Date of Award

12-2014

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Program

Speech and Hearing Science

Research Advisor

Mark S. Hedrick, Ph.D.

Committee

Jessica S. F. Hay, Ph.D. Patti M. Johnstone, Ph.D. Ilsa E. Schwarz, Ph.D. Deborah von Hapsburg, Ph.D.

Keywords

children, hearing impairment, infant directed speech, novel words, vocabulary, word processing

Abstract

Objective. Little is known about how children with hearing loss (CHL) process words. The Emergent Coalition Model (ECM) of early word learning proposes that multiple cues (e.g., perceptual, social, linguistic) are used to facilitate word learning. Because hearing loss influences speech perception, different word learning patterns may emerge in CHL relative to children with normal hearing (CNH). One perceptual cue used by young children to access word learning is infant-directed-speech (IDS). Specifically, twenty-one month-olds can learn words in IDS but not in adult directed speech (ADS); however, by 27 months children can learn words in ADS. Currently, it is unknown how CHL process words in IDS and ADS. This study examined how CHL and CNH process familiar and novel words in IDS and ADS. A Looking-WhileListening paradigm was used. We predicted that: 1) CNH would show faster reaction time (RT) and higher accuracy than CHL, 2) word processing may show different patterns for familiar versus novel words, and 3) vocabulary size would be correlated with word processing skills.

Methods. Eleven children with bilateral, sensorineural hearing loss (M=32.48 months) using hearing aids or cochlear implants, and 11 CNH, matched for age, gender, and SES participated. Each child was tested in IDS and ADS on different days. At each visit, children were trained to map two novel labels to objects, counterbalanced across visits. Following training, accuracy and RT were assessed for both novel and familiar words. Vocabulary size was assessed using the McArthur-Bates Communicative Development Inventory.

Results. In the familiar word condition, for the CHL accuracy was significantly better in IDS than ADS, and RT was faster in IDS than ADS (but not significant). For CNH, accuracy was not different in IDS than in ADS, but RT was significantly faster in ADS than IDS. A significant speech type by group interaction was found (p <.05), for both accuracy and RT. Follow-up tests showed that CNH have higher accuracy and faster RTs than CHL. Results for familiar words suggest that while IDS may lead to more efficient speech processing for CHL, CNH are more efficient at processing ADS. In the novel word condition, only 10 CHL completed the task. For CHL, accuracy was marginally better in IDS than ADS, but no significant difference was observed in RT. For CNH no differences were seen in accuracy or RT between IDS and ADS. Analysis of Variance for RT showed that CNH have significantly faster RTs than CHL for novel word processing. For CHL, vocabulary size was negatively correlated with RT to familiar words in IDS and ADS, suggesting that children with larger vocabularies processed familiar words faster than children with smaller vocabularies. For the CHN, vocabulary size was marginally correlated with accuracy and RT to novel words in ADS.

Conclusions. This study demonstrates 1) the facilitative effects of IDS on word processing for young CHL, and 2) the relationship between word processing and expressive vocabulary in young children, suggesting that children with larger vocabularies are faster and more efficient at word processing tasks. The present findings suggest that CHL do not perform as well as their normal hearing peers on word processing tasks in ADS. These findings provide empirical evidence that childhood hearing loss affects processing of IDS and ADS differently than for CNH.

DOI

10.21007/etd.cghs.2014.0262

Share

COinS