Facilitation in Word Recognition: Experiments Causing Change in the Logogen Model

The logogen model has been evolved to account for a wide variety of phenomena in the area of word recognition, and experiments on facilitation of picture recognition indicate at least two relevant levels and a two-way independence of picture and word recognition.

Abstract

The logogen has been defined as the unit which makes a particular verbal response available from whatever source. The simplest theoretical position has been that it was responsible for facilitation and the word frequency effect through either the experience or the production of words. ThuSj producing a word in response to a picture should have an effect equal to that of reading the word, on the subsequent visual duration threshold of the word. An experiment by Winnick and Daniel (1970) goes against this requirement. This experiment has been replicated, and other experiments indicate modality specificity in facilitation. These results force the adaptation to the theory of two input logogens and an output logogen to replace the original single structure. Experiments on facilitation of picture recognition indicate at least two relevant levels and a two-way independence of picture and word recognition. The logogen model (Morton, 1964, 1968, 1969, 1978) has been evolved to account for (or in response to) a wide variety of phenomena in the area of word recognition. The focal part of the model, the logogen system, is made up of devices which are responsible for producing a phonological code whenever appropriate. The linguistic unit with which the logogen is concerned is, roughly, a word. Thus, if the word "chair" is spoken, if the letter sequence CHAIR is presented, if someone is asked to name the object or free associate to "table," or if the word arises in spontaneous speech, the same logogen is terminally responsible. Since it obviates the need for multiple representations of the phonological code for a word, this is clearly the most economical hypothesis.