Various theoretical accounts assume that the mental representations of actions and their perceived effects become intertwined as a result of experience, so that activating one may activate the other (e.g., Hommel 2001). As a result, executing actions could influence perception by activating associated effect representations (i.e., motor predictions) and perceiving effects could trigger the associated actions (i.e., ideomotor effects). These mechanisms have been used as explanations for automatic influences of action on perception (e.g., sensory attenuation) and vice versa (e.g., motor resonance, mimicry, goal contagion). Based on recent work in our lab, I will argue that these mechanisms are more limited and less automatic than previously assumed. First, I will demonstrate based on previous work in our lab that motor-based forward models have their limits (Dogge, Custers, & Aarts; TICS; 2019) and that perceptual predictions of action-effects outside our body seem largely based on flexible, higher level cognitive predictions and expectations. Second, I will share recent work on ideomotor actions (e.g., Sun, Custers, Marien, & Aarts; 2020) suggesting that these ideomotor effects are mainly driven by causal models of action-effect relations that apply in the situation at hand and may often be mediated by higher cognitive mechanisms (e.g., by biasing free choices), rather than triggering actions directly. I will briefly discuss the consequences for theorizing on automatic effects of perception on action and vice versa.
(Utrecht University)Ruud Custers