Change in Subjects' Emotional States as Observed through Facial Expressions during a Seven-Day Dry Immersion Experiment

Background. The problem of precise and reliable evaluation of the emotional state of operators in spaceflight conditions still remains unresolved, and therefore relevant. This evaluation is especially important for predicting operators` professional activity effectiveness considering the current emotional state which significantly determines the accuracy and rapidity of professional tasks performance. Objectives. Our study is carried out to empirically test the new method and technology for evaluating changes in the emotional state of the 7-day experiment participants measuring facial expression changes in their DPC-video interview recordings. Study Participants. The study involved 10 men aged 25 to 35 years. Volunteers were selected by the Institute of Biomedical Problems of the Russian Academy of Sciences as participants in the research. Methods. DPC video interviews were conducted daily in the morning and in the evening. Facial expression changes evaluation was conducted with the EmoRadar software which detects FACS action units (AUs), seven basic emotions and other patterns of facial activity. Results. The description of facial expression changes in the experiment are given based on the data from two participants. These changes were compared with the emotional experience of the participants and experimental conditions. The necessity of differentiated evaluation of the positive emotion of happiness is shown since its assessments may include not only happiness as a basic emotion, but also two expressions of smiles (social smile or coy smile) similar to it. The effectiveness of the software used and the author's approach to computer analysis of facial activity is clearly shown. Conclusions. With the help of the developed technology based on the implementation of a comprehensive approach to measurement and interpretation of facial expressions, it is possible to provide objective information about characteristic changes in facial expressions as reliable behavioral indicators of a person's emotional state. The data are independent of respondents' self-assessments or expert’s experience. This opens up new methodological possibilities for aerospace psychology to solve the problem of automatic monitoring of the emotional state of operators.