Orthotactic sensitivity vs. phonological constraints on word recognition: An ERP study with deaf and hearing readers

Orthotactic sensitivity vs. phonological constraints on word recognition: An ERP study with deaf and hearing readers

First Author: Brittany Lee -- San Diego State University
Additional authors/chairs: 
Priscilla Martinez; Katherine J. Midgley; Phillip J. Holcomb; Karen Emmorey
Keywords: EEG, ERP, Deaf/Hard of Hearing, Orthography, Phonology
Abstract / Summary: 

Purpose: Fluent readers integrate phonological, orthographic, and semantic information during word recognition. A study by Frankish and Turner (2007) Illustrated the role of phonology in word recognition using a masked lexical decision task. Typical hearing readers mistook non-words for words more often if they were unpronounceable versus pronounceable, while dyslexic readers with weak phonological decoding skills showed no such effect. We used the same task and event-related potentials (ERPs) to investigate pronounceability effects in deaf adults who also had reduced access to English phonology but were skilled readers.

Method: Twenty deaf and twenty hearing adults (matched in reading skill and IQ) completed a masked lexical decision task. Non-words all contained a transposed letter (TL) and were rated as pronounceable or unpronounceable in a norming study (e.g., barve vs. brvae). We analyzed accuracy of responses as well as the N400, an ERP component associated with lexico-semantic processing.

Results: Deaf readers were more accurate at rejecting TL nonwords than hearing readers. Neither group exhibited behavioral differences in their responses to pronounceable versus unpronounceable non-words. However, pronounceable non-words elicited a larger amplitude N400 than unpronounceable non-words in both groups. Hearing readers also had a larger amplitude N400 to pronounceable non-words that they correctly rejected compared to those that they mistook for real words. This N400 difference was not seen for unpronounceable non-words and was not observed in the deaf group at all.

Conclusions: Hearing and deaf readers differ in their sensitivity to phonological constraints and possibly in the nature of orthographic representations as well.