Neural Correlates of the Pupillary Light Reflex in the Broader Autism Phenotype

Background: Autism spectrum disorder (ASD) is characterized by early emerging deficits in social communication and restricted/repetitive behavior. In addition to these core symptoms, individuals with ASD often exhibit atypical arousal regulation. As the locus coerulus (LC) has numerous projections throughout the central nervous system, with norepinephrine (NE) acting as the main neurotransmitter, the LC-NE system has been linked to arousal and attention. The LC-NE system is thought to be indexed by the pupillary light reflex (PLR) because the LC releases NE to control pupillary dilation. Adolescents with ASD exhibit a hyposensitive PLR (Fan et al., 2009); however, infants at high risk for ASD display a hypersensitive PLR (Nystrom et al., 2015); thus, the developmental trajectory of the PLR remains unclear. Finally, while the PLR itself is well characterized, associated neural response and relations to social function are poorly understood.

Objectives: To investigate relationships among PLR, electrophysiological brain correlates, and subthreshold autistic traits.

Methods: Pupil data were collected on an Eyelink-1000 system, and simultaneous EEG recording was collected with a 128-channel Hydrocel Geodesic Sensor net. Stimuli consisted of a white central fixation on a black background, which flashed white for approximately 75 ms. Participants viewed 50 consecutive trials of PLR. EEG data were decomposed and power was computed for the delta (2-4 Hz), theta (4-8 Hz), and alpha (8-13 Hz) frequency bands. Four regions of interest were created by averaging across channels: left frontal, right frontal, left posterior, and right posterior. Characteristics of the PLR were extracted from averaged trials. Autistic traits were measured with a self-report measure, the Broad Autism Phenotype Questionnaire (BAP-Q).

Results: Preliminary analyses of 10 subjects (data collection is ongoing) revealed that subjects with higher scores on the BAP-Q rigidity scale exhibited faster PLR (r -.653, p=0.041). Higher scores on the BAPQ Pragmatic score significantly correlated with slower constriction velocity (r=-.665, p=0.036). Greater pupil constriction was associated with increased power in the right frontal delta band (r=.909, p=.001). Decreased theta power in the right posterior was associated with increased impairment on the BAP-Q pragmatic language scale (r=-.747, p=.021), as well as decreased pupil constriction (r=.776, p=.014).

Conclusions: In the current study, autistic traits related to rigidity were associated with hypersensitive PLR; in contrast, autistic traits related to pragmatic language difficulties displayed a hyposensitive PLR. These distinctions suggest that previous discrepant results may be due to developmental effects, with rigidities being more active in infancy and pragmatic language becoming more focal in subsequent development. Moreover, variance in autistic traits and PLR were associated with oscillatory brain activity. These results shed light on the relationships between subcortical regions supporting arousal, cortical EEG, and social functioning, providing insight into mechanisms relevant to the development of targeted interventions and personalized medicine.