Utilizing Neural Gain as a Model for Explaining Features of Autism Spectrum Disorders: The effects of constitutive locus coeruleus activity on attention-based learning.

Abstract

The goal of this thesis is to investigate the interaction between breadth of attention and trial and error learning in individuals with autism spectrum disorders. Previous literature suggests that autistic individuals attend to and consequently learn from environmental stimuli in an atypical manner, although these studies have principally relied on measuring attentional focus using coordinate measurements from eye tracking and have generated conflicting results without elucidating a unified mechanism to explain autistic individuals’ attention and learning difficulties. Here we suggest a model for understanding some features of learning in autistic individuals that is based on neural gain. Specifically, previous work has shown that neurotypical individuals with high neural gain (a state associated with increased norepinephrine release, which can be assessed non-invasively using pupillometry) attend to and learn more readily from those stimulus dimensions to which they are predisposed to. In contrast, when in a state of low gain (a condition associated with down-regulated norepinephrine release), neurotypical individuals can integrate information and learn about multiple stimulus dimensions. On the basis of literature suggesting that autistic individuals’ significantly elevated baseline pupil sizes might be a consequence of increased norepinephrine production, we hypothesized that autistic individuals are constitutively in a state of high gain. We tested this hypothesis by comparing the performance and pupillary responses of autistic and neurotypical teenagers and young adults engaged in trial and error learning in a multidimensional environment. Contrary to our hypothesis, autistic individuals learned equally (but not efficiently) about multiple stimulus dimensions (and notably from both social and nonsocial stimuli), although their pupillary responses indeed suggested higher levels of norepinephrine. We therefore consider how the chronic release of norepinephrine that is typically implicated in autistic individuals might ultimately induce a significant decrease in gain, thereby providing a possible neural-based explanation for autistic individuals’ deficiencies at efficiently learning from environmental cues.