Abstract
This study explored the effect of varying amounts of feedback on the reconstruction of scrambled text. The subjects were undergraduate psychology students at the University of Colorado. Half the subjects reconstructed a 26-sentence passage dealing with Watergate, while the other half read either the same scrambled version or the original passage. The independent variables were amount of feedback available, reconstructing or reading, and time of testing (immediate vs delayed). Dependent variables included percentage of idea units recalled, percentage of original sentences recognized from paraphrases, and for the subjects who reconstructed the text concordance (tau) between their final sentence order and the sequence of the original passage. Among the findings were (1) no statistically significant differences between feedback groups for percent of idea units recalled, and percent recognition scores; (2) concordance (tau) with the original order of sentences increased with greater amounts of feedback but was not associated with higher recall or recognition scores; and (3) the process of reconstruction assisted recognition but not recall. The implications of these findings for current assumptions about feedback and text processing are assessed.
Get full access to this article
View all access options for this article.