This project focuses on understanding how the brain constructs networks of interacting regions (i.e., neural networks) to perform cognitive tasks, especially those associated with language, and how these networks are altered in brain disorders. These issues are addressed by combining computational neuroscience techniques with functional neuroimaging data, obtained using PET or fMRI. The network analysis methods allow us to evaluate how brain operations differ between tasks, and between normal and patient populations. This research will allow us to ascertain which networks are dysfunctional, and the role neural plasticity plays in enabling compensatory behavior to occur. During the past year, criteria for assessing the adequacy of the goodness-of-fit of some of these network methods were examined.In the last year, we began delineating the functional networks involved in the production of spontaneous narrative speech in normal subjects using PET measurements of regional cerebral blood flow (rCBF), an index of neural activity. Of special interest were left hemisphere (LH) perisylvian regions in posterior superior temporal cortex (likely corresponding to Wernickes area) and in the frontal operculum (likely corresponding to Brocas area). The Broca region had strong functional connections during speech with other language associated areas in the LH, including regions in posterior temporal and inferior parietal cortex. The Wernicke area had strong functional links during speech with LH parietal, temporoparietal and frontal perisylvian regions. Most of these strong functional connections were absent in the control task (which involved producing laryngeal and oral articulatory movements and sounds devoid of linguistic content). Interestingly, correlations with right hemisphere homologous areas were larger for the control task. These results demonstrate that LH perisylvian regions interact strongly with one another during language production, but not during a task requiring similar muscle movements and vocalizations as speech. In patients who stutter, many of these strong functional linkages seem to be absent, suggesting that LH language-production networks are abnormal during stuttering.In order to understand the relationship between what is observed in functional neuroimaging studies and the underlying neural dynamics, we had constructed a large-scale computer model of neuronal dynamics that performs a visual object-matching task similar to those designed for PET studies. The model is composed of elements that correspond to neuronal assemblies in cerebral cortex, and contain different elements that are based on types identified by electrophysiological recordings from monkeys as they perform similar tasks. It includes an active memory network involving the occipitotemporal visual pathway and a frontal circuit, and is capable of performing a match-to-sample task in which a response is made if the second stimulus matches the first. A PET study is simulated by presenting pairs of stimuli to an area of the model that represents the lateral geniculate nucleus. Simulated PET data are computed from the model as it performs the tasks by integrating synaptic activity within the different areas. Simulated PET data similar to that found in actual PET delayed match to sample visual tasks were obtained, as were the correct neuronal dynamics in each brain region. In the past year, we expanded the model to simulate fMRI studies, particularly those that use an event-related design. It was found that cognitive activity not under experimental control, especially in anterior brain regions (e.g., frontal cortex) can confound interpretation of fMRI activity in terms of its underlying neural substrate. Recently, we have begun an expansion the model so that it can also simulate auditory processing. This modeling effort will be combined with PET and fMRI experiments of the same kind of tasks used in the simulation studies. - PET, fMRI, neural networks, speech, cerebral cortex - Human Subjects
Showing the most recent 10 out of 32 publications
