Is first-person reaching experience is necessary to construe reaching as goal-directed? Across 5 experiments, we measured the responses of N = 152 three-month-old infants to reaches that varied in efficiency, goal (picking up an object vs. causing a change in its state), surface properties (wearing a mitten vs. not), and causal efficacy (acting by contacting the object vs. from a distance and after a delay). Pre-reaching expected a person to reach for and cause changes in (Exp 1-3) and pick up (Exp 4-5) an object efficiently, but only under certain conditions: When the person caused a change in the object’s state, infants only succeeded when her actions were spatiotemporally contingent, and when the person entrained the object, infants only succeeded when she used her bare hand to act. We also performed an exploratory meta-analysis across data from these experiments and past experiments using motor training (Skerry et al. 2013) (combined N = 262). This analysis showed that the spatiotemporal contingency uniquely enhances infants’ sensitivity to action efficiency. Together, these findings show that infants construe reaching as an intentional action prior to learning to reach themselves, providing strong evidence that sensorimotor experience is unnecessary for infants to form expectations about intentional action. Instead, we suggest that at the center of action understanding is the intuition that agents plan actions so as to cause changes in the world.
Analyses: Data, codebook, analyses
Materials: Stored on OSF repository (https://osf.io/rcsns/)