Brain Activation in Contrasts of Microexpression Following Emotional Contexts

Ming Zhang, Ke Zhao, Fangbing Qu, Kaiyun Li, Xiaolan FuFrontiers in Neuroscience 2020 · DOI

TL;DR

This study investigates how emotional context influences the brain's processing of microexpressions, which are fleeting facial expressions often used to conceal true emotions. Using fMRI, researchers found that while neutral contexts improve microexpression recognition, emotional contexts (both positive and negative) lead to poorer accuracy and increased brain activity in attention-related regions, suggesting a complex interplay between context and microexpression perception.

What Problem Does This Paper Address?

The recognition of facial expressions is known to be influenced by preceding emotional contexts. However, the specific impact of these contexts on the perception of microexpressions – extremely brief, involuntary facial movements that can betray underlying emotions – remains less understood. While behavioral studies have hinted at this influence, there's a need for neural evidence to explain how emotional contexts modulate the processing of these subtle expressions. This paper aims to bridge this gap by examining brain activation patterns when microexpressions are presented within different emotional contexts (negative, neutral, positive).

Methods

The researchers employed a functional Magnetic Resonance Imaging (fMRI) study to investigate brain activity. They adapted a synthesized microexpression task, based on the Micro-Expression Training Tool (METT), to present participants with target microexpressions (anger, neutral, happiness) embedded within different emotional contexts (negative, neutral, positive). Participants viewed these stimuli while in an fMRI scanner and were asked to identify the target microexpression. The fMRI data was analyzed using a standard two-level approach in SPM8, focusing on event-related analysis. Specifically, they compared brain responses to the same target microexpression presented after different emotional contexts. This contrast-based analysis aimed to isolate the neural effects of the contextual manipulation.

Datasets

The study utilized a dataset of 120 synthesized microexpression images, adapted from a previous study. These images featured 20 different models (10 male, 10 female) displaying anger, neutral, and happiness expressions. The stimuli were presented as short video clips. No new datasets were introduced; the focus was on analyzing behavioral and neural responses to existing stimuli under manipulated contextual conditions.

Key Insights and Ideas

  • Emotional Contexts Impair Microexpression Recognition: Contrary to what might be expected, emotional contexts (both positive and negative) actually decrease the accuracy of recognizing microexpressions compared to neutral contexts.
  • Neural Correlates of Contextual Interference: The fMRI data revealed increased brain activation in specific regions when microexpressions followed emotional contexts compared to neutral ones. This suggests that emotional contexts demand more cognitive resources, potentially diverting attention away from the subtle microexpression itself.
  • Attention-Related Brain Regions are Key: The increased activations were observed in areas associated with attention and emotional processing, such as the right intraparietal sulcus, extranuclear, precuneus, and inferior parietal lobule. This supports the idea that emotional contexts capture attentional resources.
  • Discrepancy in Processing: The observed differences in brain activation patterns for the same target microexpression under different contexts highlight a "discrepancy in the processing of emotional contexts," implying a complex interaction rather than a simple additive effect.

Key Results

Behavioral: * Accuracies for recognizing target microexpressions were significantly higher following neutral contexts compared to negative or positive contexts. * Response times were marginally longer following negative contexts compared to neutral contexts. * The accuracy for anger microexpressions was higher following neutral than positive contexts. * The accuracy for neutral microexpressions was higher following neutral than negative or positive contexts. * The accuracy for happiness microexpressions was higher following neutral than negative contexts.

fMRI: * Significant increases in BOLD signals were observed in contrasts of the same targets following different contexts. * For anger microexpressions, negative context > neutral context showed increased activation in the right intraparietal sulcus and extranuclear. Positive context > neutral context showed activation in the right precuneus and subgyral. * For neutral microexpressions, negative context > neutral context showed activation in the right inferior parietal lobule. Positive context > neutral context showed activation in the right precuneus and left dorsal posterior cingulate cortex.

Limitations and Future Work

The authors acknowledge a key limitation: the task design combined the forward context and target into a single "whole duration" for analysis, making it difficult to isolate the exact BOLD response to the microexpression itself. This means the findings reflect the combined effect of context and target, rather than a precise measure of microexpression processing in isolation.

Future work could focus on:

  • Exploring the processing of different functional areas' responses to microexpressions with more ecological validity.
  • Developing experimental designs that allow for a clearer separation of context and target processing.
  • Investigating the neural mechanisms underlying the effect of emotional context on microexpression recognition with more refined temporal resolution.

Summary updated

Abstract

The recognition of microexpressions may be influenced by emotional contexts. The microexpression is recognized poorly when it follows a negative context in contrast to a neutral context. Based on the behavioral evidence, we predicted that the effect of emotional contexts might be dependent on neural activities. Using the synthesized microexpressions task modified from the Micro-Expression Training Tool (METT), we performed an functional MRI (fMRI) study to compare brain response in contrasts of the same targets following different contexts. Behaviorally, we observed that the accuracies of target microexpressions following neutral contexts were significantly higher than those following negative or positive contexts. At the neural level, we found increased brain activations in contrasts of the same targets following different contexts, which reflected the discrepancy in the processing of emotional contexts. The increased activations implied that different emotional contexts might differently influence the processing of subsequent target microexpressions and further suggested interactions between the processing of emotional contexts and of microexpressions.
BibTeX
@inproceedings{zhang2020brain,
  author = {Ming Zhang and Ke Zhao and Fangbing Qu and Kaiyun Li and Xiaolan Fu},
  title = {{Brain Activation in Contrasts of Microexpression Following Emotional Contexts}},
  booktitle = {Frontiers in Neuroscience},
  year = {2020},
  doi = {10.3389/fnins.2020.00329},
}
Figure