Influence of the Level of Immersion in Emotion Recognition Using Virtual Humans

Facial affect recognition is a skill that involves the ability to perceive and distinguish between different affective facial expressions. This skill is crucial in day-to-day human relationships, and it is negatively affected by various psychiatric disorders. This has led to the development of tests that attempt to measure or train this skill. Virtual Reality (VR) is a technology that has been used for this purpose, as it can immerse the user in a virtual world that can be more similar to the real world than a traditional desktop computer. Therefore, the objective of this paper is twofold. First, a potential improvement in emotion recognition using VR instead of a traditional desktop screen condition is studied. Secondly, the impact that the presentation angle of the virtual faces has on emotion recognition is investigated. To this end, the same application was adapted to be shown in a desktop screen and in a VR environment to 36 mentally healthy participants recruited for each condition. A set of dynamic virtual faces (DVFs) were used as stimuli to show the 6 basic emotions plus the neutral expression. The DVFs were also rotated to be presented to the participants in front and side views. The results show a slight improvement in the VR environment. In addition, the accuracy is better in the front views than in the side views for both conditions, which is consistent with previous studies.