Two experiments investigated the effect of encoding two standard intervals on the performance in a temporal generalization task. Previous research has suggested that when individuals compare current time intervals with a standard encoded within both the auditory and the visual modality, they apparently use both the auditorily and the visually encoded standard for comparison so that each interval was compared with a “mixture” of both standard representations. However, the results of this study showed that memory mixing also happened within a pure visual modality. Moreover, it could be shown that the occurrence of memory mixing depended on instructional ambiguity. If in the training phase both standard durations were erroneously declared to be the same, memory mixing did occur in the testing phase. If, however, both standard durations were described as being different, memory mixing was not observed.

temporal generalization, memory mixing, instructional ambiguity, visual modality, humans