EEG, Empathy, and External Monitoring

Background:  Although it is generally accepted that empathy is impaired in autism spectrum disorders (ASD), there is not yet a complete understanding of how and when empathy becomes disrupted in the brain. Several models (e.g., Blair, 2008; Rankin et al., 2005) propose distinctions between “emotional” and “cognitive” components; however, the definition of cognitive vs. emotional aspects differs across studies. In the context of our ongoing studies of performance monitoring in ASD, we added an observational component to an explicit cognitive task.  

Objectives:  Using EEG, our aim was to investigate whether the neural signatures of children and adolescents diagnosed with an ASD differ from controls while watching a confederate research assistant perform a modification of the Eriksen Flanker task. Specifically, we hypothesized a diminished error-related negativity (ERN) in the ASD group.

Methods: Participants were 33 ASD and 28 age- and IQ-matched Typically Developing Controls (TDC) (total mean age = 13.24; Full Scale IQ = 108.22). Participants were instructed to watch their “teammate” do the Flanker task, who would be working for points to be awarded either to themselves or to the participant. Confederates deliberately made errors on about 33% of trials in order to ensure adequate numbers of error trials. Importantly, in order to make certain that participants were aware of error trials, we added a feedback display (“right” or “wrong”) after each trial. Each participant also completed the children’s version of the Interpersonal Reactivity Index.

Results:  Our major hypotheses were analyzed using a 2 (accuracy) x 2 (points type) x 2 (diagnostic group) repeated measures ANOVA. As expected, there was a significant main effect for accuracy, F = 14.48, p < .001, driven by a stronger ERP response to Error than to Correct trials in both groups. There was no significant interaction of points type, nor was there any interaction with diagnostic group. Surprisingly, there were no between-group differences on any scale of the children’s IRI.

Conclusions: We have recently reported impaired neural response in ASD on EEG tasks dependent on internal performance monitoring (South et al., 2010) but intact response when there is external feedback on the task (Larson et al., in press). In light of our added external feedback to the Flanker task, we interpret this study as further evidence for typical neural activity in ASD when the emotional valence is clearly delineated. We suggest that this cognitive component of empathy is not impaired, but that poor connectivity between social and emotional processing may lead to empathy deficits previously reported. However, in this high-functioning ASD sample there was no difference from controls on a standard, self-report empathy questionnaire, leaving further questions about the phenotype of everyday empathic behavior and attitudes in the absence of impaired verbal and cognitive ability.