Process of recognizing tachistoscopically presented words.

Abstract

A model for the recognition of tachistoscopically presented words is developed. The model is a "sophisticated guessing " model which takes explicit account of the geometry of the characters which make up the words or letter strings. Explicit attempts are made to account for word frequency effects, effects due to letter transition probabilities, and effects due to physical similarity of character strings to one another. A word recognition experiment using the set of three-letter words is reported, and the model is used to make quantitative predictions of these results as well as to give a qualitative account for a number of results in the literature. Finally, it is shown that under certain simplifying assumptions this sophisticated guessing model is isomorphic with the "criterion bias " model as developed in 1967 by Broadbent. The process whereby words are recognized has long fascinated experimental psychologists (Erdmann & Dodge, 1898; Huey, 1908; Pillsbury, 1897). Several related findings have commanded the most attention. For example, more letters per unit time may be apprehended when a word is presented than when a string of unrelated letters is presented (Huey, 1908). A letter string formed by taking a word and either deleting or replacing one or two letters is often clearly perceived as that word (Pillsbury, 1897). Other things being equal, words which occur frequently in the language are more easily perceived than those which occur less frequently (Goldia-1 This research was supported by Research