Abstract

In this talk I will examine links between action and perception in the context of object manipulation. I will first show that the sensorimotor system and the perception system maintain independent representations of object weight used to guide manipulatory actions and judge weight respectively. I will then provide evidence that, when observing an actor perform manipulation tasks, observers implement “motor plans” that closely match those of the actor. We have examined sensorimotor and perceptual representations of object weight in the context of the size-weight illusion whereby the smaller of two equally weighted objects is judged to be heavier when lifted. We have shown that when repeatedly lifting such objects, people initially scale their fingertip forces to object size but quickly adapt their force output to the true object weights and thus exhibit accurate sensorimotor predictions. Nevertheless, the size–weight illusion persists. This indicates that the sensorimotor system can operate independently of the perceptual system. We have also shown that the size-weight illusion can be inverted through extensive (multi-day) practice lifting inverted objects for which weight varies inversely with size. Thus, perceptual representations of object weight can be altered but at a far slower rate than sensorimotor representations. We conclude that both the normal and inverted forms of the illusion arise from a mismatch between expected and actual weight at the perceptual level. According to the direct matching hypothesis, observers understand the actions of others by engaging a mechanism that maps an observed action onto motor representations of that action. We have directly tested this notion by examining observers’ eye movements when watching object manipulation tasks. We have shown that the coordination between their gaze and the actor’s hand is strikingly similar to the gaze-hand coordination when they perform the task themselves. In both cases, gaze is proactively directed to key landmarks important for hand movement planning and control. These findings suggest that during action observation, people implement task-specific eye movement programs that are directed by representations of the manual actions required in the task. We have also examined observers’ gaze behaviour when watching unpredictable actions. We show that observers still use, or attempt to use, proactive eye movements. Although the gaze behaviour of observers and actors differs in the kinematic details, it is similar in terms of the goals of the eye movement.