Bullying Involvement and Social Information Processing of Dynamic Social Scenes in Children and Adolescents with ASD

Background:  

Several studies report that individuals with ASD are at an increased risk of bullying victimization in comparison to the general population (e.g. Cappadocia et al., 2011), yet few studies have used Crick and Dodge’s (1994) Social Information Processing (SIP) model to explore some of the underlying social processing difficulties in ASD. Two studies have found differences between ASD and children with typical development in the encoding and response generation stages of SIP, but not in the intent attribution stage (Embregts et al., 2006; Meyers et al., 2006).

Objectives:  

To develop a better understanding of bullying involvement and the associated encoding, attribution, and response generation stages of SIP in individuals with ASD compared with typically developing (TD) peers.

Methods:  

Participants to date are 22 children with ASD (6-16 years; 50% Asperger) and an age- and IQ-matched TD group of 20. Additional data collection is ongoing. Frequency of bullying victimization within the past month was determined using the PREVNet parent report survey. For the Social Information Processing Application (SIP-AP, innovation Research & Training, 2011), participants watched a series of eight brief videos depicting social situations involving either hostile or ambiguous provocations. The Tobii Eyetracker recorded where participants were looking during the videos. To assess encoding, participants were asked to describe what happened in the video and errors were counted. For intent attribution, participants rated the degree to which the provocateur intended to be mean on a 5-point Likert scale. To assess response generation, participants were asked what they could do if the provocation happened to them.

Results:  

Bullying involvement was significantly associated with sample χ² (1) = 8.64, p = .003.  As expected, parents of children with ASD reported more frequent physical, verbal, and social victimization (30 - 45%) than the TD comparison group (5-15%). Consistent with previous studies, the ASD group was significantly more likely than the TD group to make errors during the encoding stage of processing χ² (1) = 4.25, p = .04. Over two-thirds of children with ASD, relative to only one-third of the TD group, made encoding errors and both groups made more errors during hostile videos. Eyetracking data are currently being analysed. It is hypothesized that the ASD group will spend a smaller proportion of time attending to faces than their TD counterparts. Consistent with previous literature, independent-sample t-tests revealed no significant differences in intent attributions between the ASD and TD groups (all ps > .35). Preliminary analyses of the data reveal a greater mean number of responses generated by the TD group relative to the ASD group. The proportion of assertive responses was greater in the TD group, and the proportion of aggressive responses was greater in the ASD group.

Conclusions:  

This is the first study to examine both bullying involvement and social information processing in children with ASD and the first to incorporate an eye-tracking component with an ASD sample. Results will contribute greatly to the understanding of social difficulties in ASD and may help guide treatment planning by targeting specific processing issues.