Parafoveal information is not sufficient to produce semantic or visual priming

Abstract

Marcel (1978) has shown that semantic priming can occur in the lexical-decision task even if the prime is masked to the point at which its presence cannot be detected. The purpose of the present experiments was to determine if primes that begin four or five spaces to the right of fixation can also produce semantic facilitation even though they are very difficult to recognize. Experiment 1 showed that facilitation did occur when the parafoveal primes used in the subsequent experiments were presented foveally. In Experiment 2, the primes were moved to the parafovea, but the task demands directed the subject's attention toward the fovea. When subjects were ignoring the information presented to the parafovea (an allocation pattern that should correspond to that used during normal reading), neutral primes were just as effective as either semantically related or identical primes. In Experiment 3, the task demands were altered so that subjects were actively attending to the parafoveal primes. Although subjects were trying to process the primes, there was still no evidence that benefits could be derived from parafoveal primes. A final experiment showed that subjects given extensive practice with the materials will produce large amounts of identity and semantic priming. The results support the conclusion that readers can benefit little from the preprocessing of information in the parafovea unless that information can be supplemented with contextual expectations.Consider a reader who encounters the following sentence without any prior context: "For example, some words are much more predictable than others. " If the reader successively fixates to the positions indicated by the boldface type, it can be seen that information in the parafovea about the immediately upcoming word will sometimes be only minimally constrained by the previous context (e.g., the word "words"), but that other words (e.g., the word "others") will be highly predictable. The initial experiments of the present report concern the extent to which unanticipated words in the parafovea are processed and, equally important, whether or not any preprocessing of the parafoveal information actually speeds the reading process when an eye movement brings that part of the text into the fovea. We have tested these basic issues using a variation of the lexical-decision task to determine if a parafoveal prime can facilitate the processing of a related target Portions of these data were presented at the meeting of the Rocky Mountain Psychological Association, Tucson, April 1980. We would like to thank Susan Jones for collecting most of the data for Experiment 1 and for serving as one of our long-term subjects in Experiment 4. Earlier versions of this paper benefited from the comments of Keith Rayner and the CIP group at NMSU. We are especially indebted to Roger Schvaneveldt for the use of his equipment, computer programs, and general lore for doing lexical-decision experiments. Requests for reprints should be sent to Kenneth R. Paap, Department of Psychology, Box 3452,...

Eye movements in reading and information processing: 20 years of research.

“…However, when the parafoveal stimuli are presented 2.3 ° or 3 ° from fixation, there is no difference between the condition in which the first 2 or 3 letters are preserved across the saccade and the condition in which all letters are preserved. It was suggested (McClelland & O'Regan, 1981;Paap & Newsome, 1981) that the basic pattern of results is obtained only when there is a limited target set, but Balota and Rayner (1983) demonstrated the robustness of the effect when a large stimulus set was used.…”

Semantic codes are not used in integrating information across eye fixations in reading: Evidence from fluent Spanish-English bilinguals

“…For examples of studies showing no parafoveal semantic processing, see Inhoff (1982), Inhoff and Rayner (1980), and Paap and Newsome (1981).…”

The availability of useful information to the right of fixation in reading

“…These findings suggest that partial parafoveal information can be utilized in a variety of situations. The results of some experiments (McClelland & O'Regan, 1981;Paap & Newsome, 1981) have suggested that the facilitation observed in parafoveal naming studies by Rayner and his co-workers (Rayner, 1978;Rayner, McConkie, & Ehrlich, 1978;Rayner, McConkie, & Zola, 1980) was limited to situations in which the subject had strong expectations about what the parafoveal word would be. Most troubling was the implication that the priming from parafoveal information might be restricted to (unnatural) situations in which the set of words to be recognized (and named) was quite small.…”