Error Monitoring Predicts Executive and Social Functioning in Autism Spectrum Disorder (ASD)

Background: The error-related negativity (ERN) is an event-related potential component (ERP) indexed by a negative deflection occurring 0 to 100 ms following an erroneous response. The ability to monitor and respond to errors in childhood is important for the development of regulated executive functioning skills (Weinberg et al. 2015). Literature investigating ERN in ASD report smaller amplitudes, indicating reduced sensitivity to errors (South et al. 2010). However, more brain-behavior evidence is needed to fully understand the role of error monitoring in specific domains in and outside of executive functioning in children with ASD.

Objectives: The goals of the following analyses were to: (1) examine the relation between ERN and performance on a behavioral battery of executive function measures, and (2) examine how ERN relates to social functioning in school-aged children with ASD.

Methods: 40 (4 female) children with ASD between the ages of 7-11 years provided ERN data during a Go/No-go task. All children had cognitive ability in the average range with the Full Scale Intelligence Quotient (FSIQ) ranging from 80-135, M= 107.8 (SD= 12.8). Dependent variables included ERN latency and mean amplitude at Fz. Response-locked epochs were 100 ms long and baseline corrected 150 to 50 ms prior to response on incorrect trials for participants with a minimum of 10 errors (Olvet and Hajcak 2009). The behavioral battery included measures of interference suppression (i.e., Stroop task), behavioral inhibition and cognitive flexibility (i.e., Change Task), and working memory (i.e., Digit Span task). Social functioning was quantified using scores on the Autism Diagnostic Observation Schedule (ADOS-2) Social Affect domain.

Results: For the Stroop, children who had less negative ERNs (i.e., they exhibited dampened neural responses to their own errors) were more disrupted by the incongruent condition (i.e., less able to respond efficiently in the face of interfering information), r(39)=0.431, p=.006 . This relation persisted when age was controlled, r(36)=0.425, p=.008. No relations were detected between ERN and the ability to inhibit dominant responses or shift between response types on the Change Task. With respect to working memory, children who had longer ERN latencies (i.e., slower to detect their own errors) had larger discrepancies between their forward (i.e., short term) and backward (i.e., working) digit span scores, r(39)=0.368, p=.021, and this effect persisted when age was controlled, r(36)=0.403, p=.012. In relation to social functioning, smaller ERN amplitude related to higher social symptoms on the ADOS-2 Social Affect domain, r(37)=0.322.p=.043, and remained when corrected for age, r(37)=0.325, p=.043.

Conclusions: The above findings imply that the regulated ability to monitor errors predicts performance in domains of executive function and social behavior. Executive and social functioning challenges in children diagnosed with ASD may in part arise from atypical neural responses to self-monitoring. Future analyses will include: (1) data from correct-response negativity (CRN) to examine conflict detection on correct trials and (2) additional executive and social measures to capture abilities across multiple contexts.