Assessing the effectiveness of virtual reality tasks as stress-inducing environments

For safety reasons, the impact of stress on cognitive readiness cannot be assessed, in real-time hazardous working scenarios, by several psychophysiological methods. This study aimed to assess the validity of virtual reality environments in eliciting a significant level of stress, allowing the study of behaviours that might mimic those typically expressed by workers in an oil platform. Humans have a complex brain structure that enables learning, task performance, and responsiveness to stimuli. Cognitive readiness refers to the state in which alertness and mental preparation reach the necessary level for an adequate performance, especially in complex and unpredictable environments. However, external factors, such as pressure and stress, may impair cognitive readiness, posing a long-term challenge in several activities, particularly in offshore oil platform workers, who frequently operate in adverse and risky conditions. Therefore, developing innovative technologies to monitor cognitive readiness using accessible, simple, and quickly executed tasks could reduce work-related accidents in these and other hazardous scenarios. However, for safety reasons, psychophysiological measures, such as electroencephalographic signals and electrodermal potentials, which could provide valuable information in developing such technologies, cannot be collected during real-time activities. To partially overcome this experimental barrier, we assessed the effectiveness of a virtual reality (VR) simulation wherein a stress-inducing task was akin, in essential sensorimotor attributes, to complex procedures typically carried out in oil platforms. Materials and methods Adult volunteers were subjected to two tasks: a bomb deactivation simulation in a VR environment (submitting the volunteers to the Keep Talking and Nobody Explodes game under noisy instructions) followed by the presentation on a computer screen of a sequence of either positive or negative images selected from the International Affective Picture System (IAPS). Electrodermal potentials (galvanic skin response, GSR), heart rate (HR), and electroencephalographic activity (EEG) were acquired during the behavioural procedures. The subjective mood questionnaire State- Trait Anxiety Inventory - Scale (STAI-S) was applied before the tasks to evaluate overall anxiety. The STAI-T and the Visual Analogue Mood Scale (VAMS) were also applied after each task to assess feelings of tension, anxiety, and nervousness. Results In this preliminary study, since the main objective was the validation of the stress-inducing VR environment, we concentrated on analysing the impact of the bomb deactivating procedure and the presentation of both kinds of affectively arousing images on the mood state expressed by the volunteers. For the data obtained from the State-Trait Anxiety Inventory scale, a 2X3 two-way mixed design ANOVA, with the between-subjects factor Group (positive vs. negative images) and the within-subjects factor Time (pre-test, post-game, and post-images) revealed a highly significant main effect of the factor Time (F[84,2] = 20.33, p < 0.0001, partial eta-sqr = 0.326). A marginal interaction between the two factors (F[84,2] = 2.548, p = 0.084, partial eta-sqr = 0.057) is in agreement with a post-hoc analysis (Holm’s test) that revealed no statistically significant differences between the three levels of the factor Time for the positive image group. In contrast, a statistically significant difference (p < 0.001) was found between the pre-test average score and the two following average scores (post-game and post-images levels), which did not differ statistically from one another (p > 0.99). For the VAMS data, a 2X2 mixed design ANOVA showed similar results, with a marginal interaction between the two factors (F[41,1] = 2.843, p = 0.099, partial eta- sqr = 0.065), which agrees with a qualitatively (but non-statistically significant) decrease in the anxiety average score only for the positive image group. Conclusion Our findings have shown that the VR-simulated task was able to elicit a substantial level of anxiety that persisted throughout the presentation of the sequence of affective images only in the negative- image group, tending to decline toward normal levels in the positive-image group. Therefore, this setup can be used as a tool to investigate, by means of other psychophysiological methods, a behaviour that mimics real working scenarios.