ERIC Number: ED223992
Record Type: RIE
Publication Date: 1982-Oct-11
Reference Count: 0
A Theory of Perceptual Learning: Uncertainty Reduction and Reading.
Henk, William A.
Behaviorism cannot adequately explain language processing. A synthesis of the psycholinguistic and information processing approaches of cognitive psychology, however, can provide the basis for a speculative analysis of reading, if this synthesis is tempered by a perceptual learning theory of uncertainty reduction. Theorists of information processing have used models from computers, communication, and linguistics. Reduction of uncertainty theorists have developed two models of how information is processed, template matching and feature analysis. Template matching posits a preconceived notion of a letter, while feature analysis depends upon a series of binary decisions. Distributional and sequential redundancy reduces uncertainty in reading by limiting the possible positions letters and words can occupy. Physiologically, binary type operations selectively destroy sensory stimulation as they travel along the optic nerve. Current hypotheses about short and long term memory deal with their structure, function, and capacity. The uncertainty reduction model, in which the reader's task is to reduce the gamut of alternatives so that the original semantic intent of the author can be reconstructed, possesses an explanatory capacity consistent with evidence regarding oral language development. Although one need not accept the pure mathematical foundations of uncertainty theory, a computer paradigm is helpful in resolving the seemingly contradictory positions of the psycholinguists and information processing advocates. The ability to direct attention automatically and naturally between the graphophonic, syntactic, and semantic levels of a message links the two approaches. (JL)
Publication Type: Information Analyses; Opinion Papers; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A
Identifiers: Uncertainty Reduction
Note: Paper presented at the Annual Meeting of the Inter