Beyond Automatic Motor Mapping: New Insights into Top-Down Modulations on Action Perception

Our ability to recognize other people’s actions is central to everyday life. Transcranial magnetic stimulation (TMS) studies have shown that observing an action induces activity in the observer’s motor system that replicates the muscle selectivity, direction and temporal profile of the observed movement when executed. This motor resonant activity has long been presumed to reflect an inner, automatic replica of the observed movement. However, recent empirical evidence has challenged this view by showing that motor resonance can be tuned to high-level features, such as the overarching goal and intention of the observed action, while being simultaneously influenced by top-down contextual factors. Interestingly, current predictive coding models provide a mechanistic account to explain how action recognition is achieved, stressing the role of prior expectations and the interplay of bottom-up and top-down processes in terms of matching and mismatching predictions across hierarchical levels of action representation. In this chapter, we first provide an overview of seminal TMS findings that point to the characterization of motor resonance as an automatic fine-grained simulation of the observed movement. Second, we discuss more recent sources of evidence supporting the notion of motor resonance as a flexible phenomenon, stressing the role of top-down modulations during action perception.