Can Robots Serve as an Effective Instructional tool for decreasing Off-Task Behaviors in Young Children with Autism Spectrum Disorders?

By

Amelia K. Moody

Sharon Richter

 

University of North Carolina Wilmington

This issue of NASET’s Autism Spectrum Disorder was written by Amelia K. Moody and Sharon Richter from the University of North Carolina Wilmington. The researchers conducted a single subject reversal design to compare the Superflex® curriculum and a researcher- developed Robotics-Enhanced Superflex intervention on off-task behavior among three elementary students with autism. Researchers implemented two phases of the study, including the Robotics- Enhanced Social Skills Instruction and traditional social skills instruction across a 10-week period. The results indicate that off task behaviors were decreased using Robotics-Enhanced Social Skills instruction over a traditional social skills lesson. Results were educationally significant for decreasing off-task behaviors while increasing cognitive demands.

---

Can Robots Serve as an Effective Instructional tool for decreasing Off-Task Behaviors in Young Children with Autism Spectrum Disorders?

---

 

According to the Center for Disease Control 1 out of 59 children is diagnosed with an Autism Spectrum Disorder (ASD; CDC, 2018), with almost half having average to above average IQs (Christensen et al., 2016). Despite their intelligence level, many students with ASD have poor outcomes later in life (Dijkhuis, Ziermans, Van Rijn, Staal & Swaab, 2016). These findings may relate to one’s ability to focus and attend to instructional lessons. Students with ASD are often less engaged in instructional activities than their typically developing peers due to deficits in social, communication, sensory, and behavioral skills. It is critical that children with ASD work on their attention skills so they can focus on learning.

Attention is one’s ability to maintain focus on tasks, objects, people, or events (Patten & Watson, 2011). Maintaining attention can be challenging for children with ASD and research indicates that many are less engaged than their typically developing peers during instruction (Carnahan, Musti-Rao, & Bailey, 2009). Some reasons for this include having fixated interests on chosen objects (Liss, Sauliner, Fein, & Kinsbourne, 2006; Patten & Watson, 2011). This desire for activities that are of interest can often cause off-task behaviors like self-stimulatory behaviors (Abel, Gadomski, and Brodhead, 2016) or scripted discussions (e.g., echolalia; Bertoglio & Hendren, 2009). Another reason may be due to a child’s desire to participate in disruptive, avoidant, and escape behaviors when presented with a challenging academic activity (Geiger, Carr, & LeBlanc, 2010). Off-task behaviors can also be due to a lack of executive functioning that impacts understanding of other people’s thoughts, knowledge and beliefs (Baron-Cohen, 2001).  A lack of executive functioning can relate to impaired social learning and imitation (Rogers & Pennington, 1991; Vanvuchelen, Roeyers, & deWeerdt, 2011). Individuals with ASD can misinterpret social information and this can cause breakdowns in interactions. Decreased social communication skills can be associated with poor academic outcomes due to anxiety and social discomfort (Welsch, Parke, Widaman, & O’Neil, 2001). Effective social skills instructional groups can assist children with ASD in learning how to effectively communicate and empathize with peers.

Social Thinking

One intervention aimed at improving social communication and self-regulation skills is called social thinking. Social thinking is a methodology used to assist children in handling social communication challenges and offering them solutions (Crooke & Winner, 2016). According to researchers, combining two or more social communication interventions is often more effective than using programs in isolation (Attwood, 2000; Bauminger, 2002; Crooke et al., 2008). Many social skills programs utilize visual supports to assist in engaging and supporting positive behaviors (see Wong et al., 2005). These approaches can include scripting (Goldstein & Cisar, 1992; Smith Myles, Trautman, & Schelvan, 2004; Gresham, 2002; Rao et al., 2008; Stevenson, Krantz, & McClannahan, 2000; Sze & Wood, 2007), role plays (Anderson & Morris, 2006; Bauminger, 2002; Doctoroff, 1997), social stories (Gray, 1991; Gray, 2000; Gray & Garland, 1993), and comic strip conversations (Pierson & Glaser, 2007).

For example, Nowell and colleagues (2013) examined the use of a social thinking program to increase social communication and self-regulation in 17 first- and second-grade students using a randomized delayed treatment control group design. Differences between pre- and post-assessments indicated that children with ASD and their parents showed increased social communication and self-regulation knowledge when compared to a treatment control group and were maintained up to six months after the post assessment.

For the purposes of this study, the teacher was using the Superflex…a Superhero Social Thinking Curriculum Package® (Superflex®; Madrigal & Winner, 2008), which is used in schools across the United States. The program has gained attention (see Carey, 2011), yet few studies have yet proven its efficacy (Bolton, 2010). This program is designed to teach flexible thinking and children practice skills like determining if a problem is big or little. The curriculum includes two cartoon characters like “Superflex®” who demonstrates flexible thinking and “Rock Brain” who has difficulties using flexible thinking (Madrigal & Winner, 2008). The Superflex® curriculum is designed to improve a child’s social expectations, awareness of their own behaviors, learning of strategies to modify their own behaviors, and use of flexibility when adapting to their social environments (Bolton, 2010). Proponents of the curriculum suggest that using images can allow children to externalize descriptions of negative behaviors which can in turn reduce embarrassment while improving flexible thinking skills (Madrigal, Winner, & Knopp, 2009).

To examine the impact of Superflex®, Bolton (2010) completed a dissertation research study using a multiple case study design to determine the impact of Superflex® on social skill behaviors among 6 middle school boys with autism and social skill deficits. The dependent variables were measured using direct behavioral observations (DBO; Patterson, Jolivette, & Crosy, 2006) and a standardized measure, The Behavior Assessment System for Children – Second Edition (BASC-2; Reynolds & Kamphaus, 2004). The DBO indicated statistically significant improvements related to social and self-regulation skills. Participants’ results on the BASC-2 were not statistically significant.

Yadlosky (2012) investigated the impact of Superflex® on social communication and self-regulation among three second grade students with ADHD, one with a dual diagnosis of ASD to fulfill a master’s degree thesis. Classroom observations were collected in baseline and at bi-weekly intervals for a total of 5, 20-minute observations, across 12 sessions using a frequency count for appropriate and inappropriate behaviors (Adapted form Dowd & Tierney, 1992). Teachers also completed a pre/post questionnaire describing participants’ talking behaviors. Results support the use of the curriculum based on behavioral observations (e.g., improved understanding of social environment, hidden cues, and ability to listen). Teacher reports indicated gains across all measure including participants’ ability to adapt to social environments and understand hidden cues.

Some positive effects were noted in the previously discussed research including increases in students’ social skills and flexible thinking. However, additional research needs to be done to ascertain how to better support children’s off-task behaviors, so they can stay on task to receive the curriculum in the most effective ways.

The Use of Robotics to Increase Engagement

Research indicates that the use of robots can inspire motivation, improve learning, and increase academic outcomes (Barker & Ansorge, 2007; Chen et al., 2010; Chung, et al., 2010; Fasola & Mataric, 2010). Children with autism are often highly interested in using robotics and they can often increase motivation and interactive participation in activities (Alcorn et al., 2019). Robotics can be exciting for children and have functions that are easily to understand. Research indicates that robotics offers predictable and reliable interactions that pair well with the executive functioning of children with ASD (see Dautenhahn, 1999; Dautenhahn & Werry, 2004; Straten et al., 2018). For example, most robots have predictable patterns of behaviors and functions. They also use visual coding systems that meet the needs of students with ASD who process visually. Research also suggests that robotics can increase engagement (Robins & Dautenhahn, 2006), joint attention, and social behaviors (see Anzalone et al., 2015).

In the current study, researchers decided to pair the Superflex ® curriculum with robotics to see if there were important differences in off-task behaviors among participants. Researchers conducted a single subject reversal design to compare the Superflex® curriculum and a researcher- developed Robotics-Enhanced Superflex intervention on off-task behavior among children with autism. The research questions were as follows:

(1) Does Superflex ® or the Robotics-Enhanced Superflex intervention have a greater impact on off-task behavior among children with autism?

(2) Are the goals, procedures, and outcomes associated with this study rated as reasonable by the classroom teacher?

Method

Participants and Setting

Three participants were included in the study. Participants were (a) children with autism spectrum disorder who were enrolled in a special education classroom at a public elementary school and attended a social skills groups two times per week for 30-minutes each, (b) identified by the special education teacher as children who demonstrated off-task behavior, and (c) were in attendance for all 10 sessions of the investigation. 

According to participants’ Individualized Education Plans, one participant was identified with a Developmental Delay and two were identified with autism spectrum disorder. Participants included one female and two males. Participants’ ages ranged from 5 years, two months to 7 years, two months old (M = 6.07 years, SD= 1.03; See Table 1).

 

Table 1

Demographic data

Participant	Age/ Grade	Disability	Gender	Race/ Ethnicity	Native Language
D01 Don	5y/8mo	DD	Male	Eastern European	English
G01 Georgina	5y/2mo	ASD	Male	Caucasian	English
N01 Nico	7 y/2 mo	ASD	Male	African American	English

Don. Don was a five years and eight months old Caucasian male who is an English speaker. Don has a very happy and positive disposition and enjoys interacting with his teachers. While he will smile at his peers, he often plays alone and avoids other children without intense intervention. When presented with new and situations or routines Don often comments., “so sad” or “oh no” and requires visuals and modeling. He inconsistently participates in social skills groups and can easily become distracted (e.g., staring off, repeating TV show dialogue).

Georgina. Georgina was five years and two months old at the time of the intervention. Her preferred language is English. She is highly caring and enjoys getting her peers as they come to school. She will answer questions when she is directly asked by her peers but will not initiate or carry on conversations and enjoys playing by herself. She also requires prompts to use her kind words and is often “Bossy”. She can greet peers but needs promoting to complete a 1-2 communication exchange. One of her key objectives is to use flexible thinking when preferred choices are not available.

Nico. Nico was seven years and two months old at the time of the intervention. He is an African American boy and a native English speaker who enjoys socializing with his peers. He is social and enjoys interacting with his peers and teachers. According to his teacher, Nico’s social communication was an area identified for improvement due to poor self-awareness, social language, and impulsivity control. Nico’s flexible thinking is limited, and he can have difficulties getting back on topic. For example, he can be insistent on requests despite their relevance or appropriateness.

Dependent Variable

The dependent variable was frequency of off-task behavior, including (a) off-task motor behavior, (b) off-task verbal behavior, and (c) off-task passive behaviors among participants. In each 30-minute lesson, the researchers collected data on the number of intervals in which students demonstrated off-task behavior using momentary time sampling at 30-second intervals.

Off-task motor behavior was defined as body movements not related to the lesson. Examples of off-task motor behavior including being out of one’s seat, engaging in self-stimulatory behaviors, playing with items unrelated to the activity, making inappropriate gestures, hitting, biting, and throwing items. Off-task verbal behaviors were defined as verbalizations not related to the lesson. Examples of off-task verbal behaviors included echolalic speech, calling out, and making noises with the mouth. Off-task passive were defined as turning one’s eye gaze away from the teacher and lesson materials. Off-task passive behaviors included staring in a direction away from the teacher and lesson materials and looking away from the activity.

Experimental Design

The researcher used a reversal design with repeated reversals to investigate the impact of Robotics-Enhanced Superflex intervention on off-task behaviors including (i.e., motor, verbal, passive) among early elementary children with developmental disabilities. This design was selected because it facilitated comparison of traditional Superflex (i.e., condition A) to Robotics-Enhanced Superflex intervention (i.e., condition B). Additionally, reversal designs offer methodological rigor for single case within-subject investigations.   

Traditional Instruction. Traditional instruction comprised of the teacher leading a social skills lesson shaped around the Superflex® Curriculum which was developed to improve a child’s social expectations, behavioral understanding and modification, and adaptability to one’s environment (Botlton, 2010). The procedure included reviewing whole body listening. Children would demonstrate each behavior after it was mentioned and those who did not received a behavioral reminder. Once every child in the group was sitting quietly, looking forward, and had their hands in their lap the lesson would begin. The teacher would then introduce the daily lesson and describe an introductory story or video aligned with Superflex ® lessons on determining big and little problems, understanding how to respond to the environment, and how to control one’s behaviors. For example, the teacher would read a section of superflex ® and ask the students to offer a synopsis of what the problem was and how it was solved. Next, the teacher would offer directions of how to complete the activity with modeling and visuals. For example, the teacher would state, “Please listen quietly to the book and then we can take turns raising our hands to talk about the book”. During the activity the teacher was responsible for providing corrective feedback on incorrect responses. For example, the teacher would state, “Raise your hands before speaking”, and either point to the picture of a child raining a hand or model the practice. Finally, the children would complete a related activity to practice the skill they learned about that day. For example, the students would read scenarios and discuss what response would be appropriate. These activities always included visuals supports. Each child participated during the lesson which was ensured by having the group respond in a clockwise fashion. 

Robotics-Enhanced Superflex Instruction. The Robotics-Enhanced Superflex intervention instruction was delivered in a parallel format with the teacher: (a) reviewing whole body listening using a visual support (e.g., hand in lap, quiet mouth, etc.); (b) teacher introduced the daily activity, however the game would include the use of robotics; (c) offer directions on how to complete the activity with modeling, as appropriate; (d) provide corrective feedback on incorrect responses; using visuals supports to reinforce content (e.g., dry erase board, picture cues); (e) ensure that all students receive turns for each activity. For example. Instead of using scenarios alone with children would be responsible for programming bee bots on a grid with picture cues. The picture cue would state a problem and the children would have to determine what to do. The robot would serve as a tool to deliver the instruction. The benefit being that they learned to code while delivering the instruction.

Bee Bots® and Cozmo® were the two forms of robotics used in the study. Bee Bots® are yellow and back and look like bees with directional buttons on the top. They are designed for young children and are programmed using a simple, arrows to move forward, backward, or to turn. They are designed to teach sequencing and simple coding. Cozmo® is a social robot with human like emotions. It looks like a small front loader with a screen as a face and comes with blocks. Cozmo can play simple games (e.g., Tap it) and be programmed using drag and drop icons (e.g., move forward and then act like a dog). He can also be programmed to speak.

Procedures

Researchers implemented two phases of the study, including the Robotics- Enhanced Social Skills Instruction and traditional social skills instruction across a 10-week period. Lessons were 30 minutes each. During each session the researcher would take observational notes on the content of the traditional instruction and track off-task behaviors using a 30 second interval recording. Each child would be observed in a small group. In the next lesson, the researcher would deliver a robotics-enhanced lesson using the same content but adding robotics as a tool for engagement. For example, in one lesson children played a game where they picked a card and read about a big/little problem (e.g., hurt arm) and had to determine if it was a big or little problem. During the robotics enhanced condition, the students programmed a bee-bot to go to a chosen spot on a large grid. When the Bee Bot® arrived on the spot the child would read the picture card and identify the behavior (e.g., a student hits another student) and then determined if it was a big or little problem. Topics are shown in Table 2.

Table 2

Weekly Lesson Topics

Week	Lesson Topic
Week 1-2	Whole body listening, turn taking, and listening to others
Week 3-4	Flexible thinking and rock brain
Week 5-6	Social awareness and self-regulation
Week 7-8	Glass man and understanding emotional reactions to problems
Week 9-10	Learning about others and responding

When using Cozmo® the children were learning about taking turns and listening to one another. They played a game to practice their skills. In the Robotics-Enhanced Social Skills Instruction, children had to take turns creating stories, using Cozmo’s® drag and drop programming center. They practiced listening to one another’s ideas and taking turns. Meanwhile, there was an increase in their cognitive demand because they were required to code and use their social skills. The robot served as an engagement tool. When teaching the children to code the Bee Bots the teacher used modeling and a dry erase board and wrote out the codes as the children announced them. Then students would refer to the visual as they programmed the bots (e.g., –>, <–). This was not required for the Cozmo® because it offers visual feedback and visual coding features already. Weekly lessons were decided on by the classroom teacher and delivered by the classroom teacher. The topics included whole body listening, flexible thinking, turn taking, listening, emotional reactions, and responding. All lessons in both conditions used visual supports like Superflex© books and images, dry erase boards for coding, and visual directions for whole body listening.                        

Interrater Reliability. To assess the extent to which dependent variable data were collected with reliability, interrater reliability was assessed by a second observer for 25% of all sessions. To collect interrater reliability data, the second observer will use assess student performance using the same materials and procedures as the researcher. Interrater reliability was calculated using an item-by-item method to garner a percentage of agreements and disagreements.

Procedural Reliability. To assess the extent to which the teacher implemented the intervention with integrity, procedural reliability was assessed by a second observer in 25% of all sessions. To collect procedural reliability data, the second observer evaluated whether steps of the lesson were conducted using a researcher-made checklist, which included steps such as reviewing expectations for “whole body listening” using a visual support and providing opportunities for all students to participate in each activity. Procedural reliability agreement was determined by dividing the number of observed instructor behaviors by the total number of instructor behaviors and multiplying by 100 (Billingsley, White, & Munson, 1980). The total correct responses for procedural reliability was 98%.

Assessment of Social Validity. As recommended by Wolf (1978), the researchers assessed social validity of the goals, procedures, and outcomes associated with this investigation. To do so, the researcher asked the classroom special education teacher to provide open-ended responses to a 5-item questionnaire. The questions were as follows: (1) Is decreasing off-task behaviors important among young learners with ASD? (2) Do you feel that the intervention was reasonable in terms of time on the part of researchers and teachers? (3) Do you think that the intervention was reasonable in terms of cost of resources and supplies? (4) Do you believe that the materials were easy to use by teachers and students? (5) Do you think that the outcomes of the study were important and meaningful in terms of student behavior? The teacher’s responses indicated, “Our students that have autism primarily need high levels of engagement in order to be most successful in the world around them. Their self-directed nature impedes their inability to naturally engage both socially and academically.” She further indicated that, “Intentional instructional strategies and supports designed to increase engagement are most effective in increasing alertness, awareness, and learning. The teacher indicated that the time she allotted for researchers and teachers was adequate and that resources and supplies were reasonable in terms of costs. When asked if the materials were easy to use by teachers and students she responded, “Yes! This generation of students are driven by technology therefore, it is of high interest to them.  This naturally increases their alertness which promotes engagement…and ultimately learning.” Finally, the teacher expressed that the outcomes did provide evidence on innovative technology being a “great tool” to use in a variety of teaching skills while suggesting that, the children’s excitement and interest positioned them to be more engaged and on-task.” Her responses indicated that the study was reasonable in time, cost, and resources and assisted in adding to current innovative technology research.

Results

Off-task Behavior

Figures 1, 2, and 3 present frequency of off-task behavior in the two types of instructional sessions, Superflex and Robotics-Enhanced Social Skills instruction among the three participants, Don, Georgina, and Nico, respectively.

Figure 1. Don’s off-task behavior in Superflex and Robotics Enhanced Superflex sessions

 

Figure 2. Georgina’s off-task behavior in Superflex and Robotics Enhanced Superflex sessions

Figure 3. Nico’s off-task behavior in Superflex and Robotics Enhanced Superflex sessions

 

Don.  During the Superflex condition, Don’s frequency of off-task behavior ranged from 11 to 21 instances per session, with a mean of 16.6. During the Robotics-Enhanced Social Skills Instruction condition, Don’s frequency of off-task behavior ranged from 0-17 instances per session, with a mean of 8.4.

Georgina. During the Superflex condition, Georgina’s frequency of off-task behavior ranged from 3 to 6 instances per session, with a mean of 4.8. During the Robotics-Enhanced Social Skills instruction, Georgina’s frequency of off-task behavior ranged from 0-5 instances per session, with a mean of 1.4.

 Nico. During the Superflex condition, Nico’s frequency of off-task behavior ranged from 3 to 6 instances per session, with a mean of 3.8. During the Robotics-Enhanced Social Skills instruction, Nico’s frequency of off-task behavior ranged from 0-3 instances per session, with a mean of 1.2.

Social Validity. The special education teacher indicated that high levels of engagement among learners with autism is important, which supports the goals of this study. Additionally, the special education teacher reported that the intervention was reasonable in terms of time and cost and she indicated that materials were easy to use by the classroom professionals and students. Further, the teacher shared that the students were “driven by technology” because it “is of high interest to them.” She indicated that the students were excited about the Robotics-Enhanced Social Skills instruction, which “encouraged the students to be more engaged.” Finally, she indicated that the Robotics-Enhanced Social Skills intervention was effective in decreasing off-task behaviors among the participants.

Discussion

The results indicate that off task behaviors were decreased using Robotics-Enhanced Social Skills instruction over a traditional social skills lesson. Results were educationally significant for decreasing off-task behaviors while increasing cognitive demands. Cognitive theory suggests that an individual’s knowledge is rooted in the activity and content in which it is learned (Putnam & Borko, 2000). This theory operates under the understanding that how knowledge is applied is as important as where the knowledge is applied because the learning is authentic and representative of a real-world experience. In the current study, the intervention was embedded into the authentic environment which suggests real-world application.

Another reason this intervention was successful is that using robotics is a form of problem-based learning (PBL). PBL has garnered attention due to positive outcomes in collaboration (Boaler, 1997), student engagement (Brush & Saye, 2008), critical thinking and problem solving (Merhendoller et al., 2006), which were all part of the social skills curriculum. Thus, the robotics offered repeated practice. Furthermore, PBL can offer students learning scaffolds that enrich inquiry and increase student engagement (Brush & Saye, 2000; Ertmer & Simons, 2006; Jonassen 2011; Thomas & Mergendoller 2000; Tamim & Grant, 2013). Positive results may be linked to the interactive and problem-solving nature of using robotics and coding.

Increased cognitive demands, like the addition of robotics into a lesson, are known to cause cognitive limitations and slow down  responses in  people with ASD (Rogers & Monsell, 1995) however, if ample time to prepare for the switch can help (Poljac et al., 2006).  During this study children were introduced the robots, offered visual supports, and explicit directions so cognitive shifting did not appear to be an issue. Robotics activities offer visual supports and easy to follow instructions that leave little space for interpretations. According to Geurts et al. (2009) cognitively function at higher levels when the tasks are clearly delineated and limit the need for differing outcomes.

Recommendations for Practitioners

Special educators who serve learners with autism and developmental disabilities who demonstrate off-task behavior during instructional time should consider enhancing instruction with robotics. Additionally, given that the classroom educator involved in this study provided positive feedback on the goals, outcomes, and procedures of this study, educators might consider the Beebot® and the Cozmo® for young learners with autism.

Suggestions for Future Research

In future research studies it would be valuable to add generalization and maintenance phases to see if the effects of robotics lasted over time or if the novelty of the robots would wear off. Secondly, it would be helpful to add a computational thinking measure, possibly via a task analysis for programming. Thirdly, it would be valuable to consider interval coding versus momentary time sampling. Finally, it would be valuable to videotape sessions to evaluate student engagement with the robots in the robotics-enhanced condition.

References

 Abel, E. A., Gadomski, M., & Brodhead, M. T. (2016). A brief report of time-on-task behavior in  a child with autism: Comparing material and social reinforcement in the home             environment. Psi Chi Journal of Psychological Research, 21(3), 176-182.

Alcorn, A. M., Eloise, A., Charisi, V., Mantinioti, S.  Petrovi?, S., et al. (2019). Educators’ views on using humanoid robots with autistic learners in special education settings in England. Frontiers in Robotics and Artificial Intelligence, 6, 107. doi:10.3389/frobt.2019.00107.

Anderson, S., Morris J. (2006) Cognitive behavior therapy for people with Asperger syndrome. Behavioral and Cognitive Psychotherapy, 34(3), 293-293.

Anzalone, S. M., Tilmont, E., Boucenna, S., Xavier, J., Jouen, A. L., Bodeau, N., et al. (2014). How children with autism spectrum disorder behave and explore the 4-dimensional (spatial 3D+ time) environment during a joint attention induction task with a robot.   Research in Autism Spectrum Disorders,8, 814–826. doi:10.1016/j.rasd.2014.03.002

Attwood, T. (2000). Appropriate educational placement for children with Asperger’s syndrome. Retrieved from http://tonyattwood.com/newpage2.htm.

Baron-Cohen, S. (2000). Theory of mind and autism: A fifteen-year review. In S. Baron-Cohen,  H. Tager-Flusberg, & D. J. Cohen (Eds.), Understanding other minds: Perspectives from developmental cognitive neuroscience (p. 3–20). Oxford University Press.

Barker, B. S., & Ansorge, J. (2007). Robotics as Means to Increase Achievement Scores in an Informal Learning Environment. Journal of Research on Technology in Education, 39(3),229-243.

Baron-Cohen, S., Wheelwright, S., Hill, J., Raste, Y., & Plumb, I. (2001). Reading the Mind in the Eyes’’ Test Revised Version: A Study with Normal Adults, and Adults with Asperger Syndrome or High-functioning Autism. Journal of Child Psychology and Psychiatry,42(2), 241-251.

Bauminger, N. (2002). The Facilitation of Social-Emotional Understanding and Social Interaction in High-Functioning Children with Autism: Intervention Outcomes. Journal of Autism and Developmental Disorders, 32(4), 283-98. Doi: 10.1023/A:1016378718278.

Bertoglio, K., & Hendren, R. (2009). New developments in autism. Psychiatric Clinic of North   America, 32(1), 1-14. doi: 10.1016/j.psc.2008.10.004.

Bolton, J. B (2010). Examining the effectiveness of a social learning curriculum for improving social skills and self-regulation behaviors in middle school boys with autism spectrum disorder or social skill deficits.  PCOM Psychology Dissertations. Paper 14

Carey, T. (February 15, 2011) Unpopular children: Why don’t they like me? The Independent. Retrived form: www.independent.co.uk/life-style/health-and-families/features/unpopular- children-why-dont-they-like-me-2214956.html#.

Carnahan, C., Musti-Rao, S., & Bailey, J. (2009). Promoting Active Engagement in Small Group Learning Experiences for Students with Autism and Significant Learning Needs. Education and Treatment of Children, 32(1), 37-61.

Center for Disease Control (2018) National Center for Health Statistics. Health, United States, 2018. Hyattsville, Maryland.

Chen, N. S., Hung, I. C., & Wei, C. W. (2010). Developing Ubiquitous Learning System with Robots for Children’s Learning. In G. Biswas, D. Carr, Y. S. Chee & W. Y. Hwang (Eds.), Proceedings of the 3rd IEEE International Conference on Digital Game and Intelligent Toy Enhanced Learning (DIGITEL 2010) (pp. 61-68). Kaohsiung City, Taiwan: IEEE Computer Society Press.

Chung, P. Y., Chang, C. J., Liang, Y. D., Shih, B. Y., Lin, M. Z., Chen, T. H., et al. (2010). Design, Development and Learning Assessment by Applying NXT Robotics Multi- Media Learning Materials: A Preliminary Study to Explore Students’ Learning Motivation. World Academy of Science, Engineering and Technology, 65, 1102-1106.

Christensen, D. L.,  Baio, J., Van Naarden Braun, K., Bilder, D., Charles, J., Constantino, J. N., et al… Centers for Disease Control and Prevention (CDC).(2016). Prevalence and Characteristics of Autism Spectrum Disorder Among Children Aged 8 Years–Autism and      Developmental Disabilities Monitoring Network, 11 Sites, United States, 2012. Morbidity and Mortality Weekly Report Surveillance Summary, 65(3), 1-23. doi: 10.15585/mmwr.ss6503a1.

Crooke, P., & Winner, M. (2016). Social Thinking® Methodology: Evidence-Based or Empirically Supported? A Response to Leaf et al. (2016). Behavioral Analysis in Practice, 9(4), 403-408. doi: 10.1007/s40617-016-0151-y.

Crooke, P., Hendrix, R., Rachman, J. (2008). Brief Report: measuring the effectiveness of teaching social thinking to children with Asperger syndrome (AS) and High Functioning Autism (HFA). Journal of Autism and Developmental Disorders, 38(3), 581-591.

Doctoroff, S. (1997). Sociodramatic script training and peer role prompting: Two tactics to promote sociodramatic play and peer interaction. Early Childhood and Developmental Care, 136, 27-43.  

Dautenhahn, K. (1999). “Robots as social actors: aurora and the case of autism,” in Proceedings CT99, the 3rd International Cognitive Technology Conference (San-Francisco, CA), 359–374.

Dautenhahn, K., & Werry, I. (2004). Towards interactive robots in autism therapy: Background, motivation and challenges. Pragmatics and Cognition,12, 1-35. doi:10.1075/pc.12.1.03dau.

Dowd, T., & Tierney, J. (2005). Teaching social skills to youth. Boys Town: Boys Town Press           

Geiger, K., Carr, J., & LeBlanc, L. (2010). Function-based treatments for escape-maintained problem behavior: A treatment-selection model for practicing behavior analysts. Behavioral Analysis Practices, 3(1), 22-32. doi: 10.1007/BF03391755.

Fasola, J., & Mataric, M. J. (2010). Robot Motivator: Increasing user enjoyment and performance on a physical/cognitive task. Paper presented at Proceedings of the Ninth  IEEE International Conference on Development and Learning (ICDL 2010), Ann Arbor, MI: IEEE Computer Society Press.

Goldstein, H., & Cisar, C. L. (1992). Promoting interaction during sociodramatic play: Teaching scripts to typical preschoolers and classmates with disabilities. Journal of Applied Behavior Analysis, 25, 265-280. 

Gray, C. A., & Garand, J. D. (1993). Social stories: Improving responses of students with autism with accurate social information. Focus on Autistic Behavior, 8, 1–10. 

Gray, C. (2000). The new social stories book: Illustrated edition. Arlington, TX: Future Horizo. 

Gresham, F.M. (2002). Best practices in social skills training. In Best Practices for School  Psychologist, Volume 2 (4th ed). Thomas, A., & Grimes, J. (eds). Bethesda, MD: National Association of School Psychologists.

Rao, P.A., Beidel, D.C., & Murray, M.J. (2008). Social skills interventions for children with Asperger’s Syndrome or High-Functioning Autism: A review and recommendations. Journal of Autism and Developmental Disorders, 38, 353-361.

Liss, M., Saulnier, C., Fein, D., & Kinsbourne, M. (2006). Sensory and attention abnormalities in autistic spectrum disorders. Autism, 10, 155-172

Madrigal, S. & Winner, M. (2008). Superflex®. Santa Fe, CA: Think Social.

Madrigal, S., Winner, M. G., & Knopp, K. (2009) Superflex® takes on glassman and the team of  unthinkables. Santa Fe, CA: Think Social.

Mundy, P., Sigman, M., & Kasari, C. (1994). Joint attention, developmental level, and symptom presentation in autism. Development and Psychopathology, 6(3), 389-401. doi: 10.1017/S0954579400006003.

Smith Myles, B., Trautman, M. L., & Schelvan, R. L. (2004). The hidden curriculum. Kansas: Autism Asperger Publishing

Nowell, S. W., Watson, L. R., Boyd, B., & Klinger, L. G. (2019). Efficacy Study of a Social  Communication and Self-Regulation Intervention for School-Age Children with Autism Spectrum Disorder: A Randomized Controlled Trial Language, Speech, and Hearing Services in Schools, 1-18. doi:10.1044/2019_LSHSS-18-0093. 

Parsons, S., & Mitchell, P. (2002). The potential reality in social skills training for people with  autistic spectrum disorders. Journal of Intellectual Disabilities Research, 46, 430-443.

Patterson, D. S., Jolivette, K., & Crosby, S. (2006). Social skills training for students who demonstrate poor self-control. Beyond Behavior, 15(3), 23-27.

Patten, E., & Watson, L. (2011) Interventions targeting attention in young children with autism. American Journal of Speech-Language Pathology, 20(1), 60-9. doi: 10.1044/1058-0360

Paynter, J., & Peterson, C. C. (2013). Further evidence of benefits of thought-bubble training for theory of mind development in children with autism spectrum disorders. Research in Autism Spectrum Disorders, 7(2), 344–348. doi: 10.1016/j.rasd.2012.10.001.

Pierson, M. R., & Glaeser, B. C. (2005). Extension of research on social skills training using comic strip conversations to students without autism. Education and training in Developmental Disabilities, 279-284. 

Straten, C. L., Smeekens, I., Barakova, E., Glennon, J., Buitelaar, J., & Chen, A. (2018). Effects of robots’ intonation and bodily appearance on robot-mediated communicative treatment outcomes for children with autism spectrum disorder. Pearson. Ubiquitous Computing,     22, 379–390. doi: 10.1007/s00779-017-1060-y.

Reynolds, C. R., & Kamphaus, R. W. (2004). BASC-2: Behavior Assessment System for Children (2nd ed.). Circle Pines, MN: American Guidance Service.

Rogers, S., & Pennington, B. (1991). A theoretical approach to the deficits in infantile autism. Developmental Psychology, 3, 137–162.

Stevenson, C. L., Krantz, P. J., & McClannahan, L. E. (2000). Social interaction skills for children with autism: A script-fading procedure for nonreaders. Behavioral Interventions, 15(1), 1–20.

Sze, K. M., & Wood, J. J. (2008). Enhancing CBT for the treatment of autism spectrum disorders  and concurrent anxiety: A case study. Behavioral and Cognitive Psychotherapy, 36, 403- 409.

 Vanvuchelen, M., Roeyers, H., & de Weerdt, W. (2011). Do imitation problems reflect a core characteristic in autism? Evidence from a literature review. Research in Autism Spectrum Disorders, 5, 89–95.  doi: 10.1016/j.rasd.2010.07.010.

Welsh, M., Parke, R. D., Widaman, K., & O’Neil, R. (2001). Linkages between children’s social  and academic competence: A longitudinal analysis. Journal of School Psychology, 39(6), 463–482. doi: 10.1016/S0022-4405(01)00084-X.

Wolf, M. M. (1978). Social validity: the case for subjective measurement or how applied behavior analysis is finding its heart. Journal of Applied Behavior Analysis, 11, 203-214. 

Wong, T. A., Fairn,  G. D., Poon, P. P., Shmulevitz, M., McMaster, C., Singer, R. A., & Johnston, G. C. (2005)Membrane metabolism mediated by Sec14 family members influences Arf GTPase activating protein activity for transport from the trans-Golgi. Proceedings of the       National Academy of Sciences, USA, 102(36), 12777-12782. doi:10.1073/pnas.0506156102

Yadlosky, K. (2012). Effects of the Superflex®™ curriculum on the social cognition of primary students with attention deficit hyperactivity disorder and autism spectrum disorders. http://hdl.handle.net/10211.2/1231.