IMVEST, an immersive multimodal virtual environment stress test for humans that adjusts challenge to individual's performance

Abstract

Laboratory stressors are essential tools to study the human stress response. However, despite considerable progress in the development of stress induction procedures in recent years, the field is still missing standardization and the methods employed frequently require considerable personnel resources. Virtual reality (VR) offers flexible solutions to these problems, but available VR stress-induction tests still contain important sources of variation that challenge data interpretation. One of the major drawbacks is that tasks based on motivated performance do not adapt to individual abilities. Here, we provide open access to, and present, a novel and standardized immersive multimodal virtual environment stress test (IMVEST) in which participants are simultaneously exposed to mental -arithmetic calculations- and environmental challenges, along with intense visual and auditory stimulation. It contains critical elements of stress elicitation – perceived threat to physical self, social-evaluative threat and negative feedback, uncontrollability and unpredictability – and adjusts mathematical challenge to individual's ongoing performance. It is accompanied by a control VR scenario offering a comparable but not stressful situation. We validate and characterize the stress response to IMVEST in one-hundred-and-eighteen participants. Both cortisol and a wide range of autonomic nervous system (ANS) markers – extracted from the electrocardiogram, electrodermal activity and respiration – are significantly affected. We also show that ANS features can be used to train a stress prediction machine learning model that strongly discriminates between stress and control conditions, and indicates which aspects of IMVEST affect specific ANS components.

Keywords

• Stress test
• Sympathetic nervous system
• Virtual reality
• Machine learning
• Multimodal stress