Goals Our research focuses on the neurobiological mechanisms underlying cognitive control and its disturbance in psychiatric illness. We define cognitive control as the ability to flexibly adapt behavior to current demands, by promoting task-relevant information and behaviors over temporally-extended periods and in the face of interference or competition. This ability is central to most higher cognitive functions, and contributes to the unique character of human behavior. Our goals are to define the neural mechanisms that underlie cognitive control, to understand how these govern behavior, and to use this knowledge to improve our understanding of the relationship between brain and behavior in psychiatric disorders. Areas of Focus At the psychological level, our studies focus on attention, inhibition, working memory, language processing, and affective and evaluative, all of which are thought to be central to, or to rely critically on cognitive control. At the biological level, our work focuses on the functions of prefrontal cortex and anterior cinguate, and their regulation by brainstem neuromodulatory systems, such as those using the neurotransmitters dopamine and norepinephrine. Within the clinical domain, our work focuses on disturbances of these systems in schizophrenia. Research Approach Our work draws on methods from cognitive psychological, neuroscientific, and psychiatric research, and involves a two-pronged approach: a) the development of neural network-based computer simulation models of performance in cognitive tasks, to explore specific hypotheses about the influence of biological variables on behavior; and b) the empirical testing of these hypotheses in normal and psychiatric populations using standard cognitive tasks, as well as functional neuroimaging and psychopharmacological techniques. We use these two approaches in an iterative and interactive way: Models are used to generate empirical predictions, that can be tested in empirical studies. Empirical studies are conducted to both test predictions made by the models and to guide further development. Theoretical models are developed within the parallel distributed processing (PDP), or "neural net" framework. This permits the construction of computational models that can simulate human performance in cognitive tasks, using mechanisms that approximate those believed to be present in the brain. These models provide a conceptual link between cognitive processes and underlying neural mechanisms, and insights into how specific neurobiological abnormalities can produce the patterns of behavioral disturbances observed in psychiatric disorders.Our empirical studies make use of standard experimental methods from cognitive psychology, both alone, and in combination with non-invasive functional neuroimaging and psychopharmacological techniques. Neuroimaging techniques include functional magnetic resonance imaging (fMRI), positron emission tomography (PET), and scalp electrophysiology. Pharmacological studies focus on dopaminergic agents. All of these methods are used to study healthy volunteers and patients with schizophrenia. The goal of these studies is to deepen our understanding of normal function, and bring this to bear in our effort to understand and address the disruptions of cognitive functioning that occur in psychopathology.