On the Relationship between Letter Names and Superordinate Categories

Abstract

The sequence of processing steps in obtaining the classifications vowel-consonant and letter-digit were explored. The procedure involved measuring the RT to classify pairs of stimuli as “same”. Inferences were drawn from RT_s to letters with the same name (e.g. Bb) and from other differences between the RT structures obtained in the two classifications. When “same” was defined as both vowels or both consonants the subjects seemed to determine the letter name and then classify the name into its superordinate category. When “same” was defined as both letters or both digits the subjects appeared to classify the visual form directly into one of the two superordinate classes without first obtaining the name. The most likely explanation for the difference between conditions is the nature of training which subjects have had while learning the classifications.