Action Predictions Facilitate Embodied Geometric Reasoning.

Area of Science:

• Cognitive Science
• Educational Psychology
• Neuroscience

Background:

• Task-relevant actions can enhance mathematical thinking, including complex areas like mathematical proof.
• The action-cognition transduction (ACT) model proposes a bidirectional link between motor actions and cognitive reasoning processes.

Purpose of the Study:

• To investigate if cognitive benefits observed for performing actions extend to predicting actions.
• To test the ACT model's hypothesis that action predictions influence cognition even without physical execution.

Main Methods:

• A two-by-two factorial design was employed with 127 undergraduates.
• Participants either generated task-relevant action predictions or performed task-relevant directed actions.
• Geometry proof performance was measured as the primary outcome.

Main Results:

• Generating action predictions significantly improved participants' geometry proof production.
• Combining action predictions with directed actions did not yield additional benefits compared to action predictions alone.
• Gestural replays during explanations correlated with improved insight and proof performance for both predicted and directed actions.

Conclusions:

• Predicting task-relevant actions enhances mathematical cognition, likely through simulated transformations on imagined mathematical objects.
• The findings support the ACT model by showing overlapping cognitive processes for action prediction and performance.
• Educational interventions prompting students to imagine or perform body movements related to mathematical tasks may facilitate reasoning.