Ecological validity of VR experiments for psychological and physiological responses to audio-visual environments

Abstract

Over the past two decades, increasing research has used physiological measurements to assess the audio-visual environment. VR (virtual reality) -based laboratory experiments have been frequently employed to recreate complex audio-visual environments and collect psychological and physiological data. However, there is ongoing debate about whether VR experiments can yield generalizable results applicable to real-world scenarios. To address this concern, the present research investigated the ecological validity—defined as the extent to which laboratory data reflect real-world perceptions—of VR experiments in audio-visual environment research. A 2 × 3 within-subject design experiment was conducted, with two sites and three experiment conditions, including in-situ, room-scale VR environment, and Head-mounted display (HMD). The results indicated that although HMDs were perceived as more immersive than cylindrical VR environments, both types of VR setups were ecologically valid regarding audio-visual perceptive parameters. However, for psychological restoration metrics, neither VR tool could perfectly replicate the in-situ experiment, with cylindrical VR being slightly more accurate than HMDs. Regarding physiological parameters, both HMDs and cylindrical VR showed potential for representing real-world conditions in terms of EEG change metrics or asymmetry features. Nonetheless, HMDs were not valid substitutes for real-world settings concerning EEG time-domain features, whereas cylindrical VR was more accurate on this metric.