The influence of arousal on visuocortical processing in humans
OA Version
Citation
Abstract
Vision plays a pivotal role in our everyday lives, shaping our perception and consequently influencing our behavior and actions. However, vision is far from static. There are a number of processes that dynamically influence how we see our environment from moment to moment, with one process being states of arousal. For example, driving on a freeway, our vision is likely quite different when we are drowsy compared to when we are alert. While animal studies have found arousal states to substantially modulate visual responses, our understanding of its effects on human vision remains limited. To shed light on arousal’s influence on human visual processing, this work comprised a series of pupillometry and functional magnetic resonance imaging (fMRI) experiments aimed to untangle the influence of cognitive and emotional arousal on contrast processing in human early visual cortex (V1-V3). The goal of the first two experiments was to explore the influence of cognitive arousal – arousal induced by cognitive factors such as effort, load and task difficulty – on visuocortical contrast response functions (CRFs). Experiment 1 (n = 24) was a pupillometry study with dual objectives: first, to assess the reliability of a cognitive arousal manipulation for subsequent use in the ensuing fMRI study, and second, to determine the optimal luminance range for monitoring pupillary changes associated with cognitive arousal. This groundwork set the stage for integrating pupillometry into the subsequent fMRI study to monitor changes in arousal states. To achieve this, the experiment assessed the reliability of a cognitive arousal manipulation using auditory arithmetic problems of varied difficulties (hard, easy) while investigating its interaction with luminance on pupil size. Consistent with previous studies, pupil size was reliably modulated by math difficulty, evoking larger pupils for hard difficulty problems. Furthermore, the results revealed that there are indeed luminance regimes more ideal in observing pupillary differences by cognitive arousal—with high luminance being the least ideal. In Experiment 2 (n = 20), fMRI was employed to investigate the influence of cognitive arousal on CRFs in V1-V3. BOLD responses were measured while observers viewed stimuli that parametrically varied in contrast. Concurrently, different cognitive arousal states were induced with observers solving auditory arithmetic problems categorized as easy (low cognitive arousal) or hard (high cognitive arousal). The findings revealed a surprisingly diverse pattern of modulatory effects among individuals: some individuals displayed enhanced gain of neural response with increased cognitive arousal, while others exhibited the opposite effect—a decrease in gain with increased cognitive arousal. Furthermore, individuals' BOLD modulation patterns correlated with arousal-induced changes in their pupil size. The subsequent two experiments explored the impact of emotional arousal on pupil responses and CRFs in early visual cortex. Experiment 3 was a pupillometry study focused on assessing how emotional arousal interact with luminance in modulating pupil size. In Experiment 3A, observers (n = 36) listened to and rated sound clips from the International Affective Digitized Sounds (IADS) of neutral (low emotional arousal) and negative (high emotional arousal) valence, while the background luminance of the screen cycled through a range of luminance levels. In Experiment 3B (n = 36) and 3C (n = 33), observers viewed and rated grayscale and colored images from the International Affective Pictures System (IAPS) of neutral and negative valence. The results indicated that emotional arousal interacted with luminance in modulating pupil size differently from cognitive arousal: the effects were smaller and occurred solely at much lower luminances. Critically, there was a sizable group of individuals who did not display any response to the IAPS and IADS stimuli, across all luminances. Then, Experiment 4 (n = 12) investigated how emotional arousal, using negative and neutral auditory IADS sound clips, modulated CRFs. Overall at the group-level, emotional arousal had minimal effect on CRFs across the visual cortex. However, individual-level analysis revealed a subset of observers displaying enhanced neural gain with arousal, which contrasted with the majority showing little-to-no effect. Further analysis examining pupil size-based CRFs and CRFs in the lateral geniculate nucleus (LGN) uncovered consistent and robust modulatory effects among observers. Notably, the modulatory effects varied between different CRF delineations (valence versus pupil size) and among distinct visual areas (LGN versus V1-V3). These results collectively shed light on how arousal, with its diverse forms and individual differences, selectively modulates contrast processing.
Description
2024