Date of Completion

5-9-2015

Embargo Period

4-26-2017

Advisors

Bernard G. Grela, Emily B. Myers

Field of Study

Speech, Language, and Hearing Sciences

Degree

Master of Arts

Open Access

Open Access

Abstract

Research on perceptual learning for speech shows that lexical information can be used to modify phonological representations. Recent findings suggest that lexically-informed perceptual learning is a domain-general learning mechanism such that lexically-guided learning is also observed in the processing of printed text. The literature on lexically-informed perceptual learning has extensively investigated the nature of the change to the prelexical representation. What this literature has yet to examine, however, is how varying levels of lexical recruitment influence this learning mechanism. Here we examine this question by comparing performance on lexically-guided letter perception between two groups of readers, average readers and advanced readers. The Lexical Quality Hypothesis provides a framework for our hypothesis, which posits that more efficient and richer lexical processing occurs in skilled readers compared to average or impaired readers. Adult monolingual, English readers were assigned to either the average or advanced reading group based on performance of standardized assessments of reading and reading sub-skills. All participants made visual lexical decisions to words and nonwords and then categorized members of an H – N letter continuum. The lexical decision task involved different exposure conditions. Participants in the H-bias group saw an ambiguous grapheme midway between “H” and “N” in H-bias lexical contexts, (e.g., WEIG?), in addition to words with a clear “N,” (e.g., REIGN), whereas participants in the N-bias group saw the opposite, (e.g., REIG?, WEIGH). In order to examine the effects of orthographic transparency on perceptual learning, some participants were presented with critical words that had one-to-one letter-to-phoneme correspondences, (e.g., AHOY), and different participants did not have such orthographic transparency, (e.g., WEIGH). Results indicate that both groups of readers used lexical information to modify letter perception. Strikingly, this learning effect was more robust for the advanced readers compared to the average readers. These results suggest that lexical quality exerts a gradient influence on lexically-informed perceptual learning of letters.

Major Advisor

Rachel M. Theodore

Share

COinS