An auditory analogue of the sperling partial report procedure: Evidence for brief auditory storage

Abstract Three experiments are reported on the partial report of material presented auditorily over three spatially different channels. When partial report was required by spatial location, it was superior to whole report if the cue came less than four seconds after the end of the stimuli (Exp. I). When partial report was required by semantic category (letters/digits) the relation between it and whole report depended on whether the S was asked also to attribute each item to its correct spatial location. When location was required partial report was lower than whole report and showed no significant decay with delay of the partial report indicator (Exp. II), but when location was not required, partial report was superior to whole report for indicator delays of less than two seconds (Exp. III). This superiority was, however, much less than that found in Exp. I when partial report was required by spatial location. These results are compatible with a store which has a useful life of around two seconds and from which material may be retrieved more easily by spatial location than by semantic category.