Abstract
This report summarizes an experiment with children in Grades 5, 7, 9, and 11, testing 48 subjects at each grade level. Subjects processed sequences of letters of lengths 4 through 10, in the form of either words, pronounceable nonwords, or random sequences. Performance improves with grade level, particularly with the pronounceable nonwords and random sequences, rather than with the words. However, effects on performance of materials or grade level are minimal for the shorter strings and increase markedly as a function of length. We conclude that the older subjects have learned to read using higher order functional units. In this way, they use their limited channel capacity more effectively and take advantage of the redundancies inherent in the spelling of English words.
Get full access to this article
View all access options for this article.
