Abstract:
The discovery of “mirror neurons” in the frontal premotor cortex of monkeys demonstrates
that similar neural networks are activated when monkeys execute a goal-directed action and
when they observe the same action performed by another animal. Functional neuroimaging
studies have provided evidence for a functionally comparable system in frontal premotor
regions of the human brain. This suggests that our ability to understand the actions of others
involves the mapping of observed actions onto our own motor representations of those
actions. The overall objective of this thesis was to test and extend the “motor theory” of
action understanding using electroencephalography (EEG) and magnetoencephalography
(MEG). Experiment 1 showed that the EEG mu rhythm, generated in sensorimotor regions,
was desynchronised during the observation, execution and imitation of object-directed hand
movements. In Experiment 2 the desynchronisation of the mu rhythm was shown to be
specific to observation of object-directed hand movements as opposed to mimicry of objectdirected
movements. This is consistent with the functional properties described for monkey
mirror neurons, which only fire during the observation of object-directed actions. Experiment
3 showed that the EEG beta rhythm is also sensitive to observations of goal-directed hand
movements, providing more direct evidence for the involvement of primary motor cortex in
action observation. Experiment 4 used MEG measurements to demonstrate that the beta
rhythm is also sensitive to observations of mouth movements. Experiment 5 extended the
MEG results of Experiment 4, showing that like hand movements, object-directed mouth
movements cause stronger sensorimotor responses than non-object-directed mouth
movements. In this experiment activation of medial primary sensorimotor cortex was
observed during the observation of non-biological movements (a mechanical aperture opening
and shutting). These data are compatible with theories that claim a motor component to the
perception of structured event sequences. Observations of linguistic mouth movements showed significantly less activation in parietal and sensorimotor cortices, indicating a shift of
processing from dorsal to ventral streams. This suggests that linguistic movements are
processed with different neural networks from those recruited by normal action observation.
The implications of the results of this thesis are considered in the context of the motor theory
of speech perception, social understanding and phenomenology.