Driving behavior in immersive Virtual Environments: a study on the role of cognitive workload, sense of presence and simulation sickness in the CAVE laboratory.

Investigating human factors in the adoption of safe driving behaviors is a key factor for the reduction of road accidents. Indeed, investigating driving behavior from a simulative perspective is not trivial, since it involves several potential human-related confounders that may blur the description of the phenomena. By investigating drivers’ reactions to pedestrian crossings in predicted and unpredicted situations, the present contribution specifically focuses on the moderating role of three human factors when dealing with simulative driving in urban environments: cognitive workload, sense of presence, and simulation sickness. To this aim, a driving simulation study has been carried out in an immersive CAVE setting (Cave Automatic Virtual Environment). Forty-five participants drove in a virtual urban scenario interacting with pedestrians at zebra crosswalks or jaywalkers appearing in different sections and following an oddball-like paradigm. The correlation between cognitive workload, simulation sickness, and sense of presence has been assessed, as well as their relationship with general characteristics of the sample and their effect on drivers’ performance. Results highlight a positive correlation between workload and simulation sickness, consistently with a role on average speed and SDLP (Standard Deviation of Lateral Position) throughout the driving session. The present research provides new insights on the role of fundamental human factors when driving in an immersive virtual environment dealing with high-risk stimuli. Importantly, here we delve into the suitability and the potential of a stimulating experimental setting such as the CAVE environment, aiming to provide practical information on the design and the implementation of immersive multitool experimental studies in the field of road safety.