Linguistic processing in the human brain: Neurophysiological studies of neural organization and selective attention

Neurophysiological studies were conducted with the aim of testing whether selective attention is capable of modulating the processing of linguistic stimuli. The first three studies provided prerequisite characterization of the neurophysiological correlates of linguistic processing and the effects of semantic manipulations upon these measures. Scalp recordings taken from a comprehensive electrode array identified multiple event-related potentials (ERPs) elicited by linguistic stimuli. The ERPs discriminated between orthographically legal and illegal strings, and between words with semantic content and grammatical function. For words with semantic content, two ERPs were sensitive to semantic manipulations, including the well studied N400 potential (Kutas & Hillyard, 1980, Science, 207, 203-205). Intracranial recordings using three linguistic tasks revealed the involvement of the anterior temporal lobe in N400 generation.; In the remaining studies, a novel paradigm was used to investigate selective attention to linguistic stimuli. The tasks were visual analogues of the dichotic listening task, in which stimulus selection is maximized (Cherry, 1953, Accoustical Society of America, 25, 975-979). Adaptations made for neurophysiological investigation of selective attention to simple stimuli (Hillyard, Hink, Schwent, & Picton, 1973, Science, 182, 177-179) were incorporated to unconfound arousal variables and behavioral responses from measures. Studies using spatial and color cues converged to demonstrate that perceptual as well as semantic processing was modulated by selective attention. Visual evoked potentials differed significantly as a function of attention, such as has been reported for selection of simple stimuli (e.g., Van Voorhis & Hillyard, 1977, Perceptions and Psychophysics, 22, 54-62). In addition, ERP indices of semantic processing were significantly attenuated to the "unattended" stimuli. These findings were especially striking when spatial cues were used to determine stimulus selection. Intracranial recordings identified foci of attention in visual areas as well as in the temporal lobe. The contribution of physical features and of higher-order linguistic constraints to selective attention was studied.