Brain synchronization during perception of facial emotional expressions with natural and unnatural dynamics

Dionysios Perdikis,Timothy R Brick,Kathrin Kaulard,Christian Wallraven,Jakob Volhard,Ulman Lindenberger,Viktor Müller

doi:10.1371/journal.pone.0181225

Abstract

Research on the perception of facial emotional expressions (FEEs) often uses static images that do not capture the dynamic character of social coordination in natural settings. Recent behavioral and neuroimaging studies suggest that dynamic FEEs (videos or morphs) enhance emotion perception. To identify mechanisms associated with the perception of FEEs with natural dynamics, the present EEG (Electroencephalography)study compared (i) ecologically valid stimuli of angry and happy FEEs with natural dynamics to (ii) FEEs with unnatural dynamics, and to (iii) static FEEs. FEEs with unnatural dynamics showed faces moving in a biologically possible but unpredictable and atypical manner, generally resulting in ambivalent emotional content. Participants were asked to explicitly recognize FEEs. Using whole power (WP) and phase synchrony (Phase Locking Index, PLI), we found that brain responses discriminated between natural and unnatural FEEs (both static and dynamic). Differences were primarily observed in the timing and brain topographies of delta and theta PLI and WP, and in alpha and beta WP. Our results support the view that biologically plausible, albeit atypical, FEEs are processed by the brain by different mechanisms than natural FEEs. We conclude that natural movement dynamics are essential for the perception of FEEs and the associated brain processes.

Highlights

Human social behavior can be viewed as an ongoing interaction among individuals through a flow of coordinated processes of action and perception such as speech, gestures and facial expressions, which serve to communicate our intentions, emotions, and experiences [1]
Facial emotional expressions and brain synchronization after 700 ms), theta phase locking, delta and lower theta power, and alpha desynchronization. These results show that static stimuli were processed more strongly at early latencies, whereas natural dynamic stimuli entailed more prolonged processing towards later latencies as expected. (Note that the delta and lower theta Phase Locking Index (PLI) response to the static -unlike the natural—facial emotional expressions (FEEs) is back to baseline after 700 ms–see Figure A panel (a) of the S1 File) This conclusion agrees with previous studies [22,28] and supports the hypothesis of [19] that prolonged rather than stronger activations could explain the stronger fMRI and PET responses to dynamic FEEs over their static counterparts [17,18,19,20]
The present study used measures of amplitude (WP) and phase (PLI) of local brain oscillations to compare FEEs that differ in the naturalness of the movement dynamics

Summary

Introduction

Human social behavior can be viewed as an ongoing interaction among individuals through a flow of coordinated processes of action and perception such as speech, gestures and facial expressions, which serve to communicate our intentions, emotions, and experiences [1]. Brain processes have evolved to facilitate social coordination [1,2,3,4].

Methods

Results

Discussion

Conclusion