Recalling oral and written discourse.

Abstract

Listen

Translate article

We prescribed an expository text in two versions that varied in difficulty. The text was presented either orally or under one of two reading conditions, a normal reading or a moving window condition. The subjects were business trainees whose main communication experience was in either written or spoken communication. Immediately after presentation of the text, we asked the subjects to answer questions about it. The answers given after reading in the moving-window mode were in every respect the same as those given after self-controlled reading. There were differences between the listening condition and the two reading conditions, depending on the nature of the subjects' main communication experience and the difficulty of the text. We concluded from these results that the differences between the processing of oral and written discourse are caused neither by the difference of processing control nor by structural factors but by factors related to the communication experience of the subjects. Recall performance after reading and listening to the same discourse has been shown to be different in terms of the number of text statements recalled accurately (e.g., Green, 1981; Hildyard & Olson, 1982; Hron, Kurbjuhn, Mahdi, & Schnotz, 1985; Rickheit & Strohner, 1983). In these studies, reading resulted in more accurate free recall than did listening. Reading seems to be particularly advantageous in recalling difficult texts (e.g., Hildyard & Olson, 1978; Miisseler, Rickheit, & Strohner, 1985). But there are also results that show a better free recall after listening than after reading (Sannomiya, 1982, 1984; Thorndyke, 1976). This may be due to the ways the tasks were presented in these studies. For example, Thorndyke presented his visual text in lines. This procedure may have disrupted some units of meaning and thus prevented the text from being encoded as semantic units. The studies of Sannomiya (1982, 1984) show that omitting data at the level of micropropositions and recruiting as subjects pupils with insufficient reading ability could lead to a misinterpretation of the results. In the present study we attempted to clarify the modality effect and to establish the kind of mechanisms involved. There seem to be two plausible explanations for the modality effect: a structural advantage and a processing advantage of the visual modality. Proponents of the structural explanation claim, for example, that some inherent qualities of the visual modality, such as its neurological basis, result in more persistent traces than