EEG STUDIES OF DISABLED READERS

点击翻译按钮获取中文摘要

Over a 3-year period, we will study 100 disabled readers and 50 normal readers. The children in each group will be 8 to 10 years old. The disabled readers will be subdivided on two dimensions, phonological analysis skill and continuous naming speed which are theorized to be the two major weaknesses underlying reading/spelling disability (Bowers & Wolf, 1993). The children will be selected on the basis of standardized achievement test scores, and all will be within the normal range of intelligence. Additional tests of phonologic and orthographic analysis skill and rapid naming will be administered to provide concurrent validation data. The centerpiece of the project will be event-related potential (ERP) studies during which the children will be required to make primed rhyme judgements and lexical decisions. Three ERP components are of major interest: (1) the N400 component to target words, which is sensitive to phonological priming; (2) the N150-P200 complex, which at occipital sites is sensitive to orthographic effects (i.e., real and pseudowords versus consonant strings); and (3) the anticipatory slow wave that precedes target words in priming tasks. Both the magnitude and latency of the ERP components will be analyzed, as well as error and reaction time measures. Also to be studied are the subjects' EEG power spectra during silent and oral reading of word strings. Major quantitative EEG differences as well as performance differences (external validation) are hypothesized, between disabled subgroups and between all poor readers and normal readers. Overall, the most disabled readers are predicted to be those who have a double deficit in phonological analysis and naming speed. The more severely phonetically impaired children are not expected to exhibit consistent phonological priming to visual words. Auditory phonological priming, if found in these children, is expected to be more prominent in the right than left hemisphere, opposite to normal readers. Phonetically impaired children (whether or not slow namers) are not expected to show as consistent early activation in the left occipital area (the so-called Visual Word Form System; Posner & Raichle, 1994). Slow namers are expected to have slower reaction times, slower developing N150 and N400 components, and an attenuated ERP anticipatory component.