Carol A. Seger - US grants

Concept learning is learning about the regularities or patterns that separate stimuli into different groups, or concepts. The overall aim of this project is to differentiate between forms of concept learning and their constituent subprocesses through the identification of participating neural systems using functional Magnetic Resonance Imaging (fMRI). The first aim is to dissociate the neural substrates of rule learning (classifying on the basis of a verbalizable rule, such as "always choose the blue stimulus") from exemplar learning (classifying on the basis of similarity to previously learned stimuli), and to compare the brain activity of learners with persons who fail to learn. The second aim is to dissociate the neural systems underlying rule formation from rule application. Rule learning is dependent on executive functions subserved by the frontal lobes, and thus these studies have the potential to elucidate the pattern of deficits seen in patients with frontal lobe damage due to stroke or traumatic brain injury. The third aim is to related recruitment of striatal brain structures in concept learning to the degree that a probabilistic relationship exists between a stimulus and its concept membership. Because the striatum is damaged in disorders such as Parkinson's and Huntington's diseases, this study may provide insight into the types of learning problems seen in patients with these diseases.

[unreadable] DESCRIPTION (provided by applicant): Category learning is learning to classify stimuli into different groups, or categories. Corticostriatal networks connecting the striatum, including the caudate and putamen, with the cerebral cortex play an important role in category learning. The overall aim of this project is to differentiate between the functions that three of these networks, one linking the head of the caudate with frontal cortex, and one linking the body/tail of the caudate to visual cortex, and one linking the putamen with sensorimotor cortex, serve in classification learning using functional Magnetic Resonance Imaging (fMRI). Previous studies have linked the head of the caudate to processing feedback (i.e., being told that a classification response is correct or incorrect). The body and tail of the caudate, along with the putamen, has been linked to learning and executing associations between stimuli and categories. The first aim is to investigate the sensitivity of the head of the caudate to both verbal feedback and monetary reward, and compare how positive and negative associations with stimuli are represented. The second aim is to investigate how the body and tail of the caudate interacts with visual cortex during learning. The third aim is to distinguish between the roles of the body and tail of the caudate and putamen in categorization. The fourth aim is to separate the contributions of the striatum to categorization from those of the medial temporal lobe. The striatum is affected in many disorders, including Parkinson's disease, Huntington's disease, schizophrenia, and Tourette syndrome. Behavioral studies have shown that category learning is impaired in all of these disorders. The proposed studies may provide insight into the types of learning problems seen in patients with these diseases. [unreadable] [unreadable] [unreadable]