Bayesian Framework for Modeling Intuitive Dynamics in Collision Perception

A dataset likely containing human judgment data on object collisions, used to develop a Bayesian framework for modeling intuitive dynamics. The data was collected by researchers from University College London, MIT, and UC Berkeley for tasks involving mass and causality judgments. The dataset's size and specific structure are unknown.

Use Cases

• Train models to predict human causality judgments based on observed collision trajectories.
• Benchmark Bayesian inference models against human heuristics for mass ratio judgments.
• Analyze patterns of errors in human perception of physical variables like mass and force.

Strengths

• Framework integrates two common judgment tasks (mass and causality) into a single Bayesian model.
• Data is based on established experimental paradigms from prior studies in perception.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

paperswithcode, with authors from Gatsby Computational Neuroscience Unit (UCL), MIT, and UC Berkeley.

Collection Method

Data was likely collected from human subjects performing mass judgment and causality judgment tasks.

Time Range

null

Freshness

null

Geography

null