Fig 1.
Still frames of the stimulus videos taken from the action-effect condition (upper row) and the no-action-effect condition (middle row) of Experiment 1 and 2 as well as from the claw-action-effect condition (lower row) of Experiment 3.
The squares in the left pictures of the no-action-effect condition and the claw-action-effect condition represent examples of the AOIs used for data analysis (the squares were not visible during the experiment).
Fig 2.
Mean gaze-arrival times for the 6-, 11-, and 7-month-olds in the human hand action-effect condition and no-action-effect condition of Experiments 1 and 2, and for the 7-month-olds in the claw-action-effect condition of Experiment 3.
Positive (vs. negative) values represent gaze arrival at the goal object before (vs. after) the hand/claw arrives there. Error bars represent standard-errors, and the asterisks mark predictive gaze with gaze-arrival times significantly above 0 ms, *** = p < .001.
Fig 3.
Mean gaze-arrival times across trials 2–12 for the 6-, 11-, and 7-month-olds in the human hand action-effect condition and the no-action-effect condition of Experiments 1 and 2, and for the 7-month-olds in the claw-action-effect condition of Experiment 3.
Positive (vs. negative) values represent gaze arrival at the goal object before (vs. after) the hand/claw arrives there. Error bars represent standard-errors. The quadratic curve represents the regression function with most explained variance.