Abstract
Virtual reality is a powerful tool in human behaviour research. However, few studies compare its capacity to evoke the same emotional responses as in real scenarios. This study investigates psycho-physiological patterns evoked during the free exploration of an art museum and the museum virtualized through a 3D immersive virtual environment (IVE). An exploratory study involving 60 participants was performed, recording electroencephalographic and electrocardiographic signals using wearable devices. The real vs. virtual psychological comparison was performed using self-assessment emotional response tests, whereas the physiological comparison was performed through Support Vector Machine algorithms, endowed with an effective feature selection procedure for a set of state-of-the-art metrics quantifying cardiovascular and brain linear and nonlinear dynamics. We included an initial calibration phase, using standardized 2D and 360° emotional stimuli, to increase the accuracy of the model. The self-assessments of the physical and virtual museum support the use of IVEs in emotion research. The 2-class (high/low) system accuracy was 71.52% and 77.08% along the arousal and valence dimension, respectively, in the physical museum, and 75.00% and 71.08% in the virtual museum. The previously presented 360° stimuli contributed to increasing the accuracy in the virtual museum. Also, the real vs. virtual classifier accuracy was 95.27%, using only EEG mean phase coherency features, which demonstrates the high involvement of brain synchronization in emotional virtual reality processes. These findings provide an important contribution at a methodological level and to scientific knowledge, which will effectively guide future emotion elicitation and recognition systems using virtual reality.
Highlights
The automatic quantification and recognition of human emotions is a research area known as "Affective Computing", which combines knowledge in the fields of psychophysiology, computer science, biomedical engineering and artificial intelligence [1]
No subjects showed depressive symptoms according to their Patient Health Questionnaire (PHQ)-9 scores
With this aim in mind, we developed a realistic 3D immersive virtual environment (IVE) simulation of an art museum and performed a comparative study involving 60 subjects in a real art museum and its simulation, while they were performing a free exploration of an exhibition
Summary
The automatic quantification and recognition of human emotions is a research area known as "Affective Computing", which combines knowledge in the fields of psychophysiology, computer science, biomedical engineering and artificial intelligence [1]. Irrespective of the application, two approaches have commonly been proposed to model emotions: discrete and dimensional models. The former proposes that there is a small set of basic emotions, assuming that complex emotions result from a combination of these basics, including anger, disgust, fear, joy, sadness and surprise [6]. Dimensional models propose a multidimensional space where each dimension represents a fundamental property common to all emotions. The “Circumplex Model of Affect” (CMA) is one of the most used model, and refers to a Cartesian system of axes with two dimensions [8]: valence, i.e. the pleasantness or unpleasantness of an emotion; arousal, i.e. the intensity of the emotion in terms of activation from low to high
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
