Neuroimaging of language production using overt speech

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): The proposed research will test methods for functional magnetic resonance imaging (fMRI) studies of language production using overt speech responses. Most fMRI studies of language production have used covert speech because of the susceptibility of this imaging method to movement artifacts. The use of covert speech, however, prevents the experimenter from monitoring subjects' compliance with the task and from obtaining important behavioral measures that could be related to the MR signal. The need to evaluate the accuracy of subjects' performance is particularly critical for subjects with language impairments. The proposed experiments will evaluate single word, phrase and sentence production using tasks in which subjects can produce the target responses within a short time period following stimulus onset. An event-related analysis will be used which allows for ignoring images collected during actual speech movements that occur prior to the hemodynamic response to cortical processing. Methods will be developed for recording subjects' responses and for amplifying the signal using adaptive noise cancellation techniques. Both normal and aphasic patients will be tested in a word production experiment and normal individuals will be tested in a phrase and sentence production experiment. If responses extend into the hemodynamic response for some conditions for either normal subjects or aphasic patients, additional analysis techniques will be employed in an attempt to co-vary out signal changes induced by movement. The proposed studies will also provide preliminary information on substantive issues concerning the role of the left inferior frontal cortex in language production. Previous lesion and neuroimaging studies suggest that the left frontal cortex is involved in using semantic information to select a response from competitors and in maintaining semantic information in short-term memory. The proposed studies will address whether the same brain region is involved in both. If these methods provide reliable and interpretable results, they will provide a means of determining the neural structures involved in producing speech for single-word and multi-word utterances. These methods could be used to address a number of health-related issues including evaluating reorganization of cortical function following recovery from stroke or other brain injury and determining language areas prior to surgery that might impinge on these areas.