This study is one of the first to examine how adolescents with ASD may differ from typically developing adolescents in their neural responsivity to peer rejection. Specifically, the findings described here indicate that adolescents with ASD may differentially process peer rejection experiences at the neural level, despite evidence of similar self-reported distress responses. In the following discussion, we interpret these findings and speculate about their meaning in order to help direct future investigations of peer-related processing among individuals with ASD.
In terms of subjective responses to peer exclusion, we found that adolescents with ASD and controls did not differ in their self-reported feelings of distress following the exclusion round of Cyberball. These findings are consistent with our hypothesis, and previous research indicating, that adolescents with ASD feel just as rejected as their typically developing counterparts when they are excluded by peers (Sebastian et al., 2009
Despite these similar behavioral findings, neuroimaging analyses indicated that adolescents with ASD showed reduced neural engagement compared to typically developing adolescents during peer rejection. Specifically, ROI analyses (as well as exploratory whole-brain analyses performed at a lowered threshold) revealed less activity during exclusion versus inclusion among adolescents with ASD compared to controls in regions previously shown to be positively related to adolescents’ distress during exclusion (i.e., subACC, AI), as well as in regions previously shown to be negatively related to distress during exclusion (i.e., VLPFC, VS). This is consistent with many prior studies indicating that individuals with ASD display hypoactivation in brain regions involved in emotion processing—including the subACC and AI specifically—when performing a range of social cognitive tasks (see Di Martino et al., 2009
for a review). Furthermore, these findings are consistent with those reported by Sebastian and colleagues (2009)
indicating that individuals with ASD may show less variation in mood as a result of peer rejection. Of course, neural activity in affective-processing regions is only one potential proxy of mood, and more direct assessments of mood (such as those collected by Sebastian and colleagues) may have impacted our reported findings; however, the heightened neural activity in subACC and AI in our control group is consistent with the notion that ASDs were less affectively impacted by the rejection. Finally, although the role of oxytocin was not the primary focus of this investigation, group differences were found in several limbic areas similar to those previously linked with oxytocin production, including portions of the insula and cingulate, as well as the amygdala (see Supplementary Table 1
; Ferguson et al., 2002
; Huber et al., 2005
; Landgraf and Neumann, 2004
; Lim and Young, 2006
). Thus, our findings are consistent with those found by Andari and colleagues suggesting that neural activity related to oxytocin production may relate to the differential social processing observed among ASDs and controls.
This differential engagement of neural circuitry in response to peer rejection could also be related to a number of qualitative factors related to peer rejection experiences that adolescents with ASD have in their daily lives. For example, adolescents with ASD may be more habituated to the experience of being rejected, since it happens to them more frequently. As a result, their neural responses might be dampened, given the familiarity of the experience. A second possibility is that adolescents with ASD may actually expect to be rejected when they interact with novel groups of peers. Researchers have proposed that one important component of typical neural responses to social rejection is the detection and recognition that a social expectancy (i.e., being included) is being violated (Eisenberger and Lieberman, 2004
; Somerville et al., 2006
; Bolling et al., 2010
). Thus, the reduced neural responses to rejection observed in adolescents with ASD may reflect the fact that they have learned through experience that they are not likely to be included, even if they experience subjective distress upon later reflection. Finally, related to this possibility, adolescents with ASD might respond to social rule-breaking in unique ways. Given that they often display difficulties following social rules and initiating positive peer interactions (Attwood, 2000
; Hauck et al., 1995
; Sigman and Ruskin, 1999
), they may be responding in ways different than typically developing adolescents when they witness similar indiscretions being committed by others.
It is important to note that in the current study, we did not replicate previously found associations between self-reported distress and brain activity during peer exclusion within our typically developing group. Previous findings in our lab have indicated that adolescents display greater activity in the subACC and AI to the extent that they are more distressed by exclusion and greater activity in the VLPFC and VS to the extent that they are less distressed by exclusion (Masten et al., 2009a
); we propose two possibilities for why we did not find similar patterns here. First, the current sample was almost all male, where as the previously tested sample comprised slightly more females than males. Thus, one possibility is that girls and boys show differential associations between their neurobiological and subjective responses to peer-related social stimuli—a possibility that is not surprising given known gender differences in peer-related processes at this age (e.g., Guyer et al., 2009
; Rose and Rudolph, 2006
). Second, the current sample spanned a wide adolescent age range, whereas the previous study included only 12–13 year olds. Thus, it is possible that the correspondence between self-reported and neural affective responses to peer rejection is particularly pronounced in the years immediately following the transition to middle school (i.e., around age 13)—when peer rejection becomes particularly prevalent (e.g., Nansel et al., 2001
) and the social influence of peers increases (e.g., Masten et al., 2009b
). Of course, many other variables may have also contributed to the inconsistency between the current findings and those described in Masten et al. (2009a)
. Thus, future studies will be useful in exploring the influence of both gender and age, as well as additional factors such as pubertal status, brain development and social environment, that may impact these kinds of brain-behavior correlations.
With regard to our ASD sample, the correlational findings between self-reported distress and brain activity during peer rejection were somewhat surprising. Consistent with our previous work examining typically developing adolescents, we found a positive correlation between subACC and distress, but we also found a positive correlation between VLPFC and distress, which ran counter to previous findings in typically developing adolescents. Given that the VLPFC is typically thought to play a regulatory role in the context of social rejection (e.g., Eisenberger et al., 2003
; Masten et al., 2009a
), one possibility is that this positive correlation reflects an ineffective attempt to regulate distress resulting from peer rejection among the ASD group. Future examination of these brain-behavior correlations and regulatory strategies specifically among ASD populations will shed additional light on this possibility.
As a whole, this study provides an important first step toward understanding how adolescents with ASD may differentially perceive and respond to peer rejection—an event that is especially salient during adolescence generally, but also particularly frequent among this population. Moreover, given that adolescents with ASD and typically developing adolescents displayed differential neural responses despite reporting similar subjective ratings, these findings highlight the importance of using neuroimaging techniques to examine processes of peer functioning in ASD. Our hope is that future investigations into the neural processes involved in social deficits in ASD will lead to better understanding of the mechanisms underlying the social atypicalities that characterize this population, and eventually inform intervention and therapy practices.