George W. McConkie

The span of the effective stimulus during a fixation in reading

A computer-based eye-movement controlled, display system was developed for the study of perceptual processes in reading. A study was conducted to identify the region from which skilled readers pick up various types of visual information during a fixation while reading. This study involved making display changes, based on eye position, in the text pattern as the subject was in the act of reading from it, and then examining the effects these changes produced on eye behavior. The results indicated that the subjects acquired word-length pattern information at least 12 to 15 character positions to the right of the fixation point, and that this information primarily influenced saccade lengths. Specific letter- and word-shape information were acquired no further than 10 character positions to the right of the fixation point.

What guides a reader's eye movements?

Abstract Five categories of models of eye guidance in reading are described. Eye movement data from 10 college-age readers were analysed. Correlations between the lengths of successive saccades, between the durations of successive fixations, and between successive fixation durations and saccade lengths were very low. The number of fixations centering on a letter was related to the length of the word it was in. It is argued that eye movements in reading are under momentary, non-random control. Models attempting to account for eye movement data on a random basis, or by simple gain controls or visual buffer monitoring are not found acceptable.

Asymmetry of the perceptual span in reading

An on-line computer technique was used to determine whether three skilled readers acquired visual information equally far to the left and right of central vision during fixations in reading. None of the subjects appeared to use visual information more than four character positions to the left of the fixation point (smaller distances were not tested), though all of them acquired visual information substantially further than that to the right. Thus, the region of useful visual information in reading is asymmetric around the fixation point.

Eye movement contingent display control in studying reading

A computer system has been developed that permits experimental control over a CRT display contingent upon characteristics of the viewer’s eye movements. The display can be changed during specific saccadic eye movements or fixations. Uses of such a system for studying reading are described. The paper reviews hardware and software considerations in developing such a system.

Perception and Cognition in Reading: Where is the Meeting Point?

It has been common in the history of psychology to think of reading as involving two sets of processes, one that makes visual information available (perceptual processes) and the other that makes use of that information in support of the language processes involved in reading (cognitive processes). Interactive theories of reading (Rumelhart, 1977) have questioned the usefulness of such a distinction, suggesting that each processing activity occurs in the environment of, and can be subject to influences from, all other processing taking place. Recent work, however, has suggested the existence of different processing modules that are not directly influenced by the products of some other processing activities (Fodor, 1983; Frazier & Fodor, 1978). In particular, it has been proposed that the visual processes that make information available are not influenced by higher level processes (Forster, 1979; Stanovich, 1980). This chapter presents some data from a study we have conducted that supports the utility of maintaining a distinction in theories of reading between perceptual processes that make visually-provided information available and cognitive processes that use this information for the purposes of the task at hand.

Eye Movement Control During Reading: The Effect of Word Units

In recent years psychologists have shown a renewed interest in eye movement in reading (see reviews by Levy-Schoen & O’Regan, 1979; McConkie, 1983; Rayner, 1978a). This work has been motivated by more than a simple curiosity about the nature of eye movement control. Rather, eye movement data are regarded as having the potential for testing theories about the ongoing perceptual and language processing taking place during reading. As people read, a great deal of variability is exhibited in how far they move their eyes and in how long their eyes remain centered on different locations in the text. There is general faith in, and some evidence for, the notion that this variability reflects differences in the nature of the perceptual and cognitive processes occurring at different locations in the text. It is assumed that if we could discover the ways in which mental processes influence eye movement behavior, then we would be able to use eye movement records to infer the nature of the processing occurring at different places in the text. In effect, the eye movement pattern would then become a language by which the brain communicates some of its activities to the psychologist. The hope that this can be achieved is a strong motivator for research on eye movement control in reading (Just & Carpenter, 1980; McConkie, Hogaboam, Wolverton, Zola, & Lucas, 1979).

Investigation of Reading Strategies: I. Manipulating Strategies through Payoff Conditions

Reading strategies of several groups of college students were manipulated by using payoff structures which stressed speed vs. retention. The influence of four variables on reading speed and test performance was studied: existence of a payoff structure, type of payoff structure, clarity of payoff instructions, and presence or absence of questions after each passage. Reading rates varied substantially under these conditions, but test performance did not. The potential usefulness of payoff systems for the study of reading strategies is pointed out.

The Definition of Test Items: A Descriptive Approach

An approach is described for the characterization of test questions in terms of (1) the information in a passage relevant to answering them, and (2) the nature of the relationship of this information to the questions. The approach offers several advantages over previous proposals based on the specification of algorithms for the production of all possible members of a class of test items. Among these are (a) the ability to consider information in a passage that relates to a question but is not part of the text segment from which it was generated, and (b) a relative lack of reliance on procedures for uniquely assigning a structural description to the text.

1: Learning from Text

One of the most important goals of educational institutions in Western culture (as well as many other cultures) is to help students learn how to acquire information from written text. The person who can read well has a means of obtaining knowledge on almost any topic, with little assistance from anyone else. However, an equal burden falls on the writer, since characteristics of a passage, quite apart from the complexity of the content, greatly influence what readers understand and retain from it. An understanding of the psychological processes involved in learning from text should provide a basis for assisting people to become more effective readers and communicators. No book, let alone a single chapter, could adequately review all the research related to learning from text. Any comprehensive attempt would include research in many fields, including rhetoric, communication, education, artificial intelligence, and areas of linguistics and psychology as well as branches of philosophy and neurology. Thus this chapter must be sharply limited in its scope of review. Since Carroll (1971) has reviewed the literature related to learning from text through 1970, our focus is on publications in the first half of the present decade. We primarily consider studies involving prose, that is, connected discourse consisting of more than a single sentence. The studies included typically involved competent learners (usually adults) who read or heard individual short passages. Thus questions about the development of reading or listening competence, or about learning from more extended texts (for

Investigation of Reading Strategies: II. A Replication of Payoff Condition Effects*

An earlier study by McConkie and Rayner (1974) was replicated in which reading strategies of college students were manipulated through the use of payoff conditions. The influence of four variables on reading speed and test performance was investigated: existence of a payoff structure, the form of the payoff structure, type of payoff, and presence or absence of feedback. Essential characteristics of the earlier study were replicated. The existence of a payoff structure had little effect by itself, but the form of the structure produced significant changes in reading rate. Feedback on performance appears critical in producing substantial reading strategy changes in readers.