26.3 Memory Models

There is significant evidence that generally memory for pictures is better than memory for words. This consistent finding is referred to as the picture superiority effect. At least three significant theoretical perspectives have been used to explain the "picture superiority effect," including: (1) the dual-code model (see 16.2.1, 29.2.3), (2) the single-code model, and (3) the sensory-semantic model (see Fig. 26-1).

Proponents of the dual-coding theory argue that there are two interdependent types of memory codes, verbal and nonverbal, for processing and storing information (Paivio, 1971,1978,1990,1991).

The verbal code is a specialized system for processing and storing verbal information such as words and sentences. The nonverbal system "includes memory for all nonverbal phenomenon, including such things as emotional reactions. This system is most easily thought of as a code for images and other'picture-like' representations (although it would be inaccurate to think of this as pictures stored in the head)" (Rieber, 1994, p. I 11). If it is assumed, as Paivio does, that the dual coding of pictures in verbal and nonverbal memory is more likely to occur for pictures than words, then the "picture superiority effecf' could be explained using dualcoding theory.

Figure 26-1. Three prominent models of encoding words and pictures. From "Pictureword differences in decision latency: a test of common-coding assumptions" by J. te Linde, 1982, Journal of Experimental Psychology: Learning, Memory, and Cognition 8, pp. 585, 586. Copyright 1982 by American Psychological Association. Adapted by permission. Also from "Leaming to order pictures and words: a model of sensory and semantic encoding": by D. L. Nelson, V. S. Reed & C. L. McEvoy, 1977, Journal of Experimental Psychology: Human Learning and Memory 3, p. 486. Copyright 1977 by the American Psychological Association. Adapted by permission. Also from "Research on pictures: a guide to the literature" by W. Howard Levie, 1982. In D. M. Willows & H. A. Houghton, eds. The psychology of illustration: hasic research, p. 1. Copyright 1982 by SpringerVerlag, New York. Reprinted by permission.

Proponents of a single-code model argue that visual information is transformed into abstract propositions stored in semantic memory (Anderson, 1978; Kieras, 1978; Kosslyn, 1980, 198 1; Pylyshyn, 198 1; Rieber, 1994; Shepard, 1978). Advocates for a single-code model argue that pictures activate a single semantic memory system differently than do words. Individuals "provided with pictures just naturally spend more time and effort processing pictures" (Rieber, 1994, p. 114).

Picture superiority can also be explained using a sensory-semantic model (Nelson, 1979). There may be a more distinctive sensory code for pictures, or the probability that pictures will be processed semantically is greater than that for words (Levie, 1987; Nelson, Reed & Walling, 1976; Smith & Magee, 1980). In many cases researchers in educational communications and technology have neglected the work that has been done concerning memory models.