Priming tool actions: Are real objects more effective primes than pictures?

Humans are faster to grasp an object such as a tool if they have previewed the same object beforehand. This priming effect is strongest when actors gesture the use of the tool rather than simply move it, possibly because the previewed tool activates action-specific routines in dorsal-stream motor networks. Here, we examined whether real tools, which observers could physically act upon, serve as more potent primes than two-dimensional images of tools, which do not afford physical action. Participants were presented with a prime stimulus that could be either a real tool or a visually matched photograph of a tool. After a brief delay, participants interacted with a real tool target, either by 'grasping to move,' or 'grasping to use' it. The identities of the prime and target tools were either the same (congruent trials; e.g., spatula-spatula) or different (incongruent trials; e.g., whisk-spatula). As expected, participants were faster to initiate grasps during trials when they had to move the tool rather than gesture its use. Priming effects were observed for grasp-to-use, but not grasp-to-move, responses. Surprisingly, however, both pictures of tools and real tools primed action responses equally. Our results indicate that tool priming effects are driven by pictorial cues and their implied actions, even in the absence of volumetric cues that reflect the tangibility and affordances of the prime.