A comparison of priming effects on the nonword spelling performance of good and poor spellers

Abstract

This study examines the effects of lexical biasing on the nonword spelling performance of good and poor adult spellers of normal reading ability in an experimental (priming) phase and in a control (free-spelling) phase. Although the extent of lexical biasing demonstrated by these subjects was much less than that reported by Campbell (1983), significant lexical priming effects were noted for both groups. Poor spellers, however, showed greater susceptibility to the influence of an experimenter-provided lexical exemplar. Contrary to Campbell's findings, a negative correlation was obtained between spelling accuracy and extent of biasing. Sound-to-spelling contingency (i.e. the frequency with which spelling patterns represent vowel phonemes in words) also proved to be a significant factor underlying the performance of both groups. Results are discussed within a multiple-levels model of spelling in which performance is viewed as the outcome of an interaction between a primary routine, which rapidly generates contingency-based phoneme-grapheme subword correspondences, and an alternative whole-word routine, which generates nonword spelling by a process of analogy with a specific lexical exemplar.