Isolation of items and the stimulus suffix effect.

Thirty Ss heard and serially recalled 36 different nine-digit series. The recall cue was a buzzer, the wordzero read in the same voice that read the first seven digits, or the wordzero read in a voice different from the one that read the first seven digits. In half the series, all nine digits were read in the same voice. The final two digits of the remaining series were isolated by having them read in a voice different from the one that read the first seven digits. Zero suffixes disrupted recall of the terminal digits more than the buzzer (the stimulus suffix effect) and had a greater degrading effect when they were in the same voice as the final two digits than when the zero suffixes and the final digits were read in different voices. Isolated digits were more likely to be recalled than nonisolated digits. Results suggest the importance of perceptual analysis in both the stimulus suffix effect and the isolation effect.