THE ROLE OF ATTENTION IN MOTOR RESONANCE

Motor pathways are activated not only during the obvious task of producing voluntary movement, but also during observation of actions performed by others. Execution and observation share a common pattern of activation, so that a subliminal "motor resonance" (MR) response is evoked, in primary motor cortex and spinal circuits, which reflects the specific motor program encoding the observed actions. There is growing evidence that MR is mediated by a parieto-frontal neural network, called the “Action Observation Network” (AON). While often described as an automatic and cognitively unmediated response, there is some experimental evidence suggesting that MR can in fact be modulated by cognitive processes. In this thesis the role of attention during the observation of a grasping action in both central and peripheral vision was investigated in three different studies. In the first study, the level of attention that 56 subjects allocated to the observation of a cyclic flexion-extension hand movement was manipulated in four different experimental conditions. MR was measured as the excitability modulation of spinal motoneurones innervating a wrist flexor muscle (flexor carpi radialis), utilizing the H-reflex technique. In the first experiment (explicit observation) 14 subjects were asked to pay attention exclusively to the cyclic oscillatory movement of a hand. In the second experiment (semi-implicit observation) the attention of 14 different subjects was partly diverted from the hand movement, since they needed to monitor hand position in order to perform a parallel task. In the third experiment (implicit observation) 14 different subjects had to complete yet a different parallel task for which the hand movement was totally irrelevant. The modulation of H-reflex amplitude, i.e. of the MR response, in these experimental conditions was compared to a baseline condition, in which 14 new subjects observed the cyclic oscillatory movement of a mechanical device, which does not evoke any motor resonant response. Results show that attention manipulation in both second and third 5 experiments dramatically decreased the amplitude of the MR response, while not affecting its muscular and temporal specificity. These results support the hypothesis that MR response is not a fully automatic process, but can be modulated by top-down influences, such as selective attention. In the second study, MR was investigated during observation of actions viewed in the peripheral field, where vision is far less accurate and where they don’t automatically receive the same level of attention as in central vision. The excitability modulation of motor pathways was recorded, this time utilizing the Transcranial Magnetic Stimulation (TMS) technique, in 40 subjects who were asked to pay attention to a central fixation point on a screen while a hand grasping action was shown at 10° in their peripheral field of vision. TMS was selected for these experiments because it allows recording from more than one muscle simultaneously. Half of the subjects observed a video clip of a “natural” motor sequence (a hand grasping a red ball), and the other half observed an “impossible” version of the same grasping (a ball being grasped by flexing fingers towards the back of the hand). The presentation of an impossible movement allows us to define, through a questionnaire, to what extent subjects are able to perceive the kinematic aspects of the observed action. Motor-evoked potentials (MEPs) were elicited in the right OP and ADM muscles by stimulation of the left primary motor cortex, at different delays during the observed actions. Results show, first of all, that actions in near peripheral vision are effective in eliciting MR, but that the MR response is rough and inaccurate, compared to the response to the same actions viewed in central vision, because it does not reflect the motor program encoding the observed action. It is reasonable to hypothesize that due to their limited kinematic accuracy, these subliminal motor responses may provide information about the general aspects of observed actions, rather than specific motor information regarding the spatial and temporal activation of muscles during the observed action. Finally, in the third study the role of attention in shaping MR responses was investigated by manipulating the attention of 64 subjects during the observation of the same peripheral grasping action used in the second study. MEPs were recorded in OP and ADM muscles of their resting right hand, at different delays during the observation of the video. In the first experiment, 29 subjects were asked to maintain their gaze on a fixation point but to pay (covert) attention to the video shown in periphery. The results of the experiment show a motor facilitation only during the observation of the natural grasping and only in the OP muscle. In order to explain the different results for the observation of natural and impossible movements two more experiments were carried out. Specifically, in order to test the hypothesis that selective attention could have played a role in focusing motor facilitation only on the OP muscle, in the second experiment 16 subjects were asked to observe the video of the natural grasping in central vision, while explicitly focusing their attention only on the thumb of the moving hand. Results show a strong effect of selective attention even in central vision since motor facilitation appears only in OP muscle. Finally, in the third experiment the hypothesis that the lack of modulation recorded during the observation of the impossible movement might be due to the high cognitive demand of perceptual decoding was examined. To this aim, the impossible condition of experiment 1 was replicated, but in this case 19 subjects were allowed to see the exact kinematics of the impossible movement before the beginning of the experiment. The idea being that previous knowledge of the impossible action would decrease uncertainty and complexity in the decoding process, freeing resources needed for the development of MR during observation in peripheral vision. Results show that observation of the impossible grasping action in covert attention condition, after previous familiarization, restores normal facilitation also in the ADM muscle. Altogether these findings suggest that a minimal level of attention is sufficient for the development of MR responses with correct temporal and muscular specificity, but that the gain of the process can be modulated by selective attention. Moreover, these data indicate that responses can be completely inhibited when the decoding of the action requires demanding inferential processes, suggesting that MR is not a completely automatic, bottom-up process, but requires the deployment of sufficient attentive resources supporting cognitive processing of visuo-motor transformation.