A developmental study of the discrimination of letter-like forms.

It has generally been assumed in psychological analyses of the reading process that ability to discriminate letters is a prerequisite for reading and that children of 6 yr., when they begin to read, can in fact distinguish one letter from another with reasonable accuracy. The relevant literature is summarized by Vernon (1957, pp. 20 ff.). Some studies have used drawing as a criterion, some naming, and Nome matching. No one study has traced the development of letter differentiation as related lo those dimensions or features of letters which are critical for the task and which may present more or less difficulty. A method of studying this problem was suggested by an experiment of Gibson and Gibson i 1955) in which children of two age groups and adults identified "scribbles" which were systematically varied on three separate dimensions (number of coils, compression, and orientation). The scribbles were somewhat comparable to strokes used in cursive writing. The results showed that the difficulty of the task was greater the younger the 5, that errors varied in number with the dimension varied, and that systematic variation along at least three dimensions was possible. It also showed that confusion errors (primary generalization) provided an effective criterion for measuring ability to differentiate line drawings of this t\pe. The present experiment was designed to t̂udy the development of the ability to discriminate visually a set of letter-like forms in children 4 through 8 yr. of age. The aim was not merely quantitative comparison of different age levels, but primarily a qualitative developmental study of types of error as related to certain critical features of letters. To secure information on qualitative changes, the plan was adopted of constructing