A virtual reality study investigating the train illusion

Abstract

The feeling of self-movement that occurs in the absence of physical motion is often referred to as vection, which is commonly exemplified using the train illusion analogy (TIA). Limited research exists on whether the TIA accurately exemplifies the experience of vection in virtual environments (VEs). Few studies complemented their vection research with participants' qualitative feedback or by recording physiological responses, and most studies used stimuli that contextually differed from the TIA. We investigated whether vection is experienced differently in a VE replicating the TIA compared to a VE depicting optic flow by recording subjective and physiological responses. Additionally, we explored participants' experience through an open question survey. We expected the TIA environment to induce enhanced vection compared to the optic flow environment. Twenty-nine participants were visually and audibly immersed in VEs that either depicted optic flow or replicated the TIA. Results showed optic flow elicited more compelling vection than the TIA environment and no consistent physiological correlates to vection were identified. The post-experiment survey revealed discrepancies between participants' quantitative and qualitative feedback. Although the dynamic content may outweigh the ecological relevance of the stimuli, it was concluded that more qualitative research is needed to understand participants' vection experience in VEs.

Keywords: mixed methods; physiological responses; qualitative survey; vection intensity.

Figure 1.

The experimental set-up used in this study. Note: a participant is shown who is seated in the NLR Motion Platform, wearing the Varjo V3 head-mounted display and the Equivital galvanic skin response (GSR) sensor while using a HOTAS joystick to indicate their vection experience. The Equivital EQ02 is worn underneath the clothing.

Figure 2.

The virtual environments used in this study. Note: left: an optic flow condition which was used as an abstract motion stimulus. During this stimulus, a white point cloud moved towards the observer. Right: the train condition which aimed to contextually replicate the train illusion. During this stimulus a moving train could be seen through the window on participants' right-hand side. The windows on the left were blinded.

Figure 3.

Mean, standard deviation and compatibility interval of vection measures. Note: means (and standard deviations in parentheses) are denoted numerically for each condition in the appropriate bar. Individual data points are marked with ‘x'. Error bars represent the compatibility intervals.

Figure 4.

Distribution of participants' MISC ratings.