This paper starts by considering an interesting argument of H.A. Prichard’s against the view that to act is to cause a change; the argument is that causing is not an activity. The argument is important because of the recent emergence of an ‘agent-causation’ view according to which actions are the causing of changes by agents. I suggest a way of responding to Prichard’s argument, and then, profiting from one of his own conclusions, turn to consider the relation between neurophysiological changes and the causation of bodily movement by the agent. I consider a suggestion of Dretske's about how to understand this relation, but maintain that this suggestion cannot be got to work. I make my own suggestion about the proper way to understand the relation between the neurophysiological changes, the bodily movements and the action.