Social Isolation or Autistic Homophily: An Exploration of the Social Networks of Autistic College Students

Our understanding of autistic socialization using social network analysis has focused on autistic children and understanding support systems. The purpose of this study is to examine autistic friendship networks in adulthood without focusing on support systems. We broadly recruited autistic and nonautistic college students (n = 592 students). Students provided ego-network information, including demographics, satisfaction with their social relationships, and perceptions of their friends (e.g. perceived neurotype and demographics). Autistic college students reported similar network structure metrics (e.g. density, number of friends) as nonautistic college students. Participants of all neurotypes reported similar friendship closeness, social satisfaction, and close mixed-neurotype friendships. However, on average, autistic college students had a higher proportion of autistic and neurodivergent friends than nonautistic college students. To a lesser extent, nonautistic students reported a greater preference for neurotypical friendships. These quantitative results suggest that autistic college students prefer building friendships with other autistic or neurodivergent peers (i.e. autistic or neurodivergent homophily). This preference suggests that authentically autistic (social) spaces and the availability of neurodivergent peers would be beneficial for autistic college students. Overall, these findings contradict the pathologizing "lack of interest in friendships" narrative surrounding autism.Lay AbstractPast research on autistic friendships has either recruited children or has focused on support services. We were interested in autistic adults who might not be enrolled in support services. So, we asked college students to fill out an interview form that included information about their traits and the traits of their friends. Based on answers to questions, participants were considered not autistic (n = 325), not sure but possibly autistic (n = 166), autistic-self-identified but not diagnosed (n = 67), or autistic-diagnosed (n = 34). Overall, all the groups reported similar degrees of social satisfaction (e.g. quality of friendships). On average, the two autistic groups had the most autistic friends and the fewest neurotypical friends. The "not-autistic" group had fewer autistic friends but the most neurotypical friends. However, all groups reported close "cross-neurotype" friendships (e.g. "not-autistic" people being close friends with autistic people). We concluded that autistic college students probably like friendships with other autistic or neurodivergent people, though college students build friendships with all neurotypes.These findings contradict stereotypes about autistic people preferring to be isolated. This study shows that autistic college students do build close friendships with others. It is important to note that we focused on autistic college students rather than all autistic people. We do not know if these findings apply to other groups of autistic individuals. We also relied on what people told us. It is possible that what people are reporting about their friends is not the same thing their friends would say.