Purchase this article with an account.
Mohan Ji, Jacqueline Fulvio, Bas Rokers; The combination of 3D motion cues in Virtual Reality. Journal of Vision 2018;18(10):498. https://doi.org/10.1167/18.10.498.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
In the natural environment, many visual cues signal object motion. In the laboratory, the contribution of these cues to perception is often studied in isolation. Here we exploit virtual reality (VR) technology to study the rich interaction between perspective, disparity, and parallax cues to object motion in a large sample of participants. Naïve observers (N = 91) viewed 3D random dot motion stimuli in a VR head-mounted display (Oculus Rift). We measured psychophysical thresholds for the discrimination of approaching and receding motion by varying motion coherence of dot stimuli that contained perspective, disparity, combined (perspective and disparity), or full VR (combined + motion parallax) cues to 3D motion. Ten observers did not have a detectable threshold in any of the conditions and were removed from further analysis. Observers were sensitive to both the perspective and disparity cues to 3D motion. However, sensitivity to perspective cues generally exceeded disparity sensitivity. Critically, we did not find a significant correlation between perspective and disparity sensitivity. A model assuming the two independent cues are combined optimally predicted combined cue performance well, F(1,72) = 7.60, p < .01. Finally, we did not find evidence for greater sensitivity to the Full VR condition in our experiments. In summary, our results indicate the mechanisms for processing the perspective and disparity cues are independent. Furthermore, sensitivity to both cues combined can be characterized based on sensitivity to either cue in isolation. Finally, we did not find evidence for superior sensitivity when motion parallax cues were available, suggesting that the naïve participants in our experiments could not take advantage of the head-contingent updating feature of VR displays. Our results inform neural models of 3D motion processing under naturalistic conditions.
Meeting abstract presented at VSS 2018
This PDF is available to Subscribers Only