Three experiments examined the effects of stimulus omission and stimulus change on dishabituation of the skin conductance response. In all experiments, subjects received 17 tone-light or light-tone (S1-S2) pairings, and the primary manipulations were omission of and change in S2 on Trial 16. Responses to S1 and S2 on Trial 17 constituted the data of primary interest. Experiment 1 (N = 72) demonstrated that omission of an expected stimulus was more effective in producing dishabituation than was presentation of an unexpected but not experimentally novel stimulus. Experiment 2 (N = 60) indicated that both omission of an expected stimulus and presentation of an unexpected and experimentally novel stimulus produced dishabituation. Experiment 3 (N = 72) revealed that both the omission and the miscuing of 52 produced reliable dishabituation. The results are interpreted as indicating the importance of a comparison between current input and the stored representation of previous stimulation in the development of habituation.