ERIC Number: ED226331
Record Type: RIE
Publication Date: 1982-Oct
Reference Count: 0
Process Models of Reading: Some Data on the Initiation of Processes.
Young, Sheryl; And Others
Two experiments were conducted to distinguish empirically between the interactive and noninteractive models of reading. In the first study, 18 undergraduates divided into three groups by reading ability read stories in an RSVP mode. Two variables were manipulated within subjects--rate of presentation and amount of text presented per unit of time. Results showed that recall improved with increased unit size. Providing subjects with optimal or near optimal chunks reduced or eliminated one resource demand, thus permitting an allocation of those resources to other processes that contributed to recall. As the rate of presentation increased, recall deteriorated. To determine whether assisting the subject in chunking was responsible for the unit size effect, a new unit size was used in the second experiment with 84 different undergraduate subjects. In addition, units of text were presented so that the eyes remained fixated at the same location throughout every passage. Results showed that the elimination of eye movements improved performance. In general, the unit size manipulations in the two experiments demonstrated that assisting the subjects in chunking and thereby freeing resources used to chunk resulted in both improving text memory and increasing the number of macro propositions produced. Poor readers compensated for poor comprehension processes by slowing all processing down. It was concluded that the data were inconsistent with both serial and parallel noninteractive models of reading; that resource allocation among component processes was an integral part of any interactive model; and that resources and data were exchanged in both a top-down and a bottom-up manner during reading. (HOD)
Publication Type: Reports - Research
Education Level: N/A
Sponsor: Office of Naval Research, Arlington, VA. Personnel and Training Research Programs Office.
Authoring Institution: Colorado Univ., Boulder. Inst. of Cognitive Science.