Sustained Pupil Dilation to Sad Faces Is Associated with Self-Reported Rumination in Adults with Autism Spectrum Disorder and Adults with Current Depression

Background: Rumination, or repetitive negative thinking, is associated with the onset and maintenance of depressive disorders in the typically developing (TD) population. Autism spectrum disorder (ASD) is marked by repetitive thinking and behaviors as one of the core symptom domains that characterize this disorder. Rates of depression are also notably high in adolescents and adults with ASD. Our long-term goal is to explore repetitive thinking as a pathway to depressed mood in ASD. However, we first must lay groundwork in understanding the measurement of and mechanisms associated with depressive rumination in this special population. Pupillometry offers a neural measure of sustainedcognitive-affective processing, which may converge with repetitive thinking.

Objectives:  The current work aims to advance understanding of how people with ASD, a population marked by repetitive thinking and high rates of depression, compare to typically developing depressed (DEP) and never-depressed controls (TDC) on measures of repetitive negative thinking and sustained neural response to social-emotional material. We assessed group differences among these three diagnostic cohorts on both self-reported rumination and on pupil indices of reflexive cognitive-affective response over time, and assessed whether rumination is related to affective response in general, and within ASD specifically.

Methods:  A total of n=53 adults aged 18-35 with verbal IQ>80 were recruited from three cohorts: ASD (n=21); TD with current depressive disorders (DEP, n=13); and TD comparison participants with no depression or anxiety history (TDC, n=19). All participants completed diagnostic screening and testing (including the Autism Diagnostic Observation Schedule [ADOS-2] and the Structured Clinical Interview for DSM-5 Disorders [SCID-5]), self-reported on depressive rumination with the Ruminative Response Scale (RRS), and provided pupillary response data within a passive-viewing task in which emotionally-valenced faces were displayed for 400 ms, followed by an 8 second mask.

Results:  Participants with ASD were significantly higher than TDC and significantly lower than DEP on depressive rumination (RRS). For pupil analyses, patterns of significant cohort differences preserved across stimulus condition (Happy, Sad, Angry, Neutral) suggest a more dynamic trajectory in ASD pupil response, in which ASD resembles TDC at second 2 and grows in magnitude of pupil dilation over the trial duration to more resemble DEP by second 8 -- whereas the two TD cohorts show an immediate response (either low or high) and sustain that level on average throughout trial duration (Figure 1). Individuals who ruminate more (here, they tended to belong to the ASD or DEP cohorts) generally had greater pupil dilation to Sad faces across 8 second trials (Figure 2). When TDC and DEP were collapsed into one TD group with greater variability on rumination, the interaction was also significant for Happy and marginally significant for Angry conditions.

Conclusions:  These data suggest that both ASD and TD individuals who endorse ruminating also are likely to exhibit sustained neural activity following emotionally-salient information. This work helps to locate ASD on a continuum of rumination scores, and to evaluate the convergence between self-reported rumination and sustained neural response to emotional stimuli within and across these diagnostic cohorts.