Harnessing the power of social networks to support every lives of young adults with autism

Andrew is a 19-year-old college student with Asperger’s syndrome. He lives with his mother and younger brother. Although he is moderately independent, managing his life under his mother’s proactive guidance, he struggles with building a robust and sufficiently large network of people who can provide advice about everyday situations.

Figure 1. GroupMe allowed users to create his own focused communication circle, add members to the circle, and broadcast queries to the circle. Andrew, one of our participants, asked a question about his upcoming job interview preparation. More than one person (average: 2.5 people) replied a query within 10.3 minutes.
GroupMe was not only an empowering tool for the participants but also a convenient tool for researchers because it was freely available and offered cross-platform support for desktop, smartphone, and feature phone users.

We investigate how a social network site helps young adults with autism garner social support from members of networks online beyond their immediate family and friends. We asked three adolescents and young adults with Asperger’s syndrome to use a particular cross-platform social network site, GroupMe, for four weeks to create their own support networks and request help or advice from the network members. We explore a specific feature of GroupMe that allows users to define a small set of members, a “circle,” with whom they will participate in shared discussions.

We examine how the use of circles influences how an individual with autism might reach out to people beyond a primary caregiver for advice on everyday life skills. In addition, we analyze the types of queries asked and the patterns of communication among members of circles. Finally, we determine how conversation over GroupMe impacts the relationship between an individual and circle members.

The results of the study led to several interesting findings:

  • Limited changes to the members of the circles: All participants (three individuals with autism) created a single circle and kept the circle closed with 5-7 members.
  • Shared responsibility for initiating and responding: GroupMe encouraged individuals with autism to initiate communication. It also redistributed responsibility for responding across members within a circle.
  • The most frequently discussed topic was an ongoing issue that all of the participants with autism faced—that of socializing. Participants did not ask questions about some sensitive topics such as personal hygiene management.
  • GroupMe led to an increased sense of closeness to members whom participants did not know well prior to the study.
  • Offline socialization that occurred with GroupMe interaction was clearly positive experiences both for the young adults, and for their caregivers who want their children to seek social opportunities.
  • A shared discussion thread within a circle itself may serve as a tool for individuals with autism to learn social skills by allowing them to passively observe how people initiate a discussion topic and respond to others.

The analysis of the GroupMe field deployment study showed that the circles of communication helped participants overcome their over-reliance on their primary caregivers by increasing social closeness to others. However, engaging in conversation in a unified circle with a small set of members has a limitation. Drawing on this finding, we discuss design opportunities for building features that facilitate the formation of circles based on the topics a user want to receive advice, make circle membership transparent, and control the fine-grained communication along with the network member’s availability, tailored to support independent living of people affected by their cognitive disabilities.

We expand our research scope to explore developing a robust network of people whom the user is not likely to know but who nonetheless may be willing to commit small amounts of time to support people with special needs. We wish to investigate the potential for and the challenges of harnessing crowd-sourced volunteers that provide daily support or social advice to individuals with autism.

For more, see our full paper, Investigating the Use of Circles in Social Networks to Support Independence of Individuals with Autism.

Hwajung Hong, Georgia Tech
Lana Yarosh, AT&T
Jennifer Kim, University of Illinois at Urbana-Champaign
Gregory Abowd, Georgia Tech
Rosa Arriaga, Georgia Tech

About the author

Hwajung Hong

PhD student in Georgia Tech School of Interactive Computing HCC program with interests in HCI and Social Computing

View all posts


  • Interesting work! Was this research inspired by seeing these communities form organically? In other words, did you find that autism-spectrum individuals were already using social networks like GroupMe to seek one another out?

    • Great questions, Anand.

      We were very inspired by our previous observation of offline social networks that participants have organically formed. The young adults with autism heavily relied upon their primary caregiver and occasionally interacted with others such as their siblings, relatives, and family’s friends. We examined if existing social network sites can leverage this natural social network and if the ability to direct conversations to a set of people online could break the trend of over-reliance on the primary caregivers for the individuals with autism. At the outset of our field investigation, we asked our participants to create a list of existing social networks who would be able to help and invite them to join the communication circle of GroupMe.

      All participants in our study reported that they had social network site accounts, mostly Facebook though they were very inactive. Furthermore, none of our participants adopted Facebook for this particular purpose; connecting to someone who could provide help or advice and requesting advice when they need help. Compared to Facebook, GroupMe was more convenient tool to form a direct communication circle with existing social networks offline and to get feedback almost synchronous. This was the main reason why we picked GroupMe for the investigation.

  • Very interesting. Did you see start to see some patterns in the circle formation/did they parallel what you expected from offline support networks?

    • This is very interesting question, Manya.

      Regarding the circle formation pattern, we expected that participants might create multiple circles based upon either a common social connection offline (e.g., family, friends, co-workers) or a particular topic (e.g., health, job coaching). In contrast, the results showed that our participants formed a unified circle and kept the circle membership small and limited. This pattern might be caused by that 1) the existing social networks were not large enough to be classified into each circle 2) individuals with autism favored talking to the whole members to maximize the possibility to get rapid and diverse feedback. However, the shared discussion among circle members brought about positive impact in that 1) it improved understanding of autism-spectrum individual’s needs and desires that could shape future interaction, 2) it helped members be aware of whether the individual’s request had been handled by others or whether the poster was still awaiting a response. We will discover the effective mechanism and strategy to facilitate the formation of circle that reflects offline social network structure and practice.

      Typically individuals with autism often start communicating by a primary caregiver’s prompt. We found that GroupMe influenced them to initiate conversation by themselves or at least other members shared the responsibility of initiation while lessening the primary caregiver’s participation. Another fascinating finding was about the occurrence of new and richer social interactions for the participants. Two of the participants went on offline social outings with members in their network, and attributed the event to the social network. These social outings were clearly positive experiences for the individuals with autism, as well as exciting for their caregivers who wanted their children to be less isolated from peers.