Explicit memory for unattended information

Abstract

Explicit recognition memory of unattended information was tested in two studies. College students performed complex mental addition problems in the presence of distracting words, with instructions to concentrate on rapidly and accurately verifying the accompanying arithmetic answers. Then, they took a surprise recognition test on the words. Experiment 1 showed that a short exposure (800 msec) resulted in chance levels of recognition performance, whereas a longer exposure (1,100 msec) supported recognition barely better than chance. Experiment 2 addressed whether attended and unattended encoding are qualitatively different mental states or instead the same state, differing only in the degree of attention given. A state-dependent memory effect was observed, in which reactivating the same attentional state at the time of test as had occurred at the time of study had beneficial effects on recognition performance. This outcome adds to other types of evidence, which suggest that attended and unattended encoding differ qualitatively. It was concluded that unattended encoding supports an impoverished degree of explicit, as well as implicit, long-term memory.