Statistical learning of conditional orthographic correspondences

Abstract

The English writing system deviates widely from the alphabetic ideal of uniform one-to-one correspondence between graphemes and phonemes, but its inconsistency is greatly reduced when conditional sound–spelling rules are applied. When reading or writing one part of a word, children and adults evinced knowledge of rules sensitive to the identity of other letters or phonemes, even those appearing much later in the word. Adults also showed sensitivity to the distinction between the basic and Romance subsystems of English (Albrow’s Systems 1 and 2). Children as young as 6 years applied conditional rules that they were not taught, indicating implicit statistical learning of patterns observed in text. But learning is imperfect, and even adults did not match the frequency with which the patterns are found in English words.