A165: An Experimental Study on the Influence of Peer-MediatedIntervention on Social Interaction of Autistic Children

<bold>Purpose:</bold> Based on theories of human movement development and behaviorism, this study investigated the effects of a peer-mediated physical education intervention on the social interaction abilities of children with autism and provided recommendations for the development of large muscle group movements in children with autism. <bold>Method:</bold> This study was conducted in a special education school in Changchun. A sample of 14 male children diagnosed with autism aged 5–7 years with mild and moderate symptoms were selected and paired according to their social and motor performance and familiarity with each other. Seven peer groups were formed for an 8-month peer-intervention physical education intervention, which took the form of a large-muscle-group movement game. In this study, we collected data on the motor ability of the autistic children through the TGMD-3 scale and their social interaction ability through the SRS scale. We analyzed the data through the paired samples t-test using SPSS 26.0. <bold>Results:</bold> (1) The social interaction ability of autistic children was significantly higher after the intervention (<italic>p</italic> &lt; 0.001). Among them, social cognition (<italic>p</italic> &lt; 0.001), social communication (<italic>p</italic> &lt; 0.001), and social motivation (<italic>p</italic> &lt; 0.001) had a high increase in autistic children, while behavioral approach (<italic>p</italic> = 0.001) had a small increase, and social perception had the smallest increase (<italic>p</italic> = 0.029). (2) Displacement motor and object manipulation abilities of children with autism were significantly higher after the intervention than before (<italic>p</italic> &lt; 0.001). For displacement motor ability, autistic children showed very significant gains in side-sliding (<italic>p</italic> &lt; 0.001) versus forward-sliding (<italic>p</italic> &lt; 0.001), larger gains in running (<italic>p</italic> &lt; 0.001) and standing long jump (<italic>p</italic> &lt; 0.001), and smaller gains in one-legged hopping (<italic>p</italic> = 0.001) and bungee jumping (<italic>p</italic> = 0.006), which are difficult to accomplish for autistic children. In terms of object manipulation, the post-intervention improvement in two-handed bat swinging to hit a stationary ball was very significant for children with autism (<italic>p</italic> &lt; 0.001), with larger improvements in kicking a stationary ball (<italic>p</italic> &lt; 0.001), over-the-shoulder pitching (<italic>p</italic> &lt; 0.001), and two-handed catching (<italic>p</italic> = 0.000, <italic>p</italic> &lt; 0.001), with smaller improvements in underhand throwing (<italic>p</italic> = 0.001), and the smallest improvement in one-handed bat swinging to hit a bouncing ball (<italic>p</italic> = 0.008), making it more difficult for children with autism. <bold>Conclusion:</bold> (1) Peer-mediated sports intervention had a significant effect on the improvement of social interaction and motor skills of children with autism. (2) The weekly intervention program followed the principle of gradual progression with incremental changes, and the design of the teaching content should reflect the law of from easy to difficult.