Scanning visual images generated from verbal descriptions

Abstract

Abstract Mental scanning studies have consistently shown that the longer the distance between two points of an imaged configuration, the longer the scanning time for this distance. In typical mental scanning experiments, the images explored during the scanning task reproduce configurations which have been perceptually processed by subjects in a previous learning phase. In the present study, this condition was replicated and extended to a new condition where subjects scanned over images they had elaborated from the verbal description of a spatial configuration. This condition produced a pattern of results highly similar to those obtained in classic experiments, i.e. a linear relationship between distances and scanning times. With additional learning of the verbal description, the correlation coefficient increased, and scanning times decreased, both reaching values comparable to those obtained in the condition involving perceptually based images. These findings support the assumption that the mechanisms governing mental scanning apply to either sort of image those derived from perceptual processing and those constructed from the processing of discourse.