A Repulsive Bias in Perceived Heading Away from the Straight-Ahead as Determined by Continuous Psychophysics

Abstract

Previous studies of heading perception have reported both attraction toward and repulsion away from straight ahead. Different potential explanations for these differences have been brought forward, such as the reference frame of the response (where egocentric responses are connected to attraction and allocentric responses are connected to repulsion) or the response range (where a wider response range is related to more repulsion). In this study, we aimed to contribute to this debate using a novel Continuous Psychophysics-based task in virtual reality to investigate systematic heading biases across the full 360{degrees} range (repulsion predicted) in an egocentric task (attraction predicted), while concurrently examining the perceptual and motor contributions to these biases. Participants (n = 60 across two experiments) continuously indicated their perceived heading using a joystick while experiencing optic flow consistent with slow or fast self-motion while the simulated heading direction was changing continuously. In a CONTROL condition, participants aligned the joystick with the direction of a clearly visible arrow. The joystick was placed on the dominant-hand side in Experiment 1 (on the right for n = 28 participants) and in the middle and on the left in Experiment 2. In both experiments, we found consistent repulsion away from straight ahead in the CONTROL, FAST, and SLOW conditions, regardless of joystick position. Biases around the other cardinal directions were less consistent and generally tended to show attraction. Around straight ahead, we also observed a general progression from smaller biases in CONTROL to larger biases in FAST and SLOW, suggesting that decreased reliability of perceptual cues increased biases. The broad pattern of findings remained reasonably consistent across the three joystick positions. Our original prediction of attraction in an egocentric task was therefore not supported. Instead, the findings were more consistent with the response-range account, which links larger response ranges (like our 360{degrees} response range) to stronger repulsion. Our findings further support the use of Continuous Psychophysics to study systematic perceptual biases and provide a blueprint for the analytical steps required to do so.