Direct Evaluation of Distances within the Alphabet

Abstract

Alphabetical and numerical ratings produce discrepant psychophysical functions. The hypothesis that the discrepancy could be due to a difference in spacing of consecutive letters as compared with the corresponding consecutive numbers, was tested by asking observers direct numerical evaluations of distances between the ends of different subsets of the alphabet. The distances were correctly evaluated if the subsets contained the A. If the subsets contained the Z, the size of the subsets near the Z was very slightly overestimated, while the same size near the A was underestimated by about 1 letter. Subsets inside the alphabet with nine or more letters were underestimated by 1 or 2 letters. Five-letter subsets inside the alphabet were estimated correctly when they were near the ends of the alphabet, and were overestimated by 1 or 2 letters when they were in the middle of the alphabet. The results obtained with the sets containing the A support the idea that the distances between consecutive representational letters and corresponding representational numbers ate the same. A similar result was found by Masin, et al., by asking observers to report partitions of the alphabet. An explanation of the discrepancy between alphabetical and numerical ratings in terms of a difference in subjective distances seems implausible.