Retrieval and response criterion explanations of the effects of text organization on memory were tested in four experiments. The text materials were constructed so that the same target information was high in the content structure of one passage and low in another passage. In all experiments, the results showed that more target information was free recalled when it was high than when it was low in the content structure. In three experiments, it was found that three types of retrieval cues reduced the recall differences between information high and low in the structure, consistent with a retrieval hypothesis. In a fourth experiment, it was shown that manipulation of the subjects' response criterion had little influence on the effects of text organization on memory.Previous studies (Britton, Meyer, Simpson, Holdredge, & Curry, 1979;Meyer, 1975) have shown that free recall of information high in the content structure of a text is superior to recall of information low in the structure. Three mechanisms have been proposed for this result: selective attention, retrieval, and changes in the response criterion. The selective attention hypothesis, intuitively the most appealing, claims that information high in the content structure is selected for extra attention and processing at the time it is read. This extra processing leads to a higher probability of encoding than when the information is low in the content structure. However, some previous tests (Britton et al., 1979) did not support the hypothesis. Another possibility, considered more fully in the General Discussion section, is that the number of retrieval paths is increased by selective attention.The studies reported in the present article are concerned with the retrieval and response criterion mechanisms. According to the retrieval hypothesis, the effect of the Requests for reprints should be sent to
In most studies of relational and item-specific processing, category sorting and pleasantness rating have been the main procedures used to induce these two types of processing. Because the two types of processing have been studied in a wide range of memory phenomena (Hunt & McDaniel, 1993), it is strange that other tasks have not been proposed and tested. The present experiment demonstrates that equivalent results can be obtained with three relational processing tasks (category sorting, narrative construction, and relational imagery) and equivalent results with three itemspecific processing tasks (pleasantness ratings, familiarity ratings, and single imagery).
Experiment I demonstrated that the percentage of letters used in constructing abbreviations of common words decreased with word length (4-5, 6-7, 8-9 letters) and Thorndike-l.orge (T-L) frequency (1-24, 25-49, A, AA). The contraction principle (omitting interior letters) was preferred with short words and truncation (omitting terminal letters) with long words. Number of different abbreivations increased with word length, but not T-L frequency.Experiment II asked 5s to construct abbreviations by systematic use of contraction, truncation, or both. The results were virtually the same as before and showed that 5s can use the abbreviation rules in a consistent fashion. Experiment III required 5s to reconstitute the original words from the abbreviations produced in Experiment I. Although approximately 67% were reconstructed correctly, word reconstruction varied greatly with T-L frequency, word length, and sex of 5. The experiments were related to intraword constraint and coding processes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.