Resisting motor mimicry: Control of imitation involves processes central to social cognition in patients with frontal and temporo-parietal lesions

Abstract

Perception and execution of actions share a common representational and neural substrate and thereby facilitate unintentional motor mimicry. Controlling automatic imitation is therefore a crucial requirement of such a “shared representational” system. Based on previous findings from neuroimaging, we suggest that resisting motor mimicry recruits the same underlying computational mechanisms also involved in higher-level social cognitive processing, such as self − other differentiation and the representation of mental states. The aim of the present study was to investigate on a behavioral level whether there is a functional association between the inhibition of imitation and tasks, assessing the understanding of mental states and of different perspectives of self and other. In a sample of neuropsychological patients with frontal lesions, a correlation between the ability for mental state attribution and the control of imitation was found, with a similar effect in the control group. Temporo-parietal lesioned patients showed a highly significant correlation between imitative control and visual and cognitive perspective-taking. Even after controlling for executive functions, the results remained significant, indicating the functional specificity of this relationship. These findings provide new insight into the functional processes underlying the control of shared representations and suggest a novel link between embodied and higher-level social cognition.

Keywords:

• Perspective-taking
• Mentalizing
• Mirror system
• Shared representations
• Lesion study

This work was supported by the European Community's Sixth Framework Programme, NEST /EDICI (Contract No. 012929).

Notes

¹ The Bonferroni correction of the alpha level was computed by dividing the alpha level (one-tailed) by the number of comparisons, separately for each group and task. This resulted in adjusted alpha levels of .0166 (for p < .05) and .0083 (for p < .01) for the ToM task, and alpha levels of .0125 (for p < .05) and .0062 (for p < .01) for the other tasks.