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

Poster D72, Thursday, November 9, 6:15 – 7:30 pm, Harborview and Loch Raven Ballrooms

Electrophysiological predictors of efficacy for treatment of reading and language impairments

Paul Fillmore1, Michaela Ritter1;1Baylor University

Introduction. Much research has focused on phonological awareness intervention (PAI) for facilitating early reading skills in young children with and without language impairment (LI), and those at-risk for a reading disability (RD). However, there are limited data regarding the influence of PAI for school-age children, and efficacy in this population is a critical issue to explore. Additionally, it is crucial for both diagnosis and treatment to have an understanding of the relevant patterns of brain function associated with LI and RD, and what changes might result from successful treatment. Thus, the aims of the current study were to investigate the effects of phonological awareness intervention on reading abilities of school-age children and to identify the relevant brain systems associated with LI, RD and with treatment efficacy. Methods. The study utilized participants from Camp Success (n=55, age 5-16 years), a four-week intensive language and literacy intervention. All participants were identified with language impairment and met diagnostic criteria for reading and phonological impairment. Reading outcome measures consisted of oral reading fluency measures from Read Naturally, selected subtests of the TILLS, and the Test of Written Spelling. The intervention was an adaptation of Goldsworthy’s (2004) program, focusing on phonemic awareness and word decoding skills, and progressing in difficulty over the course of treatment. Resting electroencephalography (EEG) was collected in two five-minute intervals (eyes open, eyes closed) after completion of the four-week treatment protocol. Each session was processed using Neuroguide software for generation of EEG summary metrics (coherence, phase, power, etc.). Data were analyzed within standard frequency bands, and were compared to an age-referenced normative database via Z-scores, to assess differences between our participants and healthy controls. To evaluate the efficacy of PAI, paired t-tests were conducted for each outcome measure. To characterize the average patterns in the EEG for our sample, mean Z-scores were calculated for each measure of interest. To relate behavioral data to the EEG measures, Pearson correlations were run between each electrode/electrode pairing, and each behavioral measure, both for post-test measures and for improvement measures. Results. Most behavioral outcome measures were highly significant (mean Cohen’s d=0.67, range=0.11-1.62). Several differences between our sample and the normative EEG sample were noted, including: absolute power, amplitude asymmetry, coherence, and phase. For correlations between behavioral and EEG data, several patterns were noted. For post-test measures, non-word spelling was related to both power and phase measures, and oral narratives related to coherence. For improvement scores, written spelling, non-word spelling and written narrative recall were related to different aspects of both power and phase. Specific spatial patterns will be discussed in further detail. Conclusions. We found that PAI for school-age children is of significant benefit. Furthermore, we found several measures of difference between the EEG’s of our population and a normative sample. In addition, several of these EEG measures were predictive, not only of behavioral variance at post-test, but of response to treatment. Further research will highlight the diagnostic power of these measures, and if they can yield meaningful information for planning of therapeutic intervention.

Topic Area: Language Therapy

Back to Poster Schedule