Abstract
This experiment was based on the assumption that tachistoscopically presented non sense syllables were scanned from their persisting sensory trace or iconic memory, and then verbally rehearsed. In an experiment using the Sperling-type post-stimulus cueing procedure, three variables were manipulated: (a) accoustic similarity of the material, (b) time interval from stimulus offset to the recall cue, and (c) number of syllables presented together. The results showed linear increase in response time as well as linear decrease in recall performance, only with the increase in number of accoustically similar syllables presented. It was suggested that these results are explicable in terms of serial scanning from iconic memory and subsequent verbal rehearsal.
