CROSS-MODALITY LANGUAGE PROCESSING

点击翻译按钮获取中文摘要

The global aim of this research is to study language comprehension within and between the visual and auditory modalities. The approach used will entail the recording of event-related brain potentials (ERPs) from over several regions of the brain as well as the recordings of overt behavioral measures while subjects engage in a series of auditory and visual experiments designed to tap specific linguistic processes. Although the proposed experiments will be performed in normal adult subjects, one long term goal, for which these studies provide a critical starting point, will be to study language processes between the modalities in normal children an children with specific reading disabilities. There are 4 specific aims and corresponding sets of experiments: (1) Cross Modality Priming. These studies will seek to determine if spoken and written words engage a common lexical system or if they maintain modality specific systems. In these experiments the interval between pairs of linguistic stimuli both within and between the visual and auditory modalities will be systematically manipulated in both semantic priming and repetition priming tasks. The hypothesis is that there will be a substantial level of commonality in the systems used by the two modalities, but that there may be subtle differences that reflect distinctive initial encoding operations. (2) Picture Priming. These studies will assess the degree of overlap in the systems used to process language and non-language stimuli in the same sensory modality. In these experiments the interval between picture and word stimuli will be systematically manipulated. The hypothesis is that both ERPs and behavior will show priming effects, but the pattern will not be the same as with purely linguistic stimuli. (3) Phonological and Orthographic Priming. This study will seek to determine if commonality of systems between the modalities extends to the orthographic and phonological features of spoken and written words. In these experiments subject will be required to make phonological or orthographic similarity judgments about written and/or spoken pairs of words. The hypothesis is that to the degree these characteristics are unique to the initial encoding of linguistic events in the two modalities there will be quite different patterns of ERPs and behavior between the modalities. (4) Cohort Studies. The goal of these studies is to determine if ERPs can provide evidence for a "point of uniqueness" in spoken word recognition an to determine if contextual factors can move word recognition prior to this point. These studies will also seek to determine if the point of uniqueness of spoken words constrains visual word recognition processes. In one experiment the time point at which stimuli become nonwords or unique words will be systematically manipulated and in a second experiment the point at which sentence final words no longer fit the semantic context of the sentence will be manipulated. The hypotheses are that ERPs and behavioral measures will reflect the point of uniqueness of words and nonwords and the point at which a word no longer fits with the context of a sentence.