Predictive gaze shifts elicited during observed and performed actions in 10-month-old infants and adults

We asked whether people's actions are understood by projecting them onto one's own action programs, according to the direct matching hypothesis, and whether this mode of control functions in infants. Adults’ and infants’ gaze and hand movements were measured in two live situations. The task was either to move an object between two places in the visual field, or to observe the corresponding action performed by another person. When performing the action, infants and adults behaved strikingly similar. Hand and gaze movements were simultaneously initiated and gaze arrived at the goal ahead of the hand. When observing the actions, the initiation of the gaze shift was delayed relative to the observed hand movement in both infants and adults, but it still arrived at the goal ahead of the hand. For both the performance and observation of actions the proactiveness of gaze shifts was associated with saccades ahead of the velocity peak of the hand. The close similarity between adults’ and infants’ actions when performing the movements and the great advantage of the adults when observing them support the conclusion that one's own motor actions develop ahead of the ability to predict other people's actions.

• Infants will grasp an object to move it to a position at 10 months of age.
• For performing and observing such actions, gaze arrived at the goal ahead of the hand.
• The matching observed-performed was higher in adults than in infants.
• An observed goal-directed action is understood from one's own mastered similar action.