Making Sense of Sensory Sensitivity: New Evidence and a New Framework

Background: Atypical sensory sensitivity in autism is a puzzle. Its definition and measurement is essentially subjective, and it is unknown how it maps on to behavior: is it linked to perceptual abilities (e.g. local processing) or dysfunction? Similarly, it is unclear how an individual can be both hyper-sensitive and hypo-sensitive. Does it change over time, depend on context, or do hyper- and hypo- refer to different things (e.g. high sensitivity v. low responsivity)? Finally, although increased sensory sensitivity is assumed to reflect increased neurophysiological responses to sensory stimuli we don’t know which neural circuits are crucial, including the relative importance of sensory and affective regions.

Objectives: I will present a new framework for exploring these issues which, whilst not directly answering these questions, provides a roadmap for how they can be tackled by future research. Specifically, I will discuss how different facets of sensory sensitivity (subjective, neural, behavioral) can be understood within a signal detection framework and how existing theories fit within this (including ideas from other panel speakers). I shall present unpublished data from psychophysical studies and brain stimulation that explore the idea that individual differences in neural noise contributes to adaptation and hyper-excitability.

Methods: Adaptation (measured in terms of subjective discomfort, or subjective neutrality) to visual stimuli is measured in several paradigms. Variability in TMS phosphene thresholds from visual cortex stimulation is considered (with motor thresholds as a control). These objective measures are regressed again subjective sensory sensitivity and autistic trait measures (with the AQ).

Results: Findings show a relationship between high sensory sensitivity and low adaptation (aversive visual patterns become less aversive through repetition in most people, but this is diminished in people with high sensory sensitivity), and a negative relationship between phosphene thresholds and sensory sensitivity (low phosphine thresholds linked to high sensory sensitivity).

Conclusions: Different accounts of sensory sensitivity are not mutually exclusive and future research needs to bridge the gap between implementational theories (e.g. excitation:inhibition ratio) and those couched in cognitive/computational terms (e.g. predictive coding).