Understanding the role of sign language in reading

Understanding the role of sign language in reading

First Author: Frances Cooley -- University of Texas at Austin
Additional authors/chairs: 
David Quinto-Pozos
Keywords: Deaf/Hard of Hearing, Vocabulary, Development, Bilingualism, Efficiency
Abstract / Summary: 

Deaf signers of American Sign Language (ASL) are often bilingual in ASL and English, although much variation exists with respect to spoken English and reading abilities in this population. For many signers, bimodal bilingualism is evidenced by the daily use of ASL for face-to-face communication and the reading and writing of English for other purposes such as education, social media communication, and personal development. This population provides opportunities to examine literacy development and reading processes for readers who rely heavily on visual language for everyday communication. This symposium addresses literacy development and processes for child and adult signers of ASL while also considering signed language. We assemble investigators from multiple labs, with approaches that span various frameworks and datasets that involve behavioral measures, eye-tracking, neuro-imaging, as well as classroom studies. A unifying theme of these presentations is the examination of relationships between signed language and literacy for DHH readers.

Symposium Papers: 

Vocabulary knowledge and reading for ASL-English bilinguals

First Author/Chair:David Quinto-Pozos -- University of Texas at Austin
Additional authors/chairs: 
Frances Cooley

Purpose: Vocabulary knowledge predicts reading comprehension for hearing children; we test this hypothesis for deaf and hard of hearing (DHH) children who use American Sign Language (ASL) and read English. We also use an eye-tracking paradigm to determine if vocabulary knowledge predicts reading performance of target words in sentences.

Method: 11 DHH children (3 females, mean age 9;11, range 9;4 – 10;6) have been enrolled in an ongoing study. Children were administered 1) ASL and English Vocabulary tests that examine four dimensions of vocabulary knowledge: meaning recognition, form recognition, meaning recall/synonym test, and form recall, 2) a test of ASL fingerspelling, and 3) tests of English reading comprehension (PIAT-R) and word reading (TOSWRF-2). Children also completed portions of a homophone foil paradigm using an Eyelink Portable Due to examine sentence reading (e.g., fixation times on target words).

Results: Preliminary analyses indicate correlations between ASL and English vocabulary knowledge (r=0.6, p<.001), English vocabulary and reading (PIAT-R: r=0.71, p=0.049; TOSWRF-2: r=0.51, p=0.19 n.s.), ASL vocabulary and reading (PIAT-R: r=0.51, p=0.22 n.s.; TOSWRF-2: r=0.47, p=0.27 n.s.), and fingerspelling and reading (PIAT-R: r=0.806, p=0.0087). Eye-tracking data show correlations between English vocabulary and first fixation time in two target conditions (correct: r=-0.71, p=0.05; homophone foil: r=-0.76, p=0.028), and non-significant correlations between ASL and English vocabulary and total fixation time.

Conclusions: Bilingual vocabulary knowledge predicts reading comprehension for DHH children. Correlations between eye-gaze fixation time and English vocabulary reveal vocabulary contributions to reading efficiency. Results have implications for DHH education and the study of bilingual literacy development.

The Role of Lexical Access in Peripheral Perception: Evidence from Sign Superiority Effects in Deaf Signers.

First Author/Chair:Elizabeth Schotter -- University of South Florida
Additional authors/chairs: 
Emily Johnson; Amy Lieberman

Purpose: Deaf signers’ larger attentional span allows them to read more efficiently. Does their experience processing sign language, which involves dynamic processing across the visual field, contribute to this enhancement?
Method: We tested whether deaf signers recruit language knowledge to facilitate peripheral processing through a sign superiority effect (i.e., better handshape perception in sign than a pseudo-sign; cf.word superiority effect) and fingerspelled word superiority effect (i.e., better letter perception in a fingerspelled word than a nonword).
Results: Accuracy on both tasks was higher for deaf signers than hearing non-signing controls, higher in the near than the far periphery, and there was a three-way interaction whereby deaf signers showed lexical superiority effects only at the far eccentricity and controls showed no interaction.
Conclusions: Thus, deaf signers recruit lexical knowledge to facilitate peripheral perceptual processing, which may allow them to visually process text efficiently when reading English.

Assessing the contribution of lexical quality and sign language variables to reading comprehension in deaf adults

First Author/Chair:Zed Sehyr -- San Diego State University
Additional authors/chairs: 
Karen Emmorey

Purpose: The Lexical Quality (LQ) Hypothesis proposes that the quality of phonological, orthographic, and semantic representations impacts reading comprehension. We evaluated the relative contributions of LQ and sign language variables to reading comprehension in deaf and hearing adults using hierarchical regression models.
Method: LQ variables were assessed using tests of phonology, semantics (vocabulary), and orthography (spelling). The contribution of sign language skill was evaluated using an ASL comprehension test, ASL Sentence Repetition test, and a fingerspelling test. Reading skill was assessed via two comprehension subtests from the Peabody Individual Achievement Test (PIAT) and the Woodcock Johnson (WJ). Non-verbal IQ and education were covariates.
Results: For deaf readers, the LQ model predicted 62% of the variance in PIAT scores
(N = 86), and 52% of the variance in WJ scores (N= 84). Semantics and orthography, but not
phonological awareness, contributed to reading comprehension. For hearing readers, the
model predicted 32% of variance in PIAT scores (N = 110), and 45% in WJ scores (N = 73).
All LQ variables significantly predicted reading, except orthography was not a predictor of
PIAT scores. For the sign language variables, the model predicted 29% of variance in reading
comprehension (N = 78); fingerspelling was the only significant contributor (PIAT scores
Conclusions: Strong orthographic and semantic representations, rather than precise phonological representations, predict reading skill in deaf adults. The predictive strength of LQ variables depends on how reading comprehension is measured for both groups. Fingerspelling skill is associated with reading ability in deaf adults.

The Development and Evaluation of a Reading Comprehension Program for Deaf Children who use ASL

First Author/Chair:Amy Lederberg -- Georgia State University
Additional authors/chairs: 
Brenda Schick; Victoria Burke; Lee Branum-Martin; Nancy Bridenbaugh

Purpose. Fingerspelling Our Way to Reading (FOWR) is a 30-minute a day supplemental literacy program developed for deaf children who sign. FOWR is designed to use ASL to teach children to read English. Three days a week children learn to read words that share a rime using fingerspelling to focus on sublexical units. A randomized control trial of the FOWR word recognition component indicated treatment children had higher word recognition than control children. However, there were no differences on generalization assessments. The purpose of the current study is to assess the effect of adding a complementary two-day a week reading comprehension component on children’s literacy outcomes.
Method. Teachers of 232 DHH children in kindergarten, first, or second grade taught FOWR 5 days a week. The word recognition component was taught three days a week. The reading comprehension component embedded taught words in connected text designed to explicitly link printed English sentences to ASL.
Results. Children who received the 5 day a week FOWR showed improved word recognition and fingerspelling phonological awareness abilities compared to control children (from the previous year’s RCT). Effects for the five-day treatment ranged from 1.8 to 2.3 times the monthly gains estimated in the control groups. The children also showed progress in understanding to read taught English structures, though these results did not generalize to standardized reading comprehension tests.
Conclusion. This quasi-experimental study suggests FOWR shows promise in improving DHH children’s reading, though increased intensity is likely needed to show more extensive effects.

Mapping reading networks in deaf & hearing children considering language modality

First Author/Chair:Chris Brozdowski -- Vanderbilt University
Additional authors/chairs: 
Angela Scruggs; David Quinto-Pozos; Melanie Schuele; James R. Booth

Purpose: Little is known about the brain mechanisms for successful reading in deaf and hard of hearing (DHH) children, or whether reliance on certain mechanisms differs with communication mode. In short, why are some DHH children good readers and others are not? We adopt the triangle model of reading to guide an imaging study that intends to identify neural networks involved in reading in order to formulate a neurocognitive model of reading for DHH children.
Method: This project uses functional magnetic resonance imaging (fMRI) and ‘localizer tasks’ (signed phonology and speechreading) to test 10- to 15-year-old hearing and DHH children; subgroups of DHH children include predominant signed language, predominant oral language, or bimodal language use. Children also complete reading fMRI tasks involving rhyming judgments and meaning judgments to words presented visually.
Results: Behavioral and neuroimaging pilot data come from 7 hearing and 8 DHH children ages 10-15. The speech reading task successfully localized the superior temporal sulcus (STS) and the signed language task successfully localized the supramarginal gyrus (SG). Moreover, signing deaf individuals showed activation in both of these regions during the reading task, whereas oral and hearing individuals showed activation in STS.
Conclusions: These pilot data suggest that neural networks involved in reading can be identified for DHH children with different communication modes. We hypothesize that communication mode influences reading ability for DHH children, which has implications for educational language use for DHH children.