PURPOSE OF THE STUDY. When we reach to grasp an object we shape our hand for efficient grasping well before the contact point, implying that vision is able to influence the motor command during its execution. Inferring the shape of the grasped object from observing the action may be functionally important in anticipating the consequence of the action and in interpreting the intention of other people’s action. Here firstly we demonstrate the human visual system can retrieve the shape and size of the grasped object by observing only the visual kinematical information, and that these effects strongly depend on the view-point of the action respect to the observer.
METHODS. Visual stimuli were obtained by recording the right hand trajectories of 10 actors performing a reach and grasping task, using an optical motion capture system. The video sequences (≈ 1.5 s) were then transformed in dark point-line biological motion sequences (21 points) presented against a homogeneous background (see example sequence in figure). Four solid objects (cylinder, trunk, sphere and cube) were used, with 2 different volumes but matched height. Subsequently, the actors and additional 10 subjects were required to recognize the shape of the object in a 4AFC task. To control for the influence of dot positional and density cues of the final hand posture, we performed the task presenting as stimuli only the last frame of the video for 250 ms. The synthetic video and the last frame stimuli were presented in first or third person perspective (back and frontal view).
RESULTS. The subjects discriminated well the small (cylinder and pyramids) versus the large (sphere and cube) objects both for the moving (d’=1.64 ± 0.62) and the last frame stimuli (d’=1.16 ± 0.41). However a positive and significant (p<0.01) influence of the kinematics was present only when the view was in first person perspective. The effect of the view-point was larger for the actors when recognizing their own actions.
CONCLUSIONS. These results show that the visual system can infer the shape of an object by the reaching kinematics, but only if presented in a view-point consistent with the subject’s own movement. The view-point and the agent selectivity of these results support and strengthen the direct matching hypothesis (Rizzolatti et al., 2001) that considers that observing an action engage the same motor program used to generate it. In addition they indicate that motor programs can influence directly perception.
ESSENTIAL REFERENCES. Rizzolatti, G., Fogassi, L. & Gallese, V. Neurophysiological mechanisms underlying the understanding and imitation of action. Nature Rev. Neurosci. 2, 661–670 (2001).