Predicting Gravitational Self-Motion: Falling Down versus Falling Up
Preprint
Abstract
Humans have been shown to use an internalized representation or prior of Earth gravity to predict object motion. Often, this is shown using the fact that humans tend to neglect an object’s acceleration when predicting it’s motion. This bias tends to be ameliorated when the acceleration acting on the object is Earth gravity. It is currently unclear whether this prior can also be used when assessing one’s self-motion. We immersed two cohorts of 20 seated participants in a virtual environment of a street scene. They experienced simulated self-motion consistent with either flying straight upwards towards the ceiling (ANTI-GRAVITY) or falling downwards towards the ground (GRAVITY). After 0.8 s to 1.4 s of motion, the screen turned blank, and participants pressed a button to indicate when they thought they reached the surface they were moving towards. In Experiment 1, they were instructed to use their eye level as the reference (i.e., when they imagined the floor or ceiling to be aligned with their eyes). In Experiment 2, they used their feet or the top of theirs heads to judge when they had reached the ceiling or floor, respectively. In Experiment 1, no differences between the GRAVITY and ANTI-GRAVITY conditions were found in terms of accuracy or precision, contrary to our hypotheses. In Experiment 2 participants pressed the button significantly later in the ANTI-GRAVITY condition than in the GRAVITY condition, in line with the predictions of the use of a gravity prior, while precision remained unaffected. Speculatively, the instructions in Experiment 1 may have made the experience less ecologically valid and immersive, thus preventing participants from activating the strong gravity prior. Experiment 2 (head/feet as reference), on the other hand, provides convincing evidence for an involvement of this prior in the prediction of self-motion. In sum, the study provides evidence that humans can rely on an internalized representation of Earth gravity when predicting self-motion, at least under ecologically valid task conditions. The absence of effects in Experiment 1 and their presence in Experiment 2 suggest that activation of this gravity prior depends on the framing of the task.Keywords
This publication has 43 references indexed in Scilit:
- A binary acceleration signal reduces overestimation in pedestrians’ visual time-to-collision estimation for accelerating vehiclesHeliyon, 2024
- Auditory Information Improves Time-to-collision Estimation for Accelerating VehiclesCurrent Psychology, 2022
- Eye movements and manual interception of ballistic trajectories: effects of law of motion perturbations and occlusionsExperimental Brain Research, 2014
- Random effects structure for confirmatory hypothesis testing: Keep it maximalJournal of Memory and Language, 2013
- Ocular pursuit and the estimation of time-to-contact with accelerating objects in prediction motion are controlled independently based on first-order estimatesExperimental Brain Research, 2009
- Extrapolation of vertical target motion through a brief visual occlusionExperimental Brain Research, 2009
- Visuo-motor coordination and internal models for object interceptionExperimental Brain Research, 2009
- Internal models and prediction of visual gravitational motionVision Research, 2008
- Gravity bias in the interpretation of biological motion by inexperienced chicksCurrent Biology, 2006
- Visual processing of optic accelerationVision Research, 1992