Neural correlates of pupil dilation during human fear learning

• Published in: NeuroImage (Orlando, Fla.) Vol. 147; pp. 186 - 197
• Main Authors: Leuchs, Laura, Schneider, Max, Czisch, Michael, Spoormaker, Victor I.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 15.02.2017; Elsevier Limited
• Subjects: Anxiety; Arousal; Baas; Brain mapping; Cognitive ability; Conditioned response; Conductance; Data processing; Dorsal anterior cingulate cortex; Electromyography; Emotions; Expectancy; Extinction behavior; Eye; Fear conditioning; Fear extinction; fMRI; Functional magnetic resonance imaging; Habituation; Learning; Light; Neuroimaging; Pain; Physiology; Post traumatic stress disorder; Pupillometry; Reflexes; Salience network; Skin; Startle response; Unconditioned stimulus; Fear extinction; Dorsal anterior cingulate cortex; fMRI; Salience network; Pupillometry; Fear conditioning
• ISSN: 1053-8119; 1095-9572; 1095-9572
• DOI: 10.1016/j.neuroimage.2016.11.072

Fear conditioning and extinction are prevailing experimental and etiological models for normal and pathological anxiety. Pupil dilations in response to conditioned stimuli are increasingly used as a robust psychophysiological readout of fear learning, but their neural correlates remain unknown. We aimed at identifying the neural correlates of pupil responses to threat and safety cues during a fear learning task. Thirty-four healthy subjects underwent a fear conditioning and extinction paradigm with simultaneous functional magnetic resonance imaging (fMRI) and pupillometry. After a stringent preprocessing and artifact rejection procedure, trial-wise pupil responses to threat and safety cues were entered as parametric modulations to the fMRI general linear models. Trial-wise magnitude of pupil responses to both conditioned and safety stimuli correlated positively with activity in dorsal anterior cingulate cortex (dACC), thalamus, supramarginal gyrus and insula for the entire fear learning task, and with activity in the dACC during the fear conditioning phase in particular. Phasic pupil responses did not show habituation, but were negatively correlated with tonic baseline pupil diameter, which decreased during the task. Correcting phasic pupil responses for the tonic baseline pupil diameter revealed thalamic activity, which was also observed in an analysis employing a linear (declining) time modulation. Pupil dilations during fear conditioning and extinction provide useful readouts to track fear learning on a trial-by-trial level, particularly with simultaneous fMRI. Whereas phasic pupil responses reflect activity in brain regions involved in fear learning and threat appraisal, most prominently in dACC, tonic changes in pupil diameter may reflect changes in general arousal. •Pupil responses during fear learning are associated with dACC and thalamus activity.•Phasic pupil responses do not show substantial habituation during fear learning.•Phasic pupil responses are negatively correlated with tonic pupil size.•Pupillometry can track conditioned responses on a trial-by-trial level.