding concepts for future socialization experi-
partner. A manualized curriculum of key points and sample ences. For example, the facilitator might point out how a
stories/scripts were used to structure and guide these dis- participant’s harsh criticism caused a peer to become vis-
cussions. This portion of the group was intended to pro- ibly upset in the group, or how a fellow peer smiled when
mote joint understanding of a social skills topic and the participant used a compliment appropriately. The par-
provide opportunities to both observe and practice the skill. ticipants were then encouraged to discuss the group social
Social skills topics covered included: making introduc- skills topic with their parents to foster mutual under-
tions, maintaining a conversation, respectfully disagreeing, standing and continued use of the skill in their home

Table 2 START weekly group

Week Topic
discussion topics
1 Greeting others/making initial introductions
2 Using questions in conversation
3 Using comments in conversation
4 Showing interest—attention, eye contact, facial expressions
5 Choosing appropriate topics for conversations
6 Making and keeping friends
7 Changing topics/ending conversations/saying goodbye
8 Reducing anxiety/being comfortable during social exchanges
9 Expressing empathy
10 Complimenting others
11 Making a good impression/giving social feedback
12 Receiving social feedback
13 Respectfully disagreeing with others
14 Demonstrating good sportsmanship/being a good winner and loser
15 Working in a group/being a good team member/leader
16 Using appropriate humor and understanding/using sarcasm
17 Having social courage/joining a new group of peers
18 Using social media
19 Hosting others at your home/being a good guest at someone’s home
20 Summary of group topics and conclusion

123
J Autism Dev Disord

environment. As a final step, the social facilitator, partici- dimensions of interpersonal behavior, communication, and
pant, and parent(s) jointly established two social skills stereotypic behavior associated with ASD (Constantino and
homework goals to work on during the following week— Gruber 2005). Score reductions are associated with a
one based on the individual’s self-management goal and decrease in the severity of ASD symptoms. Internal con-
one based on the weekly topic discussed in group. Take- sistency alpha reliability coefficients for the parent forms
home sheets were provided to allow participants and par- were reported to be above 0.90 and strong correlations were
ents to track their completion of the homework objectives reported (r = 0.52–0.74) with subscales of the ADI-R.
each week. This process ensured that participants contin- Social Motivation & Competencies Scale (SMCS). The
ued to seek out additional experiential learning opportu- SMCS is unpublished rating scale that was developed by
nities in their natural social settings so that they could the current researchers for use in this study. Separate parent
process them during the next session. and adolescent self-report versions were developed that
contained identical item content. Items pertaining to com-
Fidelity of Implementation fort in social interaction, conversation skill use, empathy,
friendships, appropriate behavior, social contact, and social
While sessions were not videotaped, a fidelity checklist interest are rated on 1-5 Likert scales. See ‘‘Appendix 1’’ to
was completed during each group session to monitor review the SMCS survey items.
adherence to the established intervention protocol. Session
fidelity fell consistently above 95 %. As an additional Dynamic Conversation Probe Data
treatment integrity safeguard, project investigators also sat
in on random group sessions to observe. This investigation also employed the use of live conver-
sations to assess dynamic social skill changes on a variety
Data Collection of individualized social target skills. During the two pre-
intervention and four intervention progress meetings, par-
In addition to the two pre-intervention sessions, social ticipants engaged in two separate 5-min (600 s) ‘‘get to
skills data were also collected during progress meeting know you’’ conversations with both an unfamiliar male
sessions held every 5 weeks between regularly scheduled peer and an unfamiliar female peer. These peers were
group meetings (i.e. between the 5th and 6th sessions, 10th unaffiliated with the research project and were naı̈ve to the
and 11th sessions, 15th and 16th session, and after the final project aims or participant’s social disability status.
20th session). Participants and parents completed all survey Dynamic social data were obtained through the systematic
and conversational probe measures described below. coding of these video-recorded conversations, and data
from both conversations recorded on the same day were
Dependent Measures combined to create a mean score for that progress meeting.
Trained research assistants that were naı̈ve to the
Parent and Adolescent Social Survey Measures hypotheses of the study and treatment aims coded all
conversation videos in random order. Individualized target
A variety of self- and parent-reported survey instruments skill measures fell into four categories:
were administered to assess changes to social competence: Increasing Social Inquiries. Question asking has been
Social Skills Improvement System Rating Scales (SSIS- described in the literature as a crucial social skill (Doggett
RS). The SSIS-RS is a 75-83 item revised version of a et al. 2013; Palmen et al. 2008; Weiss and Harris 2001). A
widely-used rating scale measuring several aspects of frequency count of social inquiries made in the video clips
social skills, including Communication, Cooperation, was conducted for both the participant and the conversa-
Assertion, Responsibility, Empathy, Engagement, and Self- tional partner. A social inquiry was defined as verbal
Control (Gresham and Elliott 2008). Internal consistency question that was clearly intended to elicit a verbal
alpha reliability coefficients for the parent and self-report response from a conversational partner. Both number of
forms are reported to be in the mid to upper .90 s, with social inquiries and the percentage of the total inquiries
moderate to high correlations to corresponding social and made were calculated. The ultimate objective of this target
behavioral scales on the Behavior Assessment System for skill was to increase participant social inquiries so that a
Children, Second Edition and the Vineland Adaptive roughly equal number of questions (50 %) was exchanged
Behavior Scales, Second Edition. Both parent and adoles- between them and their conversational partners, which
cent-report surveys were completed during each progress would be an indication of a balanced, reciprocal conver-
meeting. sational exchange. Inter-rater reliability for social inquiries
Social Responsiveness Scale, Second Edition (SRS-2). was calculated for approximately 33 % of videos and
The SRS-2 a 65 item rating scale that covers various resulted in a mean of 88 % agreement (range of 82–96 %).

123
 J Autism Dev Disord

Decreasing Negative Statements. Individuals with ASD separate ratings on a 0-10 Likert scale on both (a) enjoy-
often have limited understanding of socially appropriate ment of the adolescent’s time in the group and (b) the
topics of conversation and offer overly honest opinions extent to which the adolescent’s social skills/competence
about subject matter without filtering their messages improved as a result of participation. The specific questions
(Landa 2000). Several individuals in this study struggled to provided to participant and parent were as follows:
engage in primarily positive topics of conversation.
On a scale from 0 to 10, how much did you enjoy (for
Therefore, goals were set to increase the number of posi-
parents: do you think your child enjoyed) being a part
tive and neutral statements while simultaneously decreas-
of the social club?
ing the occurrence of negative statements. A negative
statement was defined as a critical judgment or unfavorable On a scale from 0 to 10, how much did your (for
comment about oneself or another person, topic, or event. parent: your child’s) social skills/competence
Twenty-second intervals were coded by raters as focusing improve through participation in the club?
primarily on (a) negative or (b) positive/neutral subjective
content. A percentage was then calculated to determine the
ratio of intervals in the conversation that were dominated Results
by negative comments. Inter-rater reliability for negative
statements yielded a mean of 90 % agreement (range of Parent and Adolescent Social Survey Measures
81–100 %).
Increasing Verbal Contributions. The successful use of Survey data and associated effect sizes for all participants
social conversational bids is associated with higher levels are summarized in Table 2.
of peer acceptance (Burleson et al. 1986), while interper-
sonally reserved individuals generally possess a lower level SSIS-RS
of sociometric status (Newcomb et al. 1993). Participant 5
was noted to speak very infrequently during pre-interven- Parent-reported SSIS-RS Social Skills standard scores were
tion conversations and was encouraged to increase the noted to yield consistently positive score increases (mean
frequency of verbal contributions. During five-second of 5.3, range of 2.25–8.25) when comparing the mean of
intervals, raters coded whether or not the participant made the final two intervention progress meetings to the mean of
a verbal contribution of more than two words. This cut-off the two pre-intervention scores over the course of the
ensured that a participant who habitually provided only project, with four exceeding the SSIS-RS’s Standard Error
minimal verbal responses to questions (e.g. responding of Measurement (SEM = 3).
‘‘yes’’ or ‘‘not really’’) was not coded as providing appro- For the adolescent self-report SSIS-RS Social Skills, a
priate verbal contributions to a conversation. A percentage mean comparison from the final two progress meeting
was then calculated to determine the total percentage of scores to both pre-intervention scores indicated that four
time participants were speaking during each 5-min clip. participants (1, 2, 4, 5) endorsed general gains to their
Decreasing Verbal Contributions. Alternatively, an Social Skills standard score (exceeding the measure’s
excessive, pedantic speaking style is also observed in select SEM) while two (2, 6) endorsed a decrease in total score.
subgroups of individuals with ASD (Ghaziuddin and Ger-
stein 1996), particularly during discussions of highly pre-
SRS-2
ferred thematic content (Elder et al. 2006; Nadig et al.
2010). Participant 4 was encouraged to work on limiting
When comparing pre-intervention and intervention means
her excessive level of verbal information in order to allow
for the SRS-2, five participants (1, 2, 3, 5, 6) experienced a
her conversational partners more opportunities to partici-
reduction in their total score that exceeded the SRS-2’s
pate in discussions. An identical verbal contribution coding
SEM of 2.2, which is indicative of the desired decrease in
definition (described above) was implemented using five-
their endorsement of autism-related symptoms. The mean
second intervals. Inter-rater reliability for both verbal
change in SRS-2 score was -4.3 (range of -11 to 3). One
contribution categories yielded a mean of 92 % agreement
participant (6) experienced a negligible change, and one (2)
(range of 82–100 %).
experienced an increase in score.
Social Validity Ratings
SMCS
Parents and participants were both asked to provide ratings
to provide information about the acceptability of the For the parent-reported SMCS scores, a comparison of the
START program. Specifically, they were asked to provide mean of participant pre-intervention scores to the final

123
J Autism Dev Disord

intervention progress meeting scores were indicative of individualized target social skills (with medium to large
social increases (mean of 9.1 point increase with a range of effect sizes). Results are described for each individual
3.0–34.0 points). participant below (Table 3).
A comparison of the pre-intervention mean to the pro-
gress meeting scores demonstrated Four participants (1, 3, Participant 1: Increase Social Inquiries and Decrease
4, 5) endorsed gains in their adolescent self-report SMCS Negative Statements
total scores, while two (2, 6) experienced a decrease in
their mean score. The mean change in score across all Participant 1 exhibited consistently low levels of question
participants was 10.7 (range of -10.5 to 28). asking prior to beginning the intervention, asking a mean
of 5 questions (range of 4–9) during the 10-min pre-in-
Dynamic Conversation Probe Data tervention probes, representing a mean of 17.8 % (range of
11.6–24.0 %) of the total questions asked during her con-
Dynamic conversation probe data is summarized below versations. This skill was actively targeted in the first
and is generally reflective of improvements to 5 weeks of the START program, and Participant 1

Table 3 Parent and adolescent-report social survey scores

Survey measure Adol Pre Pre Pre- Week 5 Week Week Week Intervention D Effect
1 2 intervention Progress 10 15 20 mean (SD) Mean size
mean (SD) Progress Progress Progress score

SSIS-RS—parent rating 1 47 58 52.5 (7.78) 54 65 62 62 62 (0.00) 9.5 2.44

social skills standard 2 83 78 80.5 (3.54) 80 84 93 85 89 (5.66) 8.5 1.94
score
3 89 92 90.5 (2.12) 95 101 83 105 94 (15.56) 3.5 0.32
4 84 84 84 (0.00) 84 95 92 80 86 (8.49) 2 0.33
5 81 67 74 (9.90) 74 78 77 84 80.5 (4.95) 6.5 1.07
6 76 76 76 (0.00) 81 76 81 75 78 (4.24) 2 0.67
SSIS-RS—adolescent 1 86 84 85 (1.41) 103 106 102 103 102.5 (0.71) 17.5 20.21
rating social skills 2 77 91 84 (9.90) 79 82 89 86 87.5 (2.12) 3.5 0.68
standard score
3 100 100 100 (0.00) 98 95 92 85 88.5 (4.95) -11.5 -3.29
4 115 115 115 (0.00) 127 127 128 128 128 (0.00) 13.5 27.00
5 94 109 101.5 (10.61) 119 121 123 124 123.5 (0.71) 22 4.13
6 78 68 73 (7.07) 89 72 67 65 66 (1.41) -7 -1.91
SRS—parent total 1 90 90 90 (0.00) 90 90 77 88 82.5 (7.78) -7.5 -1.36
standard score (Lower 2 74 73 73.5 (0.71) 70 69 67 70 68.5 (2.12) -5 -3.24
scores indicate
3 71 78 74.5 (4.95) 69 61 85 57 71 (19.80) -3.5 -0.25
improvement)
4 65 65 65 (0.00) 64 68 68 68 68 (0.00) 3 –*
5 79 81 80 (1.41) 70 72 72 66 69 (4.24) -11 -3.57
6 71 70 70.5 (0.71) 70 74 68 69 68.5 (0.71) -2 -3.27
SMCS—parent rating 1 48 51 49.5 (2.12) 65 77 83 109 96 (18.38) 12.5 11.79
total raw score 2 69 77 73 (5.66) 73 78 79 76 77.5 (2.12) 16 3.27
3 92 87 89.5 (3.54) 94 99 70 107 88.5 (26.16) 4.5 0.40
4 81 81 81 (0.00) 85 90 96 92 94 (2.83) 5 0.83
5 67 65 66 (1.41) 76 80 81 90 85.5 (6.36) 14.5 4.06
6 79 73 76 (4.24) 85 78 84 74 79 (7.07) 2 0.54
SMCS—adolescent 1 97 95 96 (1.41) 116 118 115 119 117 (2.83) 21 13.14
rating total raw score 2 84 81 82.5 (2.12) 63 68 81 81 81 (0.00) -1.5 -1.23
3 102 102 102 (0.00) 101 100 151 104 127.5 (33.23) 25.5 0.59
4 120 120 120 (0.00) 128 128 128 127 127.5 (0.71) 7.5 18.98
5 83 89 86 (4.24) 105 118 109 119 114 (7.07) 28 4.57
6 107 105 106 (1.41) 99 105 82 109 95.5 (19.09) -10.5 -0.74
* Could not be calculated due to baseline and intervention SD of 0.00

123
 J Autism Dev Disord

increased her number of social inquiries to 10 and asked Participant 2: Increase Social Inquiries and Decrease
48.7 % of the total questions in her conversation probes. Negative Statements
After her target skill changed and social inquiries were no
longer her active target (reflected in progress meetings Participant 2 also exhibited low initial levels of social
2–4), she continued to ask a mean of 39.1 % of the total inquiries (mean of 2.3, range of 1-4), constituting a mean of
questions asked (range of 20.0–46.9 %), consisting of a 7.7 % (range of 4.6–10.7 %) of total questions asked per
mean of 12 questions per conversation (range of 8–15), conversation. After targeting this skill for the first five ses-
although a decreasing trend was noted. The effect size for sions prior to progress meeting 1, his percentage of questions
increases in her social inquiries overall was large asked increased to 18.5 % (5 total questions per conversa-
(d = 1.88). tion). At this timepoint, Participant 2’s level of negativity
Once she demonstrated initial proficiency with use of was determined to be having a greater impact on his social
social inquiries at progress meeting 1, Participant 1’s goal success than his number of social inquiries. After this change
was changed to address the amount of negative statements to focus on negative statements, Participant 2’s number and
she made. She was asked to decrease her negative state- percentage of social inquiries continued to improve during
ments by focusing on increasing her use of positive and the post-target phase (progress meetings 2–4), increasing to a
neutral statements. During the baseline phase for this new mean of 26.4 % (range of 15.7–44.4 %) with a mean of 5.5
target behavior (which included both pre-intervention ses- total questions (range of 3-10). There was a notable in-
sions and the first progress meeting), Participant 1 was creasing trend in the data, even though making social
found to discuss negative topics a mean of 30.4 % of inquiries was no longer the active target skill. The effect size
conversational intervals (range of 24.0–40.5 %). Her neg- for Participant 2’s improvement in making social inquiries
ative statements initially remained high at 35.0 % the first was large, d = 1.51.
five weeks of actively targeting this behavior and subse- Despite notable reliance on negative conversational
quently decreased at 10 and 15 weeks of intervention to a strategies during group, Participant 2 only engaged in neg-
mean of 17.3 % (range of 8.5–11.0 %). The effect size for ative statements for a mean of 4.8 % (range of 3.0–6.0 %) of
negative statements was also large, d = -1.07 (with the conversational intervals while conversing with novel social
negative value indicative of the desired reduction in neg- partners (possibly due to his need to ‘‘be polite’’ with unfa-
ative statements). Participant 1’s data is depicted in Fig. 2. miliar individuals). During the active target phase (progress

Fig. 2 Participant 1’s dynamic Pre-target phase Active target phase Post-target phase
conversation target social goals:
increase social inquiries and 80
decrease negative statements
Percentage of social inquiries

70
60
made by participant

50
40
30
20
10
0

80
Percentage of intervals with
negative statements/topics

70
60
50
40
30
20
10
0
Pre 1 Pre 2 1 2 3 4

123
J Autism Dev Disord

Fig. 3 Participant 2’s dynamic Pre-target phase Active target phase Post-target phase
conversation target social goals:
increase social inquiries and

Percentage of social inquiries

decrease negative statements 80
70

made by participant
60
50
40
30
20
10
0

80
Percentage of intervals with
negative statements/topics
70
60
50
40
30
20
10
0
Pre 1 Pre 2 1 2 3 4

meetings 2-4), his percentage of negative statements initially and during the active target phase ranged from 5 to 11.5 %
increased to 9.0 % but then decreased to 0 % for the (a mean of 7.5 %). The calculated effect size is indicative
remaining 2 progess meetings (overall mean of 3.0 %). The of a medium effect size (d = -0.41), although a visual
effect size for Participant 2’s decrease in negative statements inspection of the data reveals a significant overlap in the
was indicative of a medium effect (d = -0.48). Participant pre-intervention and intervention data and lack of a clear
2’s data is depicted in Fig. 3. treatment effect. Participant 3’s data is depicted in Fig. 4.

Participant 3: Decrease Negative Statements Participant 4. Increase Social Inquiries and Decrease
Verbal Contributions
Participant 3’s individual target focused on decreasing
reliance on negative statements throughout the entire Due to a participant unavailability due to a personal obli-
duration of the 20-week intervention. Prior to intervention, gation, Participant 4 was only able to complete a single
his percent of conversational intervals with negative pre-intervention session for conversation probes. At that
statements ranged from 3.0 to 15.5 % (a mean of 9.3 %) time, the percent of her social inquiries was 27 % (7.5
Fig. 4 Participant 3’s dynamic Pre-target phase Active target phase
conversation target social goal:
decrease negative statements
Percentage of intervals with

80
negative statements/topics

70
60
50
40
30
20
10
0
Pre 1 Pre 2 1 2 3 4

123
 J Autism Dev Disord

questions per conversation). During her active target skill was actively targeted (progress meetings 1–2), the percent
phase (following the first five sessions and measured at of social inquiries made increased to a mean of 31.0 % of
progress meeting 1) the percentage increased to a mean of the total questions asked (range of 27.8–34.2 %). She
40.0 % of total social inquiries made in the conversation asked a mean of 5.8 questions per conversation (range of
(16.5 questions). During the post-target phase when social 3–9). During the post-target phase (progress meetings 3–4),
inquiries were no longer the focus of self-management she made a total of 26.6 % of the total social inquiries in
(progress meetings 2–4), Participant 4’s percentage of total the conversation(range of 14.3–38.9 %), asking a mean of
social inquiries remained at a mean of 32.3 % (range of 5.5 questions (range of 4–7). The increase from pre-inter-
27–35 %) with a total mean of 10.8 questions (range of vention is indicative of a large effect size (d = 3.30).
5–15). This sustained improvement was indicative of a After targeting social inquiries, Participant 5’s target
large effect (d = 1.54). goal shifted to increasing her verbal contributions for the
Participant 4’s target objective shifted to focus on final two progress meetings. During the pre-target phase
reducing her verbal contributions for sessions 6–20 (and she spoke during a mean of 43.1 % of conversational
assessed during progress meetings 2–4). During the pre- intervals (range of 36.0–51.0 %). When verbal contribu-
target phase, her verbal contributions made up a mean of tions became the active treatment target, Participant 5’s
75 % of conversational intervals (range of 71–79 %). contributions increased to a mean of 62.5 % (range of
During the active target phase (progress meetings 2–4), she 57.5–67.5 %) for progress meetings 3 and 4. This change
reduced her verbal contributions to a mean of 61 % (range was indicative of a large effect size (d = 2.90). Participant
of 51–70 %), with a notable decreasing trend in her data. 5’s data is depicted in Fig. 6.
This change was indicative of a large effect (d = -1.74).
Participant 4’s data is depicted in Fig. 5. Participant 6: Increase Social Inquiries and Decrease
Negative Statements
Participant 5: Increase Social Inquiries and Increase
Verbal Contributions Participant 6 contributed minimally to the total social
inquiries that were made during his pre-intervention con-
Participant 5 did not make any social inquiries (0) during versation probes, making up a mean of 1 % of total ques-
her pre-intervention conversation probes. When this skill tions asked (range of 0–1.6 %). He made a total mean of

Fig. 5 Participant 4’s dynamic Pre-target phase Active target phase Post-target phase
conversation target social goal:
increase social inquiries and
Percentage of social inquiries

decrease verbal contributions 80

70
made by participant

60
50
40
30
20
10
0

80
with a verbal contribution
Percentage of intervals

70
60
50
40
30
20
10
0
Pre 1 Pre 2 1 2 3 4

123
J Autism Dev Disord

Fig. 6 Participant 5’s dynamic Pre-target phase Active target phase Post-target phase
conversation target social goals:
increase social inquiries and 80

Percentage of social inquiries

increase verbal contributions
70

made by participant
60
50
40
30
20
10
0

with a verbal contribution 80

70
Percentage of intervals

60
50
40
30
20
10
0
Pre 1 Pre 2 1 2 3 4

0.3 social inquiries (range of 0–1.0). Social inquiries were The participants also indicated that they highly enjoyed
actively targeted during progress meetings 1–3 and the their time in the groups (mean rating of 8.33 out of 10,
percent of inquiries made increased to a mean of 11.2 SD = 1.63). They also endorsed that their social skills and
(range of 3.8–21.4 %), with his total number of questions competence improved through participation in the group
increasing to a mean of 2.2 (range of 1–5 per conversation). (mean of 8.17 out of 10, SD = 1.33).
During his one post-target phase probe (progress meeting
4), he asked 4 % of the total questions asked (mean of 1
question). Overall, the increase in Participant 6’s question Discussion
asking was indicative of a large effect (d = 1.29).
During the final five weeks of the START group, Par- To ascertain meaningful information about social skill
ticipant 6’s target focus shifted to decreasing negative improvements from the survey data, the mean of the last
statements. During the pre-target phase (both pre-inter- two progress meeting data points (meetings 3 and 4) were
vention meetings and progress meetings 1–3), he made a compared to the mean of both pre-intervention data points.
negative statement an average of 27.8 % of conversational In this preliminary investigation, participants were noted to
intervals (range of 0–61.0 %). During the active target demonstrate some promising evidence of improvement
phase, Participant 6’s use of negative statements decreased across several of the utilized parent and adolescent survey
to 21.0 %, which was indicative of a large effect report measures. Among measures with corresponding
(d = -0.91). parent and self-report versions, the adolescents consistently
Participant 6’s data is depicted in Fig. 7. rated themselves as more socially competent than their
parents rated them at all time points. These findings align
Social Validity Ratings with previous studies (e.g. McMahon and Solomon 2015;
Lerner et al. 2012) and suggest that in general, adolescents
All parents endorsed ratings indicating that their adolescent with ASD may overestimate their relative level of social
highly enjoyed being a part of the social skills group (mean aptitude. Whereas these survey measures assessed impor-
rating of 8.58 out of 10, SD = 1.11). Additionally, they tant parent and adolescent perspectives on social func-
endorsed that their child’s social skills and competence tioning improvements, the use of data extracted from live
improved through participation in the group (mean of 7.83 conversation probes also contributed valuable information
out of 10, SD = 0.98). about their social competence profiles.

123
 J Autism Dev Disord

Fig. 7 Participant 6’s dynamic Pre-target phase Active target phase Post-target phase
conversation target goals:
increase social inquiries and
80

Percentage of socail inquiries

reduce negative statements
70

made by participant
60
50
40
30
20
10
0

80
Percentage of intervals with
negative statements/topics
70
60
50
40
30
20
10
0
Pre 1 Pre 2 1 2 3 4

The conversational skill data are indicative of encour- social competence (and the programs that aim to improve
aging improvements to a range of individualized social such competencies) incorporate live, dynamic perfor-
difficulties—making social inquiries, decreasing negative mance measures as evaluative strategies. To our knowl-
statements, and increasing or decreasing total verbal con- edge, this preliminary investigation is one of the first
tributions during social exchanges. These findings varied in socialization interventions to incorporate serial conversa-
terms of the magnitude of the treatment effect, but seem to tion probes to augment the use of more traditional social
provide preliminary evidence of the value of incorporating survey measures when assessing improvements to inter-
individualized target skills, experiential opportunities, and personal functioning.
self-management into a socialization program. The con- Autism spectrum disorder can be conceptualized as a
versational target skill data provides promising evidence of disorder of socialization, with its very diagnostic criteria
both (a) the potential benefits of individualizing some presupposing impairments in social motivation and func-
aspects of a group socialization curriculum and (b) the tioning (American Psychiatric Association 2013). Because
utility of using serial conversation probes as a novel means of the pervasive and complex nature of interpersonal vul-
to assess social competence improvements. nerabilities associated with ASD (Bauminger 2002; Bau-
Dynamic (live) social performance measures are only minger and Kasari 2000), which can often be intertwined
rarely used in the social skills intervention research liter- with a long history of social rejection or isolation (e.g.
ature, with most programs relying on survey measures as Chamberlain et al. 2007), adolescents with ASD are likely
their primary outcome measures (Bolte and Diehl 2013; to require a unique training program that simultaneously
McMahon et al. 2013a). Live measures may yield more attends to multiple treatment considerations. The diverse
ecologically valid data (Elliott and Gresham 1987; profiles of adolescents with ASD create a unique challenge
McMahon et al. 2013b; Merrell 2001) and are immune or for the development of an effective social skills interven-
less susceptible to social desirability biases, random tion program (Weiss and Harris 2001; Matson et al. 2007).
responding, and demand characteristic issues that are rec- A program that exclusively focuses on didactic social skills
ognized as limitations to survey measures (Furr and Funder acquisition is unlikely to benefit those with low motivation
2007; Moskowitz 1986). to engage with others and/or those who only have access to
Social competence is perhaps best encapsulated as nonreciprocal social partners. On the other hand, an inter-
one’s level of success within live, dynamic exchanges vention that solely provides a highly motivating experien-
with peers. Therefore, it is logical that evaluations of tial social context may be suitable for relatively

123
J Autism Dev Disord

sophisticated individuals, but is unlikely to benefit to those sample, a larger pool of participants is required to identify
who lack knowledge of basic socialization strategies. how universally applicable this intervention package would
The hybrid experiential and didactic components of the be to the broader population of adolescents with ASD.
START program were designed to directly address the Additionally, no data was collected on the quality of the
diverse needs of the population within a low-demand, high- participants’ interactions in real world social settings dur-
motivation context. The resulting social validity ratings from ing or after the conclusion of the program. Such informa-
the parent and participant adolescents of this intervention tion is necessary for a comprehensive understanding of the
were quite favorable—possibly because this preliminary generalized effects of program participation. To address
evaluation incorporated engaging components similar to these concerns, a randomized controlled trial is currently
existing afterschool clubs and utilized a cohort of peers and underway to further investigate the merits of the START
social facilitators. Both participants and their parents program with increased methodological rigor, a signifi-
endorsed a high level of satisfaction with the social group cantly increased sample size, and long-term follow-up data.
curriculum and the adolescents appeared to build strong Our initial findings suggest that the START social skills
group relationships. These observations and feedback are not program may serve as a promising intervention for
inconsequential, as an effective social skills curriculum improving the social competence of adolescents with ASD.
requires strong buy-in from both adolescent and parent Based on our preliminary data, adolescents who partici-
stakeholders. Motivation for continued participation is pated in our multi-component social club appear to expe-
essential, and the described program intentionally embedded rience indications of improvement to social functioning, as
several components for boosting and maintaining social measured by survey and live conversational measures.
interest and camaraderie (i.e., use of college-aged social Following participation, these individuals seem to be
facilitators, a variety of group activities and games, the equipped with an improved social skill set and associated
provision of snacks, and a hands-on experiential emphasis). level of social motivation, informed by an increased
The opportunity to interact with peers and jointly work on understanding of social rules and expectations. Social
social skills in an experiential setting may be preferable to competence is essential at all ages, but perhaps more so
attending a social skills group facilitated solely by adult pro- during the critical transition to young adulthood (Hendricks
fessionals. Because everyone in the program (including the and Wehman 2009). This time period is often accompanied
college-aged social facilitators) jointly completed the self- by a significant decrease in external support as adolescents
management tasks and social lessons, there may have also been graduate from high school and age out of federally-man-
a minimization of the inherent stigma traditionally associated dated special education services (Camarena and Sarigiani
with having to acknowledge one’s social deficits. The inter- 2009). Therefore, prior to this transitional time point, it
vention setting was designed to closely mirror real-world may be crucial to equip adolescents with ASD with a core
socialization settings to help build confidence and competence set of social competencies to maximize their long-term
across structured and unstructured social situations. quality of life and create strong opportunities for increased
It should be noted that there were several limitations to independent living. Ultimately, interpersonal aptitude lar-
the current preliminary investigation. The repeated mea- gely governs their ultimate level of success in many of
sure clinical case series component of this investigation life’s pursuits, including the obtainment of desirable
was non-experimental in nature, with no contrast group that employment, development of meaningful friendships, and
would allow for controlled comparisons. This limitation is establishment of romantic relationships (Burt et al. 1991;
somewhat tempered by the addition of the multiple base- Eaves and Ho 2008; Hurlbutt and Chalmers 2002).
line design for the dynamic conversation probe data, which
allowed the participants to serve as their own controls for Acknowledgments The authors would like to acknowledge all of
the adolescents and parents that participated in this research, along
their individualized target behaviors. Baseline phases were with all of the undergraduate research assistants that made this project
staggered across participants and behaviors, ranging from possible.
one to five time-points. Multiple baseline designs are
considered a rigorous single-case experimental design Author Contributions TV and AM jointly developed the START
intervention model and conducted the preliminary evaluation trial. JK
(Barlow et al. 2009). assisted with intervention model development. VW served as the lead
In the current study, the small sample size also limits the research assistant on this project and assisted with data input, orga-
generalization of the preliminary findings. While some nization, and analysis. All authors contributed to significant portions
promising initial outcomes were observed among our small of the manuscript drafting and revision process.

123
 J Autism Dev Disord

Appendix 1: Social Competency and Motivation Scale

Instructions: For each question, please indicate how much you agree with each of the following statements by selecting
ONE of the following choices. If you are unsure, please make your best guess.
Not at all Rarely Sometimes Usually Almost Always

My child is comfortable in social situations 1 2 3 4 5

My child knows how to interact with an unfamiliar person 1 2 3 4 5
My child is willing to go out in public by him/herself 1 2 3 4 5
My child knows how to make friends with other individuals his/her age 1 2 3 4 5
My child makes eye contact with people when talking to them 1 2 3 4 5
My child maintains conversations by asking questions 1 2 3 4 5
My child maintains conversations by making comments 1 2 3 4 5
My child stays on topic during conversations 1 2 3 4 5
My child brings up appropriate conversation topics 1 2 3 4 5
My child greets others appropriately 1 2 3 4 5
My child introduces him/herself appropriately 1 2 3 4 5
My child ends conversations appropriately 1 2 3 4 5
My child uses appropriate voice volume/rate intonation when speaking 1 2 3 4 5
My child exhibits appropriate facial expressions when talking with others 1 2 3 4 5
My child verbally expresses empathy for others 1 2 3 4 5
My child compliments others 1 2 3 4 5
My child uses age-appropriate humor 1 2 3 4 5
My child has an interest in other people 1 2 3 4 5
My child wants to have friends 1 2 3 4 5
My child knows how to behave when he/she is at a party 1 2 3 4 5
My child prefers to be with other people during his/her free time 1 2 3 4 5
My child loves to talk to other people 1 2 3 4 5
My child likes to learn about the experiences of other people 1 2 3 4 5
My child feels comfortable going out with peers in group outings 1 2 3 4 5
My child goes over to friends’ houses or has them come to our house 1 2 3 4 5
My child is skilled in maintaining relationships over time 1 2 3 4 5

References Blair, R. J. R., Frith, U., Smith, N., Abell, F., & Cipolotti, L. (2002).
Fractionation of visual memory: Agency detection and its
American Psychiatric Association. (2013). Diagnostic and statistical impairment in autism. Neuropsychologia, 40(1), 108–118.
manual of mental disorders: DSM-5. Washington, DC: American Bolte, E. E., & Diehl, J. J. (2013). Measurement tools and target
symptoms/skills used to assess treatment response for individ-
Psychiatric Association.
Baker, A. C., Jensen, P. J., & Kolb, D. A. (2002). Conversational uals with autism spectrum disorder. Journal of Autism and
learning: An experiential approach to knowledge creation. Developmental Disorders, 43(11), 2491–2501.
Portsmouth, NH: Greenwood Publishing Group. Burleson, B. R., Applegate, J. L., Burke, J. A., Clark, R. A., Delia, J. G.,
& Kline, S. L. (1986). Communicative correlates of peer accep-
Barlow, M. D., Nock, M., & Hersen, M. (2009). Single case
experimental designs: Strategies for studying behavior for tance in childhood. Communication Education, 35(4), 349–361.
change (No. Sirsi) i9780205474554). Burt, D. B., Fuller, S. P., & Lewis, K. R. (1991). Brief report:
Baron-Cohen, S. (1995). Theory of mind and face-processing: How Competitive employment of adults with autism. Journal of
do they interact in development and psychopathology? In Autism and Developmental Disorders, 21(2), 237–242. doi:10.
Developmental psychopathology, Vol. 1: Theory and methods 1007/BF02284763.
(pp. 343–356). Oxford: John Wiley & Sons. Camarena, P. M., & Sarigiani, P. A. (2009). Postsecondary educa-
Bauminger, N. (2002). The facilitation of social-emotional understand- tional aspirations of high-functioning adolescents with autism
ing and social interaction in high-functioning children with autism: spectrum disorders and their parents. Focus of Autism and Other
Intervention outcomes. Journal of Autism and Developmental Developmental Disabilities, 24(2), 115–128. doi:10.1177/
Disorders, 32(4), 283–298. doi:10.1023/A:1016378718278. 1088357609332675.
Carr, M. E., Moore, D. W., & Anderson, A. (2014). Self-management
Bauminger, N., & Kasari, C. (2000). Loneliness and friendship in
high-functioning children with autism. Child Development, interventions on students with autism: A meta-analysis of single-
71(2), 447–456. doi:10.1111/1467-8624.00156. subject research. Exceptional Children, 81(1), 28–44.

123
J Autism Dev Disord

Chamberlain, B., Karasi, C., & Rotheram-Fuller, E. (2007). Involve- Kim, J. A., Szatmari, P., Bryson, S. E., Streiner, D. L., & Wilson, F. J.
ment or isolation? The social networks of children with autism in (2000). The prevalence of anxiety and mood problems among
regular classrooms. Journal of Autism and Developmental children with autism and Asperger syndrome. Autism, 4(2),
Disorders, 37(2), 230–242. doi:10.1007/s10803-006-0164-4. 117–132. doi:10.1177/1362361300004002002.
Church, C., Alisanski, S., & Amanullah, S. (2000). The social, Koegel, L. K., Koegel, R. L., Hurley, C., & Frea, W. D. (1992).
behavioral, and academic experiences of children with asperger Improving social skills and disruptive behavior in children with
syndrome. Focus on Autism and Other Developmental Disabil- autism through self-management. Journal of Applied Behavior
ities, 15(1), 12–20. doi:10.1177/108835760001500102. Analysis, 25(2), 341–353.
Constantino, J., & Gruber, C. (2005). Social responsiveness scale. Los Kolb, D. A. (2014). Experiential learning: Experience as the source
Angeles: Western Psychological Services. of learning and development (2nd ed.). Upper Saddle River, NJ:
DiCicco-Bloom, E., Lord, C., Zwaigenbaum, L., Courchesne, E., Pearson Education.
Dager, S. R., Schmitz, C., & Young, L. J. (2006). The Kolb, D. A., Boyatzis, R. E., & Mainemelis, C. (2001). Experiential
developmental neurobiology of autism spectrum disorder. The learning theory: Previous research and new directions. In R.
Journal of Neuroscience, 26(26), 6897–6906. J. Sternberg & L. F. Zhang (Eds.), Perspectives on thinking,
Doggett, R. A., Krasno, A. M., Koegel, L. K., & Koegel, R. L. (2013). learning, and cognitive styles (pp. 227–247). Mahwah, NJ:
Acquisition of multiple questions in the context of social Lawrence Erlbaum.
conversation in children with autism. Journal of Autism and Landa, R. (2000). Social language use in Asperger syndrome and
Developmental Disorders,. doi:10.1007/s10803-012-1749-8. high-functioning autism. In A. Klin, F. R. Volkmar, & S.
Eaves, L. C., & Ho, H. H. (2008). Young adult outcome of autism S. Sparrow (Eds.), Asperger syndrome. New York: Guilford
spectrum disorders. Journal of Autism and Developmental Press.
Disorders, 38(4), 739–747. doi:10.1007/s10803-007-0441-x. Laugeson, E. A., Frankel, F., Gantman, A., Dillon, A. R., & Mogil, C.
Elder, L. M., Caterino, L. C., Chao, J., Shacknai, D., & DeSimone, G. (2012). Evidence-based social skills training for adolescents with
(2006). The efficacy of social skills treatment for children with autism spectrum disorders: The UCLA PEERS program. Journal
Asperger syndrome. Education & Treatment of Children, 29(4), of Autism and Developmental Disorders, 42(6), 1025–1036.
635–663. doi:10.1007/s10803-011-1339-1.
Elliott, S. N., & Gresham, F. M. (1987). Children’s social skills: Lee, S., Simpson, R. L., & Shogren, K. A. (2007). Effects and
Assessment and classification practices. Journal of Counseling & implications of self-management for students with autism: A
Development, 66(2), 96–99. meta-analysis. Focus on Autism and Other Developmental
Engström, I., Ekström, L., & Emilsson, B. (2003). Psychosocial Disabilities, 22(1), 2–13. doi:10.1177/10883576070220010101.
functioning in a group of swedish adults with asperger syndrome Lerner, M. D., Calhoun, C. D., Mikami, A. Y., & De, L. R. (2012).
or high-functioning autism. Autism, 7(1), 99–110. Understanding parent–child social informant discrepancy in
Furr, R., & Funder, D. (2007). Behavioral observation. In Handbook youth with high functioning autism spectrum disorders. Journal
of research methods in personality psychology (pp. 273–291). of Autism and Developmental Disorders, 42(12), 2680–2692.
New York: The Guilford Press. doi:10.1007/s10803-012-1525-9.
Ghaziuddin, M., & Gerstein, L. (1996). Pedantic speaking style Lord, C., Rutter, M., DiLavore, P. C., Risi, S., Gotham, K., & Bishop,
differentiates Asperger syndrome from high-functioning autism. S. (2012). Autism diagnostic observation schedule (2nd ed.).
Journal of Autism and Developmental Disorders, 26(6), Torrance, CA: Western Psychological Services.
585–595. Matson, J. L., Matson, M. L., & Rivet, T. T. (2007). Social-skills
Ghaziuddin, M., Ghaziuddin, N., & Greden, J. (2002). Depression in treatments for children with autism spectrum disorders. Behavior
persons with autism: Implications for research and clinical care. Modification, 31(5), 682–707. doi:10.1177/0145445507301650.
Journal of Autism and Developmental Disorders, 32(4), McMahon, C. M., Lerner, M. D., & Britton, N. (2013a). Group-based
299–306. doi:10.1023/A:1016330802348. social skills interventions for adolescents with higher-function-
Grainger, C., Williams, D. M., & Lind, S. E. (2014). Metacognition, ing autism spectrum disorder: A review and looking to the
metamemory, and mindreading in high-functioning adults with future. Adolescent Health, Medicine and Therapeutics, 4, 23.
autism spectrum disorder. Journal of Abnormal Psychology, McMahon, C. M., & Solomon, M. (2015). Brief Report: Parent–
123(3), 650. adolescent informant discrepancies of social skill importance
Gresham, F. M., & Elliott, S. N. (2008). Social skills improvement and social skill engagement for higher-functioning adolescents
system (SSIS) Rating Scales. Minneapolis, MN: Pearson. with autism spectrum disorder. Journal of Autism and Develop-
Gutstein, S. E., & Whitney, T. (2002). Asperger syndrome and the mental Disorders, 45(10), 3396–3403.
development of social competence. Focus on Autism and Other McMahon, C. M., Vismara, L. A., & Solomon, M. (2013b).
Developmental Disabilities, 17(3), 161–171. doi:10.1177/ Measuring changes in social behavior during a social skills
10883576020170030601. intervention for higher-functioning children and adolescents with
Hendricks, D. R., & Wehman, P. (2009). Transition from school to autism spectrum disorder. Journal of Autism and Developmental
adulthood for youth with autism spectrum disorders. Focus on Disorders, 43(8), 1843–1856.
Autism and Other Developmental Disabilities, 24(2), 77–88. Merrell, K. W. (2001). Assessment of children’s social skills: Recent
doi:10.1177/1088357608329827. developments, best practices, and new directions. Exceptionality,
Howlin, P. (2000). Outcome in adult life for more able individuals 9(1–2), 3–18.
with autism or asperger syndrome. Autism, 4(1), 63–83. Miller, A.R., Vernon, T.W., Wu, V., Russo, K. (2014) Social skill
Hurlbutt, K., & Chalmers, L. (2002). Adults with autism speak out: group interventions for adolescents with autism spectrum
Perceptions of their life experiences. Focus on Autism and Other disorders: A systematic review. Review Journal of Autism and
Developmental Disabilities, 17(2), 103–111. doi:10.1177/ Developmental Disorders.
10883576020170020501. Moskowitz, D. S. (1986). Comparison of self-reports, reports by
Jones, W., & Klin, A. (2009). Heterogeneity and homogeneity across knowledgeable informants, and behavioral observation data.
the autism spectrum: The role of development. Journal of the Journal of Personality, 54(1), 294–317.
American Academy of Child & Adolescent Psychiatry, 48(5), Nadig, A., Lee, I., Singh, L., Bosshart, K., & Ozonoff, S. (2010). How
471–473. does the topic of conversation affect verbal exchange and eye

123
 J Autism Dev Disord

gaze? A comparison between typical development and high- Reichow, B., Steiner, A. M., & Volkmar, F. (2012). Social skills
functioning autism. Neuropsychologia, 48(9), 2730–2739. groups for people aged 6 to 21 with autism spectrum disorders
National Autism Center. (2009). National Standards Report. Ran- (ASD). The Cochrane Database of Systematic Reviews, 7, 1.
dolph, MA: Self-published. doi:10.1002/14651858.CD008511.pub2.
Newcomb, A. F., Bukowski, W. M., & Pattee, L. (1993). Children’s Rutter, M., Le Couteur, A., & Lord, C. (2003). Autism diagnostic
peer relations: A meta-analytic review of popular, rejected, interview, revised. Torrance, CA: Western Psychological
neglected, controversial, and average sociometric status. Psy- Services.
chological Bulletin, 113(1), 99. Starr, E., Szatmari, P., Bryson, S., & Zwaigenbaum, L. (2003).
Newman, B., & Ten Eyck, P. (2005). Self-management of initiations Stability and change among high-functioning children with
by students diagnosed with autism. Analysis of Verbal Behavior, pervasive developmental disorders: A 2-year outcome study.
21, 117–122. Journal of Autism and Developmental Disorders, 33(1), 15–22.
Ojanen, T., Sijtsema, J. J., Hawley, P. H., & Little, T. D. (2010). Webb, B. J., Miller, S. P., Pierce, T. B., Strawser, S., & Jones, W. P.
Intrinsic and extrinsic motivation in early adolescents’ friendship (2004). Effects of social skill instruction for high-functioning
development: Friendship selection, influence, and prospective adolescents with autism spectrum disorders. Focus on Autism
friendship quality. Journal of Adolescence, 33(6), 837–851. and Other Developmental Disabilities, 19(1), 53–62. doi:10.
Ollendick, T. H., Weist, M. D., Borden, M. C., & Greene, R. W. 1177/10883576040190010701.
(1992). Sociometric status and academic, behavioral, and Weiss, M. J., & Harris, S. L. (2001). Teaching social skills to people
psychological adjustment: A five-year longitudinal study. Jour- with autism. Behavior Modification, 25(5), 785–802. doi:10.
nal of Consulting and Clinical Psychology, 60, 80–87. 1177/0145445501255007.
Ozonoff, S., & Miller, J. N. (1995). Teaching theory of mind: A new White, S. W., Koenig, K., & Scahill, L. (2010). Group social skills
approach to social skills training for individuals with autism. instruction for adolescents with high-functioning autism spec-
Journal of Autism and Developmental Disorders, 25(4), trum disorders. Focus on Autism and Other Developmental
415–433. Disabilities, 25(4), 209–219. doi:10.1177/1088357610380595.
Palmen, A., Didden, R., & Arts, M. (2008). Improving question Williams, D. M., Lind, S. E., & Happé, F. (2009). Metacognition may
asking in high-functioning adolescents with autism spectrum be more impaired than mindreading in autism. Behavioral and
disorders: Effectiveness of small-group training. Autism, 12(1), Brain Sciences, 32(02), 162–163.
83–98. doi:10.1177/1362361307085265.
Parker, J., & Asher, S. (1987). Peer relations and later personal
adjustment: Are low accepted children ‘‘at risk’’? Psychological
Bulletin, 102, 357–389.

123