The promise and pitfalls of studying the neurophysiological correlates of automatic imitation

Abstract

The automatic copying of other people – automatic imitation – is one of the most widely studied topics in psychology. In this chapter, we review current evidence for the neurophysiological correlates of automatic imitation. To do so, we focus on one heavily- used and influential paradigm that manipulates stimulus-response compatibility and is conducive to investigation using neuroscientific methods. We distinguish between two distinct mental processes that occur during imitation: action representation and action selection. Observed actions are perceived and represented and a particular action needs to be selected to be executed. Neuroscientific evidence concerning action representation appears relatively robust and engages a widespread and distributed network of visual and motor regions that span the ventral visual stream, as well as frontoparietal cortex. In contrast, the neurophysiological correlates of action selection during imitation are far from clear. The dominant view in the literature is that this particular task indexes processes relating to the control of automatic imitative tendencies, which rely on a self-other distinction mechanism that is uniquely tied to human social interaction and engages the theory-of-mind network. However, our analysis shows that this claim lacks essential evidence for four forms of validity (construct, internal, external and statistical-conclusion validities). Instead, given current evidence, the best estimate is that this task engages domain-general forms of control that are underpinned by the multiple-demand network. For claims to be supported regarding socially-specific forms of control when using this task, there is a burden of proof on researchers to show robust evidence for each of the four validities that we have outlined. Fortunately, with the emergence of the meta-science movement over the past 10 years, there are more resources than ever to help achieve this aim. More generally, even though we focus on one imitation task to provide a thorough test-case example, given the widespread and well-established lack of validity in psychology in general, we fully expect our analysis to be relevant across the full range of imitation tasks covered in this book, including imitation of speech, gestures and emotions.