Presented at ISEC 2000

Let's Play : Children with Autism and Their Play Partners Together

Serene (Hyun-Jin) Choi - Schonell Special Education Research Centre, University of Queensland, Australia

Contributions from: Anne Jobling and Annemaree Carroll.

Abstract

Since 1990, a debate has simmered over how best to reform special education services for "full inclusion". Does physical mainstreaming mean full inclusion?

Many children with developmental disabilities, especially autism, have limited interactions with others. They can be described as outsiders, lonely, or disruptive by typically developing peers. If typically developing peers do not know what, when, and how children with autism express themselves, social interactions between these two groups may not occur. Thus reciprocal exchanges may be infrequent, unsatisfactory, and limiting for both groups in inclusive education settings.

As a design teaching strategy for inclusive schooling, this paper will present the development of a peer-mediated play strategy aimed at facilitating appropriate social interactions between children with autism and cross aged peer play partners. The paper will examine the following: 1) the selection of peer play partners using the class teacher's recommendations and a standard social scale rating system; 2) the development of the play training package; and 3) the play training package to train the peer play partners. It is considered that this strategy could significantly assist the interactions of children with autism and their peers in inclusive educational settings.

1. Rationale for the study

Many children with disabilities are often unpopular among their playmates. They seem to struggle with exclusion from games and activities with their same-age peers. Cutts (1998) states that educational practitioners including school staff, and parents who have children with disabilities report that children with disabilities are teased, ignored, and rejected by typically developing children. Children with disabilities often express the feeling that they are lonely, isolated, and less confident to make friends. Typically developing children also describe classmates with disabilities as outsiders, lonely, or disruptive.

Why are children with disabilities often teased or rejected by typically developing children? Why do typically developing children have negative attitudes toward children with disabilities? Why do these undesirable interaction relationships occur between them? Why do children with disabilities fail to be included in inclusive schooling? Is it because of the lack of opportunities to interact with each other? Due to the lack of chances to meet each other in segregated education settings? Why then does it occur in inclusive settings as well? Is it possible to establish desirable social interaction between all children? How? Will this occur only when children with disabilities master appropriate interaction skills? If they cannot master appropriate interaction skills, are they unable to make friends with typically developing children? Why does the child with a disability still play alone, even though she/he has learned the skills?

Many children with disabilities have limited interactions with peers when they play together (Guralnick & Weinhouse, 1984). Children with disabilities have a variety of obstacles to effective communicative interaction with others due to their physical limitations or cognitive problems. Also, some children with disabilities require special methods (eg. a communication board) to interact or communicate with others. If typically developing children, even typically developing adults, do not know what kinds of difficulties the children with disabilities have and how they express themselves and interact with others, social interactions between children with and without disabilities may not occur. Social interaction, as the term itself indicates, means reciprocal exchanges, not one-way relationships.

There may be many more opportunities and possibilities of maintaining social interaction between children with disabilities and typically developing children, if typically developing children support and encourage children with disabilities in their play activities. Barriers for social interaction between children with disabilities and typically developing children can be broken down by the support and encouragement from typically developing children in play. Simultaneously, the new relationship called 'friendship' may be formed through the activities between them.

In developmental disability groups, individuals with autism exhibit a fundamental failure in social development (American Psychiatric Association, 1994). The qualitative aspect of the social play behaviours of children with autism seems to be affected by the lack of close relationships, and lack of appropriate language and social models. From the first years in life, they show deviant patterns of reciprocal gaze behaviour, and an absence of expected patterns of attachment and social relatedness.

Later in life, they may fail to develop friendships and thus have a limited capacity to interact with other people (Rumsey, Rapoport, & Sceery, 1985). Their impairment of social communication has been labeled as "semantic and pragmatic deficits" (Tanguey, 1990). In addition, the absence of limited babbling, failure to use vocalizations for social engagement including vocal turn-taking and vocal imitation, and monotonic sound have been commonly reported in the early history of children with autism (Prizant, 1996). Their language, even if syntactically correct when used, did not seem to be used as a method of communication and social interchange (Tager-Flusberg, 1981). Children with autism show deficits in comprehending nonverbal communication such as gestures (Buffington, Krantz, McClannahan, & Poulson 1998), facial expression (Celani, Battacchi, & Arcidiacono, 1999), social perception (Pierce, Glad, & Schreibman, 1997), and voice intonations (Garfin & Lord, 1986). Children with autism appear to be relatively deficient in role-taking tasks and social competence measures that are related to social interaction difficulties. These impairments may result from deficits in social cognitive abilities (Oswald & Ollendick, 1989).

Research using peer mediation has been implemented to examine social interaction skills and their generalization in children with autism in play activities (Belchic & Harris, 1994; Lee & Odom, 1996; Pierce & Schreibman, 1995, 1997; Strain, 1997; Strain & Kohler, 1995). Much of the research has reported success in improving the social interaction skills in children with developmental disabilities when typically developing peers were employed as peer tutors. The peer tutors engaged in a training program to learn the skills to invite the target children into play, to maintain social interactions during their activities, and to extend social interaction skills in the target children. The research has also recommended using multiple peers as play partners (Belchic & Harris, 1994; Pierce & Schreibman, 1997) and natural play settings to enhance generalization. Peer mediated intervention was positively compared to teacher based intervention in aspects of longer interaction and greater generalization (Teitzel & Terkelsen, 1997).

2. Purpose of the study

This study has five main functions. These are:

1) To design and examine the effectiveness of peer mediated play intervention which could facilitate the development of appropriate play skills and positive social interaction skills in children with autism.

2) To explore which categories of cognitive play (functional, constructive, pretend, and games with rules) can be easily facilitated for children with autism in response to peer mediated-play intervention.

3) To examine the correlation between improvements in cognitive play skills and improvements in social interaction skills in children with autism.

4) To find what kinds of interaction skills are used between autistic children and peer play partners in indoor and outdoor playing.

5) To investigate possible generalizations of the acquired play and social interaction behaviours by children with autism as a result of play interaction with trained and untrained peer play partners.

3. Methodology of the study

1) Selection of Participants

i) Children with autism Five young children with typical autism who are 4 to 6 years old and attending segregated special education settings such as Special Education Developmental Units (SEDU) in Brisbane in Australia were chosen. The Childhood Autism Rating Scale (Schopler, Reichler, & Renner, 1988) was selected for assessing children with autism in preference to medical diagnosis. Transdisciplinary Play-Based Assessment (Linder, 1933) was modified to measure cognitive abilities, communication skills, as well as the play and social skills of children with autism in 'general' play activities through observation techniques and discussion with teachers. In addition, a questionnaire for parents was used to collect detailed information about play behaviours in children with autism.

ii) Peer Play Partners Thirteen eight to nine year old typically developing children who attend a primary school associated the special education settings were chosen to participate in the study as cross-aged peer play partners. These play partners were chosen through the recommendation of their class teachers, with their own and their parents' consent. A standard social interaction scale, Social Skills Rating System (Gresham & Elliott, 1990), was used to measure general social skills in the peer play partners. The peers' gender, number of siblings, and socio-economic status (SES) of the family were considered in order to match the demography of the children with autism. In the first visit to the classroom, the researcher observed and discussed with their class teacher their behaviours in school activities.

2) Procedure

An AB design with two groups (trained peer play partners and untrained peer play partners) in two situations (indoor setting and outdoor setting) was used in this single subject experimental research.

This study was divided into 5 phases of data collection: pre-observation and orientation (preparation phase); participants' natural play sessions (baseline phase); play training sessions for half of the peers (play training phase); comparison play sessions between dyad groups with trained peer play partners and another dyad group with untrained peer play partners (intervention phase); and follow-up questionnaires for teachers and parents to collect any changing play behaviour in children with autism and peer play partners (follow-up phase).

C/A = Child with autism

PPP = Peer play partners: TPPP=Trained Peer play partners

UTPPP=Untrained Peer play partners

3) Settings and play objects

Play sessions between children with autism and peer play partners took place at the special education setting where the children with autism attend and the primary school setting where peer play partners attend. Because this study was implemented in natural education settings, not a laboratory, the setting of the play session depended on the educational contexts such as extra room facilities, school time schedule, and school activities.

Play sessions were carried out in an outdoor area (e.g. playground) and an indoor area (e.g. a classroom). In the outdoor play setting, play equipment including simple sports equipment in the playground and sandpit play were used for play in a functional, constructive, pretend, or games with rules manner. Common play objects used in a functional, constructive, pretend, or games-with-rules manner in preschool education settings were set up for indoor play sessions. Usually 4-5 toy sets including cars, blocks, dolls, cooking toys, and matching cards were set up. One set of toys was considered as a representation for each play category. For example, blocks may be easily used in a constructive manner.

However, when some children with autism showed appropriate functional and constructive play manner in the cognitive dimension of play, four toy sets used mainly in pretend or games with rules manner were set up. There were two reasons for this. One purpose was to investigate the improvement in cognitive play skills that were weak play skills in the children with autism. Therefore, if the children with autism already had appropriate cognitive play skills, they were not included in this study. Another reason was that peer play partners did not have enough time to deal with the persistence of children with autism due to the limited project duration. If children with autism persist in using only one toy because they can use it, it was not expected that the peer play partners were to deal with this in this study.

4) Play groups and play time

One play group consisted of one child with autism and one peer play partner (dyad play group). One child with autism had two play partners (one trained and one untrained). Each dyad play group had two play sessions, indoor and outdoor play for about six minutes each. However, when one child did not want to keep playing in the session, the researcher recommended continuing play. If the child still did not want to keep playing, the session was stopped. On the other hand, if the dyad group wanted to keep playing longer, then they were allowed to play for a couple of minutes more.

Half of the peer play partners had training in the training phase. The training play partners were chosen randomly and play sessions were arranged for them according to their school schedule. After the training phase, peer play partners were divided into two groups, trained peer play partners and untrained peer play partners. In the intervention phase, each child with autism had the same play sessions as they did in the baseline phase.

4. Data collection

Every session of the study was videotaped by the researcher. Each play dyad (the child with autism and the peer play partner) was told "It is play time with friends! You can play with these toys and have fun!" The researcher typically followed the dyad group to record the play session and to inform the children of available play objects if they used only one item. In natural school environments, other objects were set up around the dyad play group for teaching by other staff. Even though the researcher let the dyad group know which objects were available for this study, the dyad group sometimes needed to be reminded in the sessions. Furthermore, teacher aids from special education development unit were involved for safety reasons. The teacher aid took the children with autism to the primary school and sat in the corner when the play session occurred. If difficulties arose between the children such as fighting, the teacher aid was supposed to take care of it.

Four months after the peer play program was completed, a follow-up interview with teachers of the children with autism, the teacher of the peer play partners, and the autistic children's parents was conducted, utilising a questionnaire form. In addition, telephone interviews with parents, and face-to-face interviews with teachers, were conducted along with the questionnaires. Any changes in play behaviours of children with autism before and after the peer play program were discussed. Any evidence of the level of understanding about disabilities in the peer play partners after participating in the program was ascertained from the teachers of the peer play partners.

5. Data coding and analysis

The play data from all play dyad groups were categorized in three ways: first, cognitive dimensions based on Piaget's theory (1962), Smilansky's theory (1968), and Vygotsky's theory (1962); second, social dimensions modified on Parten's categories (1932); thrid, social interaction skills in play were developed by the researcher from the literature (Kohler, Strain, Maretsky, & DeCesare, 1990; Odom, Hoyson, Jamieson, & Strain, 1985; Sainato, Goldstein, & Strain, 1992).

6. Design for play training package and training for peer play partners

A training package was designed for the peer play partners to train play and interaction skills. In the package, play skills and strategies were adopted from the work of Pierce and Schreibman (1995) and set up as follows: paying attention to target children; giving choice of play materials to target children (e.g., "Which toy do you want to play with, this car or this doll?"); modeling play activities including verbal statements by peer players (e.g., "This game is fun, just like this…"); narration of play activity to target child (e.g., "I'm going to cook the pizza," or "You drive a car to go to the shop."); and, reinforcement of attempts to encourage and extend appropriate play behaviours and social interaction skills in target children (e.g., "Great bouncing!" or "Thank you for sharing the car with me.").

The training package included 16 training sessions. It consisted of watching videotapes, discussions, role-playing, worksheets, and play practices. Videotapes that show general information about friendship, differences in people, and typical behaviours of autistic children were used to help the peer play partners' understanding of play interaction with children with autism. Discussion about play and social interaction was guided in a player-centered manner. The researcher supported and facilitated their discussion by simple questions under the theme of each session. The topic of discussion was started with general friendship in typically developing children's lives, and expanded to special friendships with children with disabilities, especially children with autism. Also, it focused not on 'what children with disabilities are not able to do due to their disabilities' but 'what children with disabilities are able to do.' The researcher guided the peer play partners to develop ways to play with children with autism. The peer play partners exchanged their experiences and ideas of play with their siblings or other friends in order to use these when playing with the children with autism. Role-playing was also used to assist the peer play partners to understand more about the playing situations. While peer play partners experienced disabilities, they could recognize the abilities and different needs of individuals with disabilities from individuals without disabilities. The worksheet enabled peer play partners to summarize the session.

On the general understanding and experiences of disabilities, peer play partners practised and elaborated the play skills and strategies by performing role plays. When one peer play partner demonstrated the play skills and strategies, other peer play partners monitored him/her and scored them on the Play Note. This monitoring activity was also carried out visa versa. When one peer play partner demonstrated play skills and strategies, others found out what skill was used and scored on Play Note. Eventually, the researcher role-played an autistic child. The peer play partners then performed the play skills and strategies learned with the researcher. All these practices were carried out with fun!

During 16 intervention sessions, 5 training sessions were performed on 'fading' schedule (one to two days without feedback and the 3rd day given feedback, four to six day without feedback and the 7th day given feedback, etc) between trained peer play partners and the researcher before or after the session for about 20 minutes. The written play skills and strategies were also read by the peer play partners before they played with the children with autism. The peer play partners watched videotapes that showed the play activities with the children with autism, and monitored their play skills and strategies on the Play Note. They were allowed to bring along their friends to watch the play tapes together if they wanted. Also, untrained peer play partners watched their own play sessions of tapes with their friends. However, they were not provided with the play skills and strategies of the Play Note. All peer play partners enjoyed watching the videotapes and their friends wished to be involved in the play program. It was good time for all peer play partners to show off as actors or actresses with their movies!

7. Implications of the study

1) Benefits of the peer mediated play program Since 1990, a debate has simmered over how best to reform special education services for 'full inclusion'. Physical integration can be the first step towards realizing the concept of full inclusion for children with disabilities (Skinner, 1996). However, unless appropriate programs are used for both children with disabilities and typically developing children, qualitatively effective interaction, including friendship between them, might not develop (Fuchs & Fuchs, 1995; Shanker, 1995; Tansey, 1995). Such a peer play program can be thought of as "software" necessary to make proper use of the "hardware" of physical integration.

The peer mediated play program developed in the present research may provide benefits not only to children with autism and play partners without disabilities but also to teachers. While children with autism and typically developing children are playing together, special education teachers could utilize their time for other tasks. It is hypothesized that children with autism would learn to use play objects and enjoy play activities and learn to interact with play partners through this play program. Typically developing children who are play partners would acquire several social cognitive skills by interacting with children with autism in this research. They are pro-social behaviours such as helping others; leadership, such as management skills for interpersonal relationships; and, investigating skills for social science subjects such as "Studies of Society and Environment", particularly the aspect related to culture and diversity. The findings from this study are expected to contribute to strategies that may be applied to facilitate full inclusion for children with autism.

2) Limitations of the study

i) Segregated settings and cross-aged peer play partners The study involved segregated educational settings such as special education developmental units, special schools, and primary schools. Since children with autism are spread around local preschools in Brisbane, it was not possible for the researcher to implement the program in each local preschool due to time limitations. Compared to local preschools, a number of children with autism are in SEDUs where typically developing children are nearby. There are no activities for interaction between children with disabilities and typically developing children.

Due to the choice of the segregated settings, same aged peers could not be recruited. However, there was another reason to have cross-aged peers. Cross-aged peers were considered as more cognitively and socially mature play partners to understand the play training within the limited time schedule for the study. If same-aged peers were involved in the study, the research duration might need to be longer. Therefore, the play program may need to be implemented in inclusive educational settings by teachers who have typically developing children and children with disabilities.

ii) Researcher with video camera and teacher aid Play sessions were carried out in spare rooms and playgrounds of the natural educational settings. The researcher recorded the play sessions with a small video camera beside the dyad play groups. Although the video camera is hand-sized, it may cause players to show unnatural play interactions. For example, typically developing children were aware of the video camera, and tried to perform some different actions from natural play behaviours. On the other hand, the video camera was used for a child with autism as a reinforcement to keep waiting for his play partners. While he was waiting for the next play partner, he got bored easily and did not want to stay in the play setting. The researcher showed his play sessions on the screen of the video camera, and he enjoyed watching it for about three minutes untill next play partner came.

Due to safety reasons, the teacher in charge let a teacher aid attend all play sessions. This may have influenced play behaviours in players. For example, a typically developing child was extremely shy when the teacher aid attended the play sessions. She said that she could not use the play skills and strategies if somebody watched her. When the teacher aid was removed from the session, she played with the partner with autism with confidence including proper volume of voice. Also, the child with autism was less resistant to his partner's attempts for interaction. When the teacher aid attended, he used to hug the teacher aid and avoided his partner's attempts at interaction. When the teacher aid was removed, he behaved in the same way towards the researcher. The researcher ignored him and he complied with his peer partner.

8. Conclusion

Data presently being analysed. Preliminary observation results show encouraging patterns of development in play and social interaction skills between children with autism and trained peer play partners.

References

American Psychiatric Association (4th Ed.) (1994). 'Diagnostic and Statistical Manual of Mental Disorders'. Washington, DC: American Psychiatric Association.

Belchic, J. K. & Harris, S. L. (1994). The use of multiple peer examplars to enhance the generalization of play skills to the siblings of children with autism. Child and Family Behavior Therapy, 16(2), 1-25.

Buffington, D. M., Krantz, P. J., McClannahan, L. E., & Poulson, C. L. (1998). Procedures for teaching appropriate gestural communication skills to children with autism. Journal of Autism and Developmental Disorders, 28, 535-546.

Celani, G., Battacchi, M. W., & Arcidiacono, L. (1999). The understanding of the emotional meaning of facial expressions in people with autism. Journal of Autism and Developmental Disorders, 29, 57-66.

Cutts, S (1998). The social competence and peer interactions of Adolescents with intellectual impairments in a regular high school. Master thesis. The University of Queensland.

Fuchs, D. & Fuchs, L. S. (1995). Sometimes separate is better. Educational Leadership, 52(4), 22-26.

Garfin, D. G. & Lord, C. (1986). Communication as a social problem in autism. In E. Schopler & G. B. Mesibov (Eds.). Social behavior in autism. (pp. 133-152). New York: Prenum Press

Gresham, F. M. & Elliott, S. N. (1990). Social Skills Rating System. USA: American Guidance Service.

Guralnick, M.J. & Weinhouse, E. (1984). Peer-related social interactions of developmentally delayed young children: Development and characteristics. Developmental Psychology, 20(5), 815-827.

Kohler, F. W., Strain, P, S., Maretsky, S., & DeCesare, L. (1990). Promoting positive and supportive interactions between preschoolers: An anlysis of group-oriented contingencies. Journal of Early Intervention, 14(4), 327-341.

Lee, S. & Odom, S. L. (1996). The relationship between stereotypic behavior and peer social interaction for children with severe disabilities. Journal of the Association for Persons with Severe Handicaps, 21, 88-95.

Linder, T. W. (1933) Transdisciplimary Play-Based Assessment: A functional Approach to Working with Young Children, Paul H. Brookes: Baltimore.

Odom, S. L., Hoyson, M., Jamieson, G., & Strain, P. S. (1985). Increasing handicapped preschoolers' peer social interactions: Cross-setting and component analysis. Journal of Applied Behavior Anaysis, 18(1), 3-16.

Oswald, D P. & Ollendick, T. H. (1989). Role taking and social competence in autism and mental retardation. Journal of Autism and Developmental Disorders, 19(1), 119-127.

Parten, M. (1932). Social participation among preschool children, Journal of Abnormal and Social Psychology, 27, 243-269.

Piaget, J. (1962). Play, dreams and imitation in childhood. New York: Norton.

Pierce, K., Glad, K. S., & Schreibman, L. (1997). Social perception in children with autism: An attentional deficit? Journal of Autism and Developmental Disorders, 27, 265-282.

Pierce, K. & Schreibman, L. (1995). Increasing complex social behaviors in children with autism: Effects of peer-implemented Pivotal Response Training. Journal of Applied Behavior Analysis, 28(3), 285-295.

Pierce, K. & Schreibman, L. (1997). Multiple peer use of pivotal response training to increase social behaviors of classmates with autism: Results from trained and untrained peers. Journal of Applied Behavior Analysis, 30, 157-160.

Prizant, B. M. (1996). Brief report: Communication, Language, Social, and emotional development. Journal of Autism and Developmental Disorders, 26(2), 173-178.

Rumsey, J. M., Rapoport, J. L., & Sceery, W. R. (1985). Autistic children as adults: psychiatric, social, and behavioral outcomes. Journal of American Academy of Child Adolescent Psychiatry, 24, 465-473.

Schopler, E., Reichler, R. J., & Renner, B. R. (1988). The Childhood Autism Rating Scale. USA: Western Psychological Services.

Shanker, A (1995). Full inclusion is neither free nor appropriate. Educational Leadership, 52(4), 18-21.

Skinner, M. F. (1996). Full inclusion and students with disabilities: One size fits all? Reading and Writing Quarterly: Overcoming Learning Difficulties, 12(2), 241-244.

Smilansky, S. (1968). The effects of sociodramatic play on disadvantaged preschool children. New York: John Wiley.

Strain, P. S. (1997). An experimental analysis of peer social initiations on the behavior of withdrawn preschool children: Some training and generalization effects. Journal of Abnormal Child Psychology, 5, 445-455.

Strain, P. S. & Kohler, F. W. (1995). Analyzing predictors of daily social skill performance. Behavioral Disorders, 21, 79-88.

Tager-Flusberg, H. (1981). On the nature linguistic functioning in early infantile autism. Journal of Autism and Developmental Disorders, 11(1), 45-56.

Tanguey, P. E. (1990). Early infantile autism: What have we learned in the past 50 years? Brain dysfunction, 3(5), 197-207.

Tansey, K (1995). This can't be my responsibility: it must be yours! An analysis of a reintegration programme for a school refuser. British Journal of Special Education, 22(1), 12-15.

Teitzel, T & Terkelsen, J. (1997). A PEER IN. N.S.W.: Tweed District Early Intervention Centre.

Vygotsky L. S. (1962). Thought and language. (E. Hanfman & G. Vakai, Trans.). Cambridge: MIT Press. (Original work published 1934).

Werts, M. G., Caldwell, N. K., & Wolery, M. (1996). Peer modeling of response chains: Observational learning by students with disabilities. Journal of Applied Behavior Analysis, 29(1), 53-66.

Index