Motor Preparation and Motor Imagery: Functional Equivalence is not Computational Equivalence

Vividly imagining action is thought to be functionally equivalent with preparing to act. Although evidence has shown that imagination can prime subsequent action, this is also the defining feature of action preparation. Curiously, neither theory nor experiment have thus far examined to what extent functional equivalence between these phenomena also constitutes a computational equivalence. To test this hypothesis, a response precuing experiment was designed whereby real responses were primed either with relevant imagination or preparation of an action. The information content and validity of precues were manipulated, allowing us to evaluate the relative costly, beneficial, and total congruency effects of these two conditions on subsequent performance. In four experiments, we observed consistency between conditions in their effects on the patterns of response time costs and benefits, demonstrating empirical evidence in line with the functional equivalence hypothesis. Additionally, the strength of priming, measured as the total congruency effect, was consistently larger in the imagery condition. This observation was not dependent on temporal uncertainty (Experiment 2), was a general feature of the motor system (Experiment 3), and could not be explained as a function of cognitive load (Experiment 4). To our knowledge this is the first evidence of a reliable difference between motor imagery and motor preparation, and therefore provides a novel qualitative aspect to the functional equivalence hypothesis. The implications for existing and future theories and models of motor imagery are discussed, with particular focus on the potential value of predictive coding frameworks.