With National Science Foundation support, Dr. Wagner will conduct a three study of the role of the pre-frontal cortex in selection of task-relevant knowledge mediating goal-directed behavior. Human cognition is marked by complex mental behavior. Such behavior frequently depends on cognitive control processes that permit an individual to access and work with internal representations in a goal-directed manner, especially within the context of competing goal irrelevant representations. Given the centrality of control mechanisms in cognition, a fundamental challenge is to specify the cognitive and neural architectures of control processes. Initial efforts point to prefrontal cortex (PFC) as a locus of control, with theories positing that PFC is a component of the neural circuitry that guides access to and selection of task- relevant knowledge through interactions with posterior association cortices that may represent target knowledge. Moreover, in the process of mediating goal-directed behavior, PFC-control processes may contribute to the encoding of information in episodic memory. Understanding of PFC contributions to cognitive and mnemonic control has markedly advanced over the past few decades, with significant insights often deriving from the integration of multiple experimental techniques, such as the recent combination of cognitive behavioral paradigms with functional brain imaging methods. While these efforts have yielded important advances, fundamental questions remain regarding the nature of PFC control mechanisms, their necessity for goal-directed behavior and memory formation, and the temporal dynamics of these operations. Leverage on these outstanding questions can be gained through the integration of multiple neuroimaging methods when each provides a different perspective on brain function. In the proposed research, a unique multi-modal neuroimaging approach will be adopted. Spatio-temporal imaging will combine the spatial resolution of event-related functional magnetic resonance imaging (fMRI) with the temporal resolution of electromagnetic imaging techniques - magnetoencephalography (MEG) and electroencephalography (EEG). FMRI will characterize the functional properties of specific PFC sub-regions, and will guide the solutions for MEG/EEG analysis of their temporal characteristics. In addition, transcranial magnetic stimulation (TMS) -a technique permitting the momentary disruption of local neural activity in healthy individuals -will be implemented to address questions regarding the necessity of PFC regions for cognitive control, interference resolution, and memory formation. Single-pulse TMS will be applied to fMRI-identified PFC structures at MEG/EEG-determined temporal periods. The proposed research, which will build on our preliminary studies as well as the broader literature, will address three specific aims: Aim 1. Capitalizing on the spatial resolution of fMRI, we aim to test specific hypothesis regarding the nature of PFC contributions to stimulus processing, controlled retrieval and selection, interference resolution, and episodic encoding. FMRI will adjudicate between models of PFC cognitive control by (a) exploring whether PFC subserves content-specific or contentgeneral bias mechanisms, (b) testing whether PFC mediates controlled retrieval or selection of task relevant knowledge, and (c) assessing the relation between these control mechanisms and behavioral susceptibility to interference. Aim 2. Capitalizing on the temporal resolution of MEG/EEG, we aim to test the hypothesis that PFC provides a "top-down" bias on neural computations in posterior cortices, thus subserving access to, selection of, and episodic memory for target knowledge. MEG will provide complimentary evidence to fMRI, thus further addressing the central questions in Aim 1. Aim 3. Using TMS, we aim to test the hypothesis that PFC subregions - identified with fMRI and MEG to be correlated with specific control processes - are indeed necessary for cognitive control, interference resolution, and memory formation. We will test whether specific subregions are differentially necessary depending on (a) the nature of the required control processes, (b) the content of the to-be-encoded experience, and (c) the strength of relevant and irrelevant associations. It is anticipated that the combination of spatiotemporal imaging and disruptive stimulation will provide important new evidence regarding the nature, necessity, and temporal dynamics of PFC mechanisms that subserve cognitive and mnemonic control.
