Semantic control of eye movements in picture scanning during sentence-picture verification

Abstract

Eye movements were monitored while people inspected a picture to determine whether it confirmed or disconfirmed a sentence they had just read. The sentences were quantified statements like Few of the dots are red, and the picture was a display of a large and a small subset of dots. It was found that the subset of dots a person fixated was determined by the semantic representation of the sentence rather than by its superficial referent. Since Few of the dots are red is semantically represented as a denial that the larger subset is red, people verifying the sentence fixate the larger subset, even though it is the small subset that is superficially referenced. By contrast, since A minority of the dots are red is semantically represented as an assertion that the smaller subset is red, people verifying this sentence fixate the smaller subset. A model of how people derive and compare representations of sentences and pictures is outlined. © 1972 Psychonomic Society, Inc.