Behavioral study demonstrates that aligned self-action predictions drive dynamic coordination in paired adults, suggesting shared internal models outweigh physiological synchrony.
Key Points
To determine the relative contributions of predictive processing, physiological synchrony, motor kinematics, and social factors to dynamic real-world joint motor coordination.
Assessed 70 participants (64 unique pairs) during solo and joint sessions of a naturalistic, turn-taking ball-hitting task.
Collected multimodal behavioral, physiological, demographic, and social closeness metrics alongside Bayesian generative modeling of predictive processing.
Coordination success was most strongly associated with anticipation of one's own action outcomes, outperforming physiological synchrony, motor behavior, demographic traits, and social closeness.
Partners whose self-action outcome predictions were closely aligned coordinated significantly better by consistently fulfilling mutual behavioral expectations.
Bayesian generative modeling revealed that well-coordinated pairs relied more heavily on prior expectations than on incoming sensory evidence during joint action.