Psychological theories of actual causation aim to characterize which of multiple causes of an event is singled out as the pri- mary cause. We present one such theory called the continu- ity account of actual causation. The continuity account treats events as changes of state in continuous time and traces a se- quence of stage changes backwards through time from an event to its primary cause. The account is broadly compatible with the physical process view of causation and we test it by ask- ing people to identify the primary cause of events occurring in simple physical systems. An initial experiment confirms that root causes are more likely to be chosen as primary causes than are immediate causes. A second experiment demonstrates that root causes that have temporal continuity with the effect are preferred even when probability raising accounts would pre- dict otherwise. The results of both experiments are consistent with the continuity account, and suggest that inferences about changes of state in continuous time may underpin an important class of actual causation judgments.