Using pupil size and heart rate to infer affective states during behavioral neurophysiology and neuropsychology experiments

Abstract

Background Nonhuman primates (NHPs) are a valuable research model because of their behavioral, physiological and neuroanatomical similarities to humans. In the absence of language, autonomic activity can provide crucial information about cognitive and affective states during single-unit recording, inactivation and lesion studies. Methods standardized for use in humans are not easily adapted to NHPs and detailed guidance has been lacking. New Method We provide guidance for monitoring heart rate and pupil size in the behavioral neurophysiology setting by addressing the methodological issues, pitfalls and solutions for NHP studies. The methods are based on comparative physiology to establish a rationale for each solution. We include examples from both electrophysiological and lesion studies. Results Single-unit recording, pupil responses and heart rate changes represent a range of decreasing temporal resolution, a characteristic that impacts experimental design and analysis. We demonstrate the unexpected result that autonomic measures acquired before and after amygdala lesions are comparable despite disruption of normal autonomic function. Comparison with Existing Methods Species and study design differences can render standard techniques used in human studies inappropriate for NHP studies. We show how to manage data from small groups typical of NHP studies, data from the short behavioral trials typical of neurophysiological studies, issues associated with longitudinal studies, and differences in anatomy and physiology. Conclusions Autonomic measurement to infer cognitive and affective states in NHP is neither off-the-shelf nor onerous. Familiarity with the issues and solutions will broaden the use of autonomic signals in NHP single unit and lesion studies.

Autonomic arousal tracks outcome salience not valence in monkeys making social decisions

“…Across two laboratories, with different monkeys, different versions of a social reward allocation task, and different stimuli, we found that monkeys’ pupils were paradoxically narrower in anticipation of the preferred prosocial outcome ( other trials) relative to the less preferred antisocial outcome ( neither trials). This is contrary to what is usually observed in studies that manipulate reward magnitude, in which pupil size continually increases as outcomes become more preferred 10 . In this task, vicarious reward does not correspond with increasing pupil diameter.…”

“…Accordingly, they choose reward to the self most often because it is most arousing, reward to other moderately often because it is moderately arousing, and reward to neither least often because it is least arousing. This would be consistent with how monkeys’ pupil size, a common indicator of autonomic arousal, behaves during nonsocial tasks: pupil dilation reliably increases with the amount of juice predicted by a stimulus 10 . Neurally, it would be consistent with the population average activity of ACCg neurons; these neurons are most active for rewards to the self , moderately active for rewards to the other , and least active for rewards to neither 7 .…”

Shared neural codes for eye-gaze and valence

“…Fig.2), in agreement with the stressful and defensive responses that are traditionally induced by direct-gaze of a human intruder 8 . Heart-rate and heart-rate-variability (HRV) further confirmed anxietyrelated responses 42 (Fig.1j,k, t-test, p<0.05).…”

A Review of Pupillometry for Human-computer Interaction Studies

“…All parameters of the PLN are tied to stimulus intensity. There is a logarithmic relation between stimulus intensity and constriction amplitude, constriction time, constriction speed and the inverse of latency time [19]. Light adapted pupil size can be estimated before running any participants.…”

“…Emotional changes are another important influence on pupil reactivity [19,22]. According to the latest studies, the valence of the emotion (pleasant or unpleasant stimulus) is not important only the intensity is [23].…”