Automatic Processing of Changes in Facial Emotions in Dysphoria: A Magnetoencephalography Study.

Qianru Xu,Xueqiao Li,Elisa M Ruohonen,Wenbo Luo,Kairi Kreegipuu,Gabor Stefanics,Chaoxiong Ye,Piia Astikainen

doi:10.3389/fnhum.2018.00186

Abstract

It is not known to what extent the automatic encoding and change detection of peripherally presented facial emotion is altered in dysphoria. The negative bias in automatic face processing in particular has rarely been studied. We used magnetoencephalography (MEG) to record automatic brain responses to happy and sad faces in dysphoric (Beck’s Depression Inventory ≥ 13) and control participants. Stimuli were presented in a passive oddball condition, which allowed potential negative bias in dysphoria at different stages of face processing (M100, M170, and M300) and alterations of change detection (visual mismatch negativity, vMMN) to be investigated. The magnetic counterpart of the vMMN was elicited at all stages of face processing, indexing automatic deviance detection in facial emotions. The M170 amplitude was modulated by emotion, response amplitudes being larger for sad faces than happy faces. Group differences were found for the M300, and they were indexed by two different interaction effects. At the left occipital region of interest, the dysphoric group had larger amplitudes for sad than happy deviant faces, reflecting negative bias in deviance detection, which was not found in the control group. On the other hand, the dysphoric group showed no vMMN to changes in facial emotions, while the vMMN was observed in the control group at the right occipital region of interest. Our results indicate that there is a negative bias in automatic visual deviance detection, but also a general change detection deficit in dysphoria.

Highlights

Depression is a common and recurring disorder
The negative bias in the dysphoric group, which was demonstrated as a relative difference in M300 amplitude for sad and happy faces in comparison to the control group, seems to be associated with change detection, as the deviant stimulus responses, but not the standard stimulus responses, were larger for sad than happy faces in the dysphoric group
Our finding of the negative bias in emotional face processing extends from the previous findings involving attended stimulus conditions (Bistricky et al, 2014; Chen et al, 2014; Zhao et al, 2015; Dai et al, 2016) to ignore condition

Summary

Introduction

Depression is a common and recurring disorder. Decades ago, Beck (1976) suggested that negatively biased information processing plays a role in the development and maintenance of depression. The P2 component ( labeled as P250), a positive polarity ERP response approximately at 200–320 ms in the temporo-occipital region, is followed by the N170 and reflects the encoding of emotional information (Zhao and Li, 2006; Stefanics et al, 2012; Da Silva et al, 2016) It has a counterpart in MEG responses, sometimes labeled M220 (e.g., Itier et al, 2006; Schweinberger et al, 2007; Bayle and Taylor, 2010). In an ERP study with an emotional face intensity judgment task, depressed participants showed larger P2 amplitude for sad faces than happy and neutral ones, reflecting a negative bias in their attentive level, which was not found in the control group (Dai and Feng, 2012)

Objectives

Methods

Results

Conclusion