The sound of actions: a mismatch negativity (MMN) study

The ability to derive the intentions of others from the sound produced by their actions is quintessential to effective social behaviour. Many neuroscientists believe that this ability depends on the brain’s mirror-neuron system, which provides a direct link between action and perception. Precisely how intentions can be inferred through actionperception, however, has provoked much debate. One challenge in inferring the cause of a perceived action, is the fact that the problem is ill-posed, because identical movements can be made to perform different actions with different goals. Here, we show how, in the auditory modality, identification of most likely cause of a human action-related sound is highly subject to inferences. Using multi-channel, event-related potentials (ERPs), we determined the temporal dynamics of the ability to decipher action sounds by recording the mismatch negativity (MMN) generated in response to multi-deviant stimuli consisting of 3 different human action-related sounds (click of the tongue, hand clapping, and footsteps) and a non-human action-related sound (water drop). Subjects listened to the original sound-stimulus and to sounds obtained by altering 1 (low degree of disguise) or more complex (high degree of disguise) acoustic parameters of the original sound.