Gina R. Kuperberg - US grants

DESCRIPTION: (Adapted from applicant's abstract)This is a request for an NIMH Mentored Patient-Oriented Research Career Development Award (K-23) entitled 'The Cognitive Neuroscience of Language Processing in Schizophrenia'. A disorder of thought and language has long been considered a core feature of schizophrenia. The candidate will test the hypothesis that such language abnormalities arise from specific cognitive and neural deficits in processing meaning (semantics). To address this question, she aims to acquire training in three complementary methodologies. First, psycholinguistic paradigms will be used to define the specific nature of language processing deficits in schizophrenia at a cognitive level. Second, electrophysiological experiments will determine the timing 01 neurophysiological abnormalities during language processing in schizophrenia, by examining the latency and amplitude of the N400- an event related potential that is known to be sensitive to semantic context. Third, event-related fMRI studies will characterize the functional neuroanatomy of language processing deficits in schizophrenia, by examining activity within the temporal and the inferior frontal cortex (particularly on the left) regions that are known to mediate semantic processing. The use of these three methodologies to address the same fundamental questions will give insights into the cognitive and neural basis of language dysfunction in schizophrenia, in both spatial and temporal domains. Additional didactic instruction and expert mentorship, in cognitive neuroscience, psycholinguistics and advanced statistics, will provide the theoretical framework and the analytical tools to integrate across these three methodologies, both conceptually and quantitatively. This training and research program will advance the candidate to the stage when she can establish herself as an independent investigator of the cognitive neuroscience of schizophrenia.

A disorder of thought and language has long been considered a core feature of schizophrenia. The broad objective of this proposal is to identify abnormalities in the neural dynamics of language processing in schizophrenia using multimodal imaging techniques. It is hypothesized that abnormally increased neural/synaptic activity within a temporal-prefrontal network (left superior and middle temporal, bilateral inferior temporal and left inferior prefrontal cortices) to semantic relationships between individual words leads to abnormal modulation of temporal-prefrontal activity as context is built up, word by word, within and across sentences. This cognitive neurobiological theory of language dysfunction in schizophrenia will be tested in a series of experiments studying (a) semantic relationships between word pairs, (b) the build up of context within sentences and (c) contextual integration across sentences (discourse). In order to investigate "when" and "where", a multimodal approach will be taken, integrating the advantages of two complementary neuroimaging techniques. First, electrophysiological experiments will determine the time course of neurocognitive abnormalities during language processing in schizophrenia by examining the N400 and late positive components - event related potentials that are known to be sensitive to contextual integration. Second, event-related functional (magnetic resonance imaging studies will characterize the functional neuroanatomy of language abnormalities in schizophrenia by examining the extent, magnitude and temporal dynamics of the hemodynamic response within the temporal-prefrontal semantic network. Third, data obtained using these complementary techniques will be directly compared, allowing inferences to be made about underlying neural/synaptic activity, and quantitatively combined, allowing the delineation of spatiotemporal patterns of activation across temporal and prefrontal cortices during language processing in schizophrenia.

DESCRIPTION (provided by applicant): Abnormalities of language comprehension characterize many neurodevelopmental and acquired disorders, leading to severe social communicative dysfunction. Many of these same disorders are also characterized by executive deficits of cognitive control, such as conflict resolution, selection, monitoring and adaptation. Yt we know little about the basic relationships between language processing and cognitive control. This grant uses multimodal neuroimaging techniques to ask how, when and where the neural systems mediating language comprehension and cognitive control interact in healthy adults. Our basic hypothesis is that we use context whenever we can to predict upcoming information at multiple levels of representation, and that, when an incoming word disconfirms some or all of our prior prediction(s), we engage in prolonged attempts to integrate that word into its context. We further hypothesize that the neural mechanisms mediating this prolonged integration are closely linked to core mechanisms of cognitive control, and that they depend on both the strength/certainty as well as the nature of our prior predictions (lexical vs. event structure). Finally, we hypothesize that they play an essential role in enabling us to rapidly and flexibly adapt our comprehension mechanisms to the ever-changing demands of our overall communicative environment. To test these hypotheses, we propose a series of experiments that systematically manipulate both the constraint and the nature of linguistic context, and we use three complementary neuroimaging techniques--event-related potentials (ERPs), functional MRI (fMRI) and magneto-encephalography (MEG)--to probe the temporal and spatial patterns of brain activity recruited to incoming words in these contexts. Under Aim 1, we probe the neural mechanisms that detect and resolve conflict between a strong, high certainty prediction and an incoming word that violates this prediction. Under Aim 2, we probe the neural mechanisms that select an incoming word that matches a medium certainty prediction and suppress alternative competing prediction(s). Under Aim 3, we ask whether different individuals can use the same context to predict with different strengths/certainties, leading some people to mount a high conflict neural response (see Aim 1), and others to mount a selection neural response (see Aim 2) to integrate the same linguistic input. We also ask whether these individual differences can be linked to behavioral performance and brain activity associated with executive tasks that tap into more general conflict control mechanisms (Stroop; AX-CPT). Finally, under Aim 4, we ask whether the same individuals can adapt to their wider communicative environment by predicting with different strengths/certainties, leading them to engage different neural mechanisms (a high conflict or a selection response) to integrate the same word into the same local contexts. By addressing these questions, this project will identify core neural mechanisms that can break down and lead to comprehension deficits in multiple different disorders. It therefore lays the foundation for the development of targeted, theoretically motivated neurocognitive strategies to prevent and improve communication deficits in these disorders.