Rapid trial-and-error learning with simulation supports flexible tool use and physical reasoning

Area of Science:

• Cognitive Science
• Robotics
• Human-Computer Interaction

Background:

• Animals and artificial agents show object manipulation skills.
• Humans uniquely exhibit flexible and creative tool use for problem-solving.

Purpose of the Study:

• To investigate the mechanisms behind human flexible and efficient physical problem solving.
• To introduce a novel game, Virtual Tools, for studying general physical problem-solving capabilities.

Main Methods:

• Development of the Virtual Tools game with challenging physical puzzles.
• Introduction of the "sample, simulate, update" (SSUP) model.
• Testing the SSUP model against human performance data across 30 game levels.

Main Results:

• Humans solve complex physical puzzles in the Virtual Tools game efficiently.
• The SSUP model accurately captures human problem-solving performance.
• The model demonstrates the interplay between imagined action effects and updated action priors.

Conclusions:

• Human physical problem-solving flexibility stems from imagining hypothetical action outcomes.
• Efficient search in physical problem-solving is driven by prior knowledge updated through observation.
• The SSUP model offers a framework for understanding how general physical knowledge is applied to specific tasks.