Newtonian Predictions Are Integrated With Sensory Information in 3D Motion Perception

Area of Science:

• Cognitive Neuroscience
• Visual Perception
• Physics Perception

Background:

• Newtonian mechanics governs everyday object motion.
• The brain may internalize physical laws for motion perception.
• Previous research showed the visual system fills information consistent with kinematic constraints.

Purpose of the Study:

• Investigate if internalized physical regularities influence motion perception beyond filling in missing data.
• Determine if the brain's internal physics model overrides conflicting sensory information.
• Test the hypothesis that perception integrates sensory data with internal predictions.

Main Methods:

• Adult participants viewed 3D billiard-ball collisions with varying Newtonian consistency.
• Perceptual judgments of postcollision trajectories were collected.
• Data analyzed using a maximum-likelihood sensory integration model.

Main Results:

• Perceptual judgments of trajectories were biased towards the Newtonian outcome.
• This bias occurred even when motion information was inconsistent with physical laws.
• Results supported a model where perception averages sensory input and internal predictions.

Conclusions:

• The brain's perception of motion relies on internalized physical regularities, including Newtonian mechanics.
• Internal predictions consistent with physical laws influence perception, especially when sensory data is ambiguous or conflicting.
• Sensory integration models can explain how the brain reconciles sensory input with internal physical knowledge.