Teaching Social Skills: Peer Network Interventions in Inclusive Elementary School Settings

By Suzanne Murphy 

It is a well-known fact that people with an autism spectrum disorder (ASD) face significant challenges in the areas of social skills and social interactions. This is a real concern for parents, school staff and community service providers alike. Deficits and social differences are evident from an early age. Individuals with an ASD can show a range of social characteristics, from aloofness and avoiding social situations to a desire for friendships and actively seeking social interactions although usually lacking the skills to do this successfully.

Social deficits and the failures that result from them impact the short-term and long-term well being as well as quality of life of individuals with ASD. Where to start? What to teach? What is the most effective level of intervention? How to ensure generalization of skills from one environment to another or from one person to another? These are only some of many questions that arise when attempting to address the multiple social needs of individuals with ASD.

Another type of question touches upon peer involvement in social learning.  How can the importance of peers in social learning be underlined?  What structures are necessary to fully benefit from this naturally occurring social network?  During her presentation at the 2012 Association for Behavior Analysis International (ABAI) Autism Conference, Debra M. Kamps (Ph. D.) associate director and senior scientist at the University of Kansas clearly presented the necessity for social intervention starting at a young age. Peer mediated interventions are a key strategy for social learning. Circles of Friends and social skills groups made up of children or youth with ASD and neurotypically developing peers are two examples of this type of intervention.

According to Kamps, neurotypical peers interact with one another about 40 percent of the time, which is not the case for children and youth with ASD. She shared results of research which examined the impact of peer networks, defined as small groups of peers with specific roles including offering support and assistance, tutoring or acting as reinforcing agents with the students with ASD.  Intervention based on peer networks differs from other interventions in two important ways: training of peers and peer involvement in the choice of activities and goals. Seventeen studies involving forty-five participants attributed progress in social communication and exchanges to the length of interactions between the student with ASD and their peer network and to the fact that peers considered themselves ‘agents of change’.

Reasons for training peers are many, most notably that they are part of the natural environment. Peers spend the entire school day with the student with ASD and they move up together from year to year. However, the fact of being physically in the same classroom offers no guarantees of successful interactions or positive attitudes towards students with special needs. Furthermore, students with special needs are more likely to be victims of bullying than their neurotypical peers. Peers play an important role in social learning, but without guidance they may unknowingly encourage interfering behaviour of the student with ASD or fail to reinforce desired behaviour.

Results of a five-year longitudinal study, carried out by Kamps and her colleagues, demonstrated the positive impact that peer networks can have. The study followed forty-five students with ASD in elementary school and each student was part of one to four peer networks per year. Peer networks included social groups or games, peer tutoring sessions and recess or lunch groups and were active three to four times per week. Results showed significant differences in interactions between the students with ASD and peers who received training in comparison to interactions with unfamiliar or untrained peers.  Interval recording demonstrated interaction in 30% of intervals with unfamiliar peers, 40% of intervals with familiar peers but almost 70% of intervals with trained peers, an impressive increase. 

Kamps noted a few key evidence-based features of interventions based on peer networks:


  • Sessions and activities are structured and include one student with ASD and two to four peers;
  • Direct instruction is used to teach communication and social skills;
  • Visual supports are used and include graphics and words;
  • Scripts are used for the student with ASD and with peers;
  • Peers interact with the student with ASD. The role of the adult is to encourage peers to prompt the students with ASD to participate;
  • Reinforcement is incorporated into the intervention.


In addition to these features, Kamps underlines the following:


  • The ‘30 second rule’ is in effect. This means that if the student with ASD stops participating for more than 30 seconds, the adult prompts the peers to encourage participation. Peers can prompt the student with ASD a maximum of two times. If this does not result in participation, the adult intervenes directly;
  • Peer networks can exist with different frequencies.  It is not necessary for them to occur every day. Having said that, Kamps adds that a ‘dose’ of regular interaction is a way to overcome challenges related to autism;
  • Peer networks increase social interactions, can contribute to improved language and provide tools useful for better functioning in society.


It was also noted that although peer networks are effective, teaching staff struggle with incorporating them into the school schedule.  Kamps ended with a vision for moving forward and some questions to ponder:


  • How does one adapt interventions based on peer networks so they can be less intensive and less costly?
  • What might peer networks look like at the secondary school level and in adulthood?
  • What are target skills within community contexts?
  • What teaching strategies are essential to include?
  • What are the priorities to achieve significant results?


Although research can lead to additional questions, results of research to date support implementation of peer networks. It is impossible to negate the characteristics of ASD, the complexity and enormity of the social realm and the need to include a social component to programs designed for children with ASD.  The National Research Council (2001) recommends using peer mediated techniques to improve interactions and promote social development of children with autism.  Essential elements of peer networks are evidence-based and generally have a positive impact on the student with ASD as well as on the peers, two key parts of successful interactions.

Suzanne Murphy is consultant with over thirty years of experience in the field of child and youth work.  She specialises in the fields of autism spectrum disorders (ASD) and behaviour.

Disclaimer: This document reflects the views of the author. It is Autism Ontario’s intent to inform and educate. Every situation is unique and while we hope this information is useful, it should be used in the context of broader considerations for each person. Please contact Autism Ontario at info@autismontario.com or 416-246-9592 for permission to reproduce this material for any purpose other than personal use. © 2012 Autism Ontario  416.246.9592  www.autismontario.com.