A robust automatic micro-expression recognition system would have broad applications in national safety, police interrogation, and clinical diagnosis. Developing such a system requires high quality databases with sufficient training samples which are currently not available. We reviewed the previously developed micro-expression databases and built an improved one (CASME II), with higher temporal resolution (200 fps) and spatial resolution (about 280×340 pixels on facial area). We elicited participants' facial expressions in a well-controlled laboratory environment and proper illumination (such as removing light flickering). Among nearly 3000 facial movements, 247 micro-expressions were selected for the database with action units (AUs) and emotions labeled. For baseline evaluation, LBP-TOP and SVM were employed respectively for feature extraction and classifier with the leave-one-subject-out cross-validation method. The best performance is 63.41% for 5-class classification.

This study examines trust in technology‐supported groups from the perspectives of culture, social presence and group composition. Our results demonstrate that, in culturally homogeneous groups, individualism has a negative impact on the level of interpersonal trust; however, in culturally heterogeneous groups consisting of Chinese and U.S. participants, individualism has a positive impact on interpersonal trust among members. There were also significant differences in the level of trust between homogeneous Chinese groups and heterogeneous groups consisting of Chinese and U.S. participants. In addition, the mediating role of communication quality was identified to explain the effect for trust‐of social presence on trust. These findings have important implications on building and communication in global technology‐supported decision‐making groups.

Micro-expression spotting is a fundamental step in the micro-expression analysis. This paper proposes a novel network based convolutional neural network (CNN) for spotting multi-scale spontaneous micro-expression intervals in long videos. We named the network as Micro-Expression Spotting Network (MESNet). It is composed of three modules. The first module is a 2+1D Spatiotemporal Convolutional Network, which uses 2D convolution to extract spatial features and 1D convolution to extract temporal features. The second module is a Clip Proposal Network, which gives some proposed micro-expression clips. The last module is a Classification Regression Network, which classifies the proposed clips to micro-expression or not, and further regresses their temporal boundaries. We also propose a novel evaluation metric for spotting micro-expression. Extensive experiments have been conducted on the two long video datasets: CAS(ME) ² and SAMM, and the leave-one-subject-out cross-validation is used to evaluate the spotting performance. Results show that the proposed MESNet effectively enhances the F1-score metric. And comparative results show the proposed MESNet has achieved a good performance, which outperforms other state-of-the-art methods, especially in the SAMM dataset.

The purpose of this study was to investigate the effects of the duration of expressions on the recognition of microexpressions, which are closely related to deception. In two experiments, participants were briefly (from 20 to 300 ms) shown one of six basic expressions and then were asked to identify the expression. The results showed that the participants’ performance in recognition of microexpressions increased with the duration of the expressions, reaching a turning point at 200 ms before levelling off. The results also indicated that practice could improve the participants’ performance. The results of this study suggest that the proper upper limit of the duration of microexpressions might be around 1/5 of a second and confirmed that the ability to recognize microexpressions can be enhanced with practice.

Facial expression recognition is mediated by a distributed neural system in humans that involves multiple, bilateral regions. There are six basic facial expressions that may be recognized in humans (fear, sadness, surprise, happiness, anger, and disgust); however, fearful faces and surprised faces are easily confused in rapid presentation. The functional organization of the facial expression recognition system embodies a distinction between these two emotions, which is investigated in the present study. A core system that includes the right parahippocampal gyrus (BA 30), fusiform gyrus, and amygdala mediates the visual recognition of fear and surprise. We found that fearful faces evoked greater activity in the left precuneus, middle temporal gyrus (MTG), middle frontal gyrus, and right lingual gyrus, whereas surprised faces were associated with greater activity in the right postcentral gyrus and left posterior insula. These findings indicate the importance of common and separate mechanisms of the neural activation that underlies the recognition of fearful and surprised faces.