Reaction time measures of pictorial and verbal encoding

Abstract

The relative availability of verbal and pictorial memory representations for pictures and words was examined in a visual matching task with reaction time (RT) as the dependent variable. Unfilled retention intervals of .5, 2, or 8 seconds were employed to estimate the effect of temporal parameters upon such memory representations. In Experiment I, Ss were unaware of the format (picture or word) of the first or second stimulus, while Experiment II included blocks of trials in which the format of the second stimulus was known. The results from both experiments suggest that visual and verbal memory codes are in general equally available for pictures, while the verbal code is more available for words. Thus, verbal labelling of pictures is easier than image generation for words. In addition, the visual code for pictures seems to decrease in availability more rapidly than the verbal code with increasing delay of the second stimulus, a result which might be due to a higher tendency to use verbal rather than visual rehearsal in short-term memory.