The Establishment of Weak Ecological Microexpressions Recognition Test (WEMERT): An Extension on EMERT

The JACBART (Japanese and Caucasian Brief Affect Recognition Test) microexpression recognition test only examines facial expressions under the neutral expression background and the ecological validity is not high. The EMERT (Ecological MicroExpressions Recognition Test) microexpression recognition test examined six microexpressions under seven backgrounds but does not detect the intensity of expressions. In the current study, a weak ecological microexpression recognition test was established to examine the recognition features of six weak microexpressions in all seven high intensity basic expressions. The results found: (1) the test had good retest reliability, criterion validity and ecological validity; and (2) the reliability and validity tests revealed a lot of characteristics of weak microexpression recognition. There were training effects in some weak microexpression recognition. Weak microexpression recognition was generally positively related to the microexpression recognition of JACBART but were generally negatively related to approximate common expressions. The backgrounds main effects in all weak microexpressions were significant and pairwise comparisons show there were a wide range of differences between weak microexpressions under different backgrounds. The standard deviations, of the accuracy of weak microexpressions in different backgrounds, were used to define the fluctuations of the weak microexpression recognition and we found that weak microexpression recognition had many fluctuations. (3) Personality openness and its subdimensions (O1, O2, O3, and O5) were generally positively related to some weak microexpression recognition, except O1, which was significantly negatively related to surprise under neutrality. O1 was positively related to the standard deviation of the weak anger microexpression recognition accuracies and O6 was negatively related to the standard deviation of the weak happiness microexpression recognition accuracies in the first measurement.