The ability to anticipate others’ actions relies on predictive coding mechanisms that integrate evolvingkinematic cues with internal motor representations. In high-speed sports such as tennis, thesecomputations must occur within tens of milliseconds. Using high-density EEG and a temporal occlusionparadigm, we investigated the neural dynamics that support predictive action monitoring in 25 experttennis players by systematically manipulating the moment at which visual information about anopponent’s stroke became unavailable. Participants watched groundstroke videos occluded at six timepoints (−320 to +80 ms relative to ball–racket contact) and indicated the predicted landing location.Behaviorally, anticipation accuracy increased and response times decreased as occlusion approached orexceeded contact, confirming the decisive contribution of late kinematic information. Time-frequency analyses of EEG signals showed a temporally structured engagement of large-scalecortical networks. Early in the action sequence, theta (3–5 Hz) synchronization was significantlyenhanced in centro-parietal regions during observation, consistent with the recruitment of top-downpredictive and attentional control processes. As the action unfolded, alpha (8-12 Hz) and beta (15-25 Hz)desynchronization emerged across sensorimotor and posterior regions, indexing progressive activationof the action observation and motor resonance systems. Critically, immediately preceding video offset,theta synchronization peaked in the +80 ms condition—i.e., when contact cues were fully available—suggesting that theta oscillations encode the integration of high-informational kinematic evidence forprediction. During the post-video period, theta power further increased in frontal and sensorimotor sites duringresponse preparation, and theta amplitude in this early motor-initiation window negatively correlatedwith the inverse efficiency score, linking stronger theta engagement to superior anticipatoryperformance. Beta desynchronization in sensorimotor cortices also intensified as the motor response approached, consistent with the release of inhibitory tone and the preparation of task-specific motorplans.Together, these findings delineate a hierarchical temporal architecture through which the braintransforms observed movement into predictive motor representations: early theta-mediated evaluativeprocessing, mid-stage alpha/beta-mediated sensorimotor resonance, and late integrative theta burstssupporting decision formation and action selection. The results provide novel electrophysiologicalevidence for predictive coding mechanisms governing action anticipation in expert athletes, highlightingtheta oscillations as a key neural substrate for perceptual–motor prediction in fast, open-skillenvironments.
Cortical dynamics underlying temporal extraction of kinematic evidence and predictive coding during observation of imminent actions
Filippo Zappasodi
Primo
;Camilla Scaramuzza;Sergio Costa;Pierpaolo Croce;Marika Berchicci;Maurizio Bertollo
2026-01-01
Abstract
The ability to anticipate others’ actions relies on predictive coding mechanisms that integrate evolvingkinematic cues with internal motor representations. In high-speed sports such as tennis, thesecomputations must occur within tens of milliseconds. Using high-density EEG and a temporal occlusionparadigm, we investigated the neural dynamics that support predictive action monitoring in 25 experttennis players by systematically manipulating the moment at which visual information about anopponent’s stroke became unavailable. Participants watched groundstroke videos occluded at six timepoints (−320 to +80 ms relative to ball–racket contact) and indicated the predicted landing location.Behaviorally, anticipation accuracy increased and response times decreased as occlusion approached orexceeded contact, confirming the decisive contribution of late kinematic information. Time-frequency analyses of EEG signals showed a temporally structured engagement of large-scalecortical networks. Early in the action sequence, theta (3–5 Hz) synchronization was significantlyenhanced in centro-parietal regions during observation, consistent with the recruitment of top-downpredictive and attentional control processes. As the action unfolded, alpha (8-12 Hz) and beta (15-25 Hz)desynchronization emerged across sensorimotor and posterior regions, indexing progressive activationof the action observation and motor resonance systems. Critically, immediately preceding video offset,theta synchronization peaked in the +80 ms condition—i.e., when contact cues were fully available—suggesting that theta oscillations encode the integration of high-informational kinematic evidence forprediction. During the post-video period, theta power further increased in frontal and sensorimotor sites duringresponse preparation, and theta amplitude in this early motor-initiation window negatively correlatedwith the inverse efficiency score, linking stronger theta engagement to superior anticipatoryperformance. Beta desynchronization in sensorimotor cortices also intensified as the motor response approached, consistent with the release of inhibitory tone and the preparation of task-specific motorplans.Together, these findings delineate a hierarchical temporal architecture through which the braintransforms observed movement into predictive motor representations: early theta-mediated evaluativeprocessing, mid-stage alpha/beta-mediated sensorimotor resonance, and late integrative theta burstssupporting decision formation and action selection. The results provide novel electrophysiologicalevidence for predictive coding mechanisms governing action anticipation in expert athletes, highlightingtheta oscillations as a key neural substrate for perceptual–motor prediction in fast, open-skillenvironments.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


