Parafoveal visual information and semantic contextual constraints.

Two experiments are reported that examine the influence of semantic contextual constraints on an individual's ability to use parafoveal visual information. Subjects were presented either a word (reptile) or a row of Xs in foveal vision along with a parafoveal nonword (snckks) centered 2.3 degrees or 5 degrees to the left or right of fixation. The subjects were asked to pronounce the parafoveal stimulus aloud. During their eye movement to that stimulus, the nonword was replaced by a word that was either (a) semantically related to the foveal item and visually related to the parafoveal preview nonword (snakes), (b) semantically unrelated to the foveal item and visually related to the preview (sneaks), (c) semantically related to the foveal item and visually unrelated to the preview (lizard), or (d) semantically unrelated to the foveal item and visually unrelated to the preview (limits). In Experiment 1, subjects were only given 250 msec to use the semantic context, whereas in Experiment 2, subjects were given 1,250 msec. The results of both experiments yielded highly significant effects of contextual constraints and parafoveal visual information. However, the first experiment yielded additive effects of the two variables, whereas the second experiment yielded interactive effects. The results are discussed in light of recent arguments regarding the importance of contextual constraints for the use of parafoveal visual information.