This paper addresses the question of the most appropriate theoretical account of the phenomena of orienting and habituation. Several lines of evidence are reviewed. First, it is argued that the effects of stimulus omission require a comparator theory in which it is asserted that responses to iterated events result from a comparison between predicted and actual stimulus input. Second, the data from studies in which paired stimulus events are employed seem, at least at first sight, to be best explained in terms of a comparator theory in which a key role is ascribed to associative processes. Third, secondary task probe reaction time data indicate that events that elicit orienting also command processing resources, and that habituation involves changes in the manner in which events are processed. Finally, recent data on the effects of intermodality change indicate that electrodermal responses are larger on the change trial than on the first habituation training trial; these results seem problematical for noncomparator theories. However, other data on the context-specificity of habituation and on the effects of stimulus miscuing cast doubt on the usefulness of an associative analysis as a general account of habituation phenomena. Nevertheless, the weight of evidence seems to indicate that an adequate theory of human habituation must include a comparison process and must acknowledge that orienting and habituation involve a re-allocation of attentional resources.