You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

Poster D63, Thursday, November 9, 6:15 – 7:30 pm, Harborview and Loch Raven Ballrooms

Individual Differences in Subphonemic Sensitivity and Reading Ability

Monica Li1,2, David Braze1,2, Anuenue Kukona2,3, Donald P. Shankweiler1,2, Whitney A. Tabor1,2, Julie Van Dyke2, W. Einar Mencl2, Clinton L. Johns2, Kenneth R. Pugh1,2, James S. Magnuson1,2;1University of Connecticut, Storrs, Connecticut, USA, 2Haskins Laboratories, New Haven, Connecticut, USA, 3De Montfort University, The Gateway, Leicester, UK

Introduction: The link between phonological abilities and the development of skilled reading is well-established in both typical and atypical language development. However, the nature of phonological deficits in low ability readers remains a debated topic. One hypothesis is that poor readers have underspecified (or “fuzzy”) phonological representations (Tallal et al., 1998) that highly overlap with each other due to lack of detail in phonetic encoding. An alternative hypothesis is that reading difficulty derives from overspecified phonological representations, with perceptual categories of speech sounds as allophones instead of phonemes, resulting in mismatch between written symbols and spoken sound categories (Serniclaes, 2006). Methods: To examine hypotheses of under- vs. over-specified representations, we used a visual world eye tracking paradigm closely modeled after one used by Dahan et al. (2001) to investigate individual differences in sensitivity to subphonemic information in young adults with a wide range of reading ability. We recruited a community based sample of 67 college-aged native English speakers. They were enrolled in GED programs or community college, or not in school at all when they participated. In order to examine individual differences in our sample, we administered a comprehensive set of standardized measures with known connections to reading ability, such as listening comprehension, vocabulary, decoding skills, and phonological skills. During the experimental eye-tracking task, participants saw pictures of four objects (including a target, a competitor, and two unrelated distractors) on a computer screen, while following auditory instructions to click on the target picture. The auditory stimulus for each target word was cross-spliced with either another recording of the target word (W1, e.g., ‘cat’), a competitor (W2; e.g., ‘cab’), or a nonword (N3; e.g., ‘cag’), such that there were three versions of each target word with varying coarticulatory information: consistent coarticulation (W1W1; e.g., ‘catt’), misleading competitor coarticulation (W2W1; e.g., ‘cabt’), and misleading nonword coarticulation (N3W1; e.g., ‘cagt’). We expected that differences in eye-tracking trajectories across the three conditions would relate to individuals’ sensitivity to subphonemic information. Results: Consistent with Dahan et al.’s findings, our overall results showed that individuals were sensitive to subphonemic coarticulatory information: participants’ target fixations in the W1W1 condition were greater and faster than in N3W1, followed by W2W1. In addition, individuals’ phonological skills (as a proxy for reading ability) were negatively correlated with their subphonemic sensitivity (measured by target fixation differences between conditions), indicating that less skilled readers had higher subphonemic sensitivity. Conclusion: Our measures of the fine-grained time course of lexical activation during spoken word recognition in response to misleading coarticulation suggest that poorer reading abilities are associated with higher sensitivity to subcategorical phonetic detail, consistent with the overspecification hypothesis in developmental reading impairment proposed by Serniclaes (2006). Individual differences in reading ability and phonological representations implicated in the current study may guide future computational modeling and neurobiological work, deepening our knowledge for the underlying mechanisms and factors that contribute to the dynamic between phonological processing and reading skills.

Topic Area: Perception: Speech Perception and Audiovisual Integration

Back to Poster Schedule