Gravity and Known Size Calibrate Visual Information to Time Parabolic Trajectories

Area of Science:

• Cognitive Neuroscience
• Motor Control
• Visual Perception

Background:

• Catching objects in parabolic trajectories requires precise time-to-contact (TTC) estimation.
• Previous models often bypass TTC estimation, relying on complex visual transformations.
• Estimating TTC from 2D retinal data to 3D space is computationally intensive.

Purpose of the Study:

• To describe information sources for parabolic trajectory interception.
• To demonstrate how prior knowledge aids TTC estimation using a Bayesian framework.
• To assess the robustness of these predictions in dynamic environments.

Main Methods:

• Simulations exploring the role of prior distributions (gravity, physical size).
• Bayesian encoding-decoding framework to integrate prior knowledge with visual data.
• Analysis of predictive capacities from minimal early visual information.

Main Results:

• Exploiting prior knowledge of gravity and object size simplifies TTC estimation.
• Predictive capabilities are enhanced with minimal early visual input.
• Prior information disambiguates visual cues for more accurate predictions.

Conclusions:

• Prior knowledge significantly simplifies the visual processing required for interceptive actions.
• This approach enables robust TTC predictions, even in complex dynamic settings.
• Future research should further investigate prior knowledge calibration for action control.