Pupil response to negative emotional information in individuals at risk for depression

Abstract

Using pupil dilation as a physiological gauge of cognitive and emotional load, currently depressed individuals have previously shown sustained processing of negative emotional information. In this study, pupil dilation data from 24 recovered-depressed individuals were compared to those of 25 never-depressed individuals during a task in which they labelled the valence of emotional words before and after a negative mood-state induction. Before the mood induction, recovered-depressed participants evidenced more sustained pupil dilation in response to negative stimuli than did never-depressed participants. However, after the mood induction, sustained dilation to negative stimuli decreased for recovered participants in comparison to never-depressed participants. These findings suggest that, for people at risk for depression, small amounts of negative information are highly salient and yield increased physiological responses. However, larger amounts of negative information lead to decreased reactions potentially associated with emotional and cognitive blunting.

Keywords:

• Depression
• Depression vulnerability
• Information processing
• Emotion
• Pupil dilation

Acknowledgements

The authors thank members of the Cognitive Clinical Research Lab at the University of Kansas for assistance in study design and data collection. In addition, we are grateful to Doug Denney, Steve Ilardi, Jutta Joormann and two anonymous reviewers for comments on earlier versions of this manuscript.

Notes

¹Participants endorsing qualifying DSM-IV-TR criteria for a past Major Depressive Episode who reported the symptoms in response to Bereavement were including in the sample. The rationale for inclusion was that the processes of interest (information and emotional processes), are not theorised to be different in Bereavement than in Major Depressive Disorder.

This unpublished scale is available from the author at PMB#2657, 8033 Sunset Blvd., Los Angeles, CA 90046, USA. E-mail: [email protected].

³Prior to data collection, all interviewers administering the SCID underwent a training procedure that included coding mock interviews and viewing a SCID instructional video.

⁴Five 2×2 analyses of variance (ANOVAs) were conducted. This approach was used rather than a 5 × 2 ANOVA because we were interested in all the possible comparisons. In addition, no significant differences were expected for the comparisons (with the exception of an increase in sad mood from the pre- to post-mood ratings for both groups, which was already established with the MAACL-R ratings). Therefore, inflating type I error rate was not a concern. In fact, a Bonferroni correction was not used because when predicting no significant differences, a higher alpha value would serve as a more stringent test.

⁵A peak dilation analysis in which the averages of the single largest amount of pupil dilation were compared between groups was not significant.