Role of serial order in the irrelevant speech effect: Tests of the changing-state hypothesis.

Abstract

The effects of irrelevant speech were examined on a range of memory tasks. A missing-item task, which relied on a nonserial strategy for recall, proved less sensitive to the effects of irrelevant speech than one calling on memory for serial order. The finding that the effect of irrelevant speech both on a recognition task and on a paired-associates task was modified significantly by articulatory suppression further suggested that memory for serial order is the dominant feature of these tasks and that it renders them vulnerable to disruption by irrelevant speech. Taken together, the results of the experimental series support the notion that tasks involving memory for serial order are particularly susceptible to disruption by irrelevant speech. These and other findings converge on the notion that interference with information processing by irrelevant sound is based on similarity of process rather than similarity of content.