Decision-making behaviour is considerably affected by dynamic aspects of the task environment. First of all, as a dynamic situation continuously changes, a decision maker has to take time into consideration. Second, a decision maker can use feedback providing information on the effect of own actions on system change, elaborating the set of strategies that can be used to cope with the decision problem. Third, in dealing with uncertainty a trade-off has to be made between costs of action, for example information search, versus the risks involved in doing nothing. The present paper describes an experiment in which subjects had to control a system that changed over time. Deteriorations in system performance could either result from changes in system parameters or from false alarms. Of major interest was decision behaviour as a function of time pressure, in this case, speed of system decline. Decision strategies are described in terms of the time allocated to decision phases and in terms of behavioural indices related to information requests and actions. The results showed a general speedup of information processing as time pressure increased. The decision strategy remained constant across all experimental conditions: subjects waited until a specific value of overall system performance was reached, then requested information on the underlying cause, and subsequently executed an action. Under high levels of time pressure, however, this strategy led to a significant increase in system crashes. The findings indicate that people do not optimally react to the time dimension of decision problems. It is concluded that future research should investigate the effects of a priori probabilities of false alarms, and of the costs involved in the decision-making process.