Remote identification of psychophysiological parameters for a cognitive-emotional conflict

Abstract

The contradiction between the conscious and emotional assessment of situations often arises when there is a discrepancy between reality and ideal view. The purpose of this study is to identify a system of remotely diagnosed parameters (by video recording using Skype tools), which can objectively indicate the presence of a cognitive-emotional conflict during a dialogue. The current study has become particularly relevant in the COVID-19 environment, where face-to-face communication is limited, and all interactions must occur in a remote online format. In such an environment, the identification of formal signs indicating latent conflicts in online interactions comes to the fore. To stimulate an outburst and fix the cognitive-emotional conflict remotely, various schemes were used: interviews based on the self-attitude technique, discussions based on the transactional analysis model, free reasoning about situations that were selected by the age, and other parameters of participants, as well as watching video clips with an unexpected end. For automatic marking of people's emotional reactions in video data, a computer program was developed using neural network algorithms based on VGG16.Furthermore, the neural network for emotion recognition was pretrained based on open sources by the P. Ekman FACS model. Finally, video recordings were marked according to the expert assessment results regarding the presence of cognitive-emotional conflict. We conducted 27 interviews with the participation of both genders, including ten high school students (from 12 to 17 years old), a total of 338,075 frames were received. As a result, it was found that the two facial expression parameters identified by the program reliably distinguish the presence of a cognitive-emotional conflict. Furthermore, in all the studied schemes of telecommunication dialogue, the «happy» parameter was significantly more pronounced in a situation of conflict.