M.A. (Master of Arts)
Department of Psychology
Information theory in psychology
Paivio (1971) has proposed that information may be processed differently as a function of its concreteness. Specifically, concrete information may be encoded as a visual image and processed in parallel while abstract information is encoded into a verbal code and processed serially. A recent study by Seamon (1972) lent support to this hypothesis. Seamon found that, in a task in which subjects were required to indicate whether a word was or was not a member of a previously presented memory set, decision times increased linearly with increase in memory set size when subjects were instructed to rehearse or generate separate images of the items in the memory set. However, when subjects were instructed to generate an integrated image using the items in the memory set, decision times did not increase with increase in set size. It was concluded that, consistent with Paivio's hypothesis, subjects in the first two groups encoded the information into a verbal format and retrieved it serially while subjects in the third group encoded the information into a visual image and retrieved it in parallel. The purpose of this study was to further investigate Paivio’s hypothesis, especially in relation to concrete and abstract phrases. To verify Seamon’s (1972) findings, a replication of his study was carried out (Experiment I). While the results failed to unequivocally support Seamon's findings, decision times in the integrated image group differed markedly in form from that of the other two groups. In an attempt to clarify and further extend the findings of the first experiment, a second study (Experiment II) was carried out in which meaningful concrete and abstract phrases were used in place of a memory set. The study comprised a3X2X2X3 factorial design, with three instructional conditions (rehearsal, imagery, and no instruction) constituting the between-groups factor, and probe type (positive or negative), phrase concreteness (concrete or abstract), and phrase length (one, two, and three nouns) constituting the within-groups factors. According to Paivio's (1971) hypothesis, decision times for concrete phrases should be similar to those found by Seamon (1972). Decision times for abstract material, however, should be independent of instructions, since abstract material can not be easily encoded into a visual image. The results of the second study failed to support Paivio's (1971) hypothesis. Thus, although decision time functions for concrete phrases were found to have less slope than abstract phrases, these differences were not found to interact with instructional condition. The differences in slope were considered in relation to three possible hypotheses of encoding and retrieval. In addition, some conceptual and methodological problems in investigating Paivio's hypothesis were presented. Finally, a possible role for phrase comprehension was discussed in relation to two trends in the data of the second experiment.
Hillinger, Michael L., "Information retrieval in concrete and abstract phrases" (1975). Graduate Research Theses & Dissertations. 3575.
Northern Illinois University
Rights Statement 2