Sharon L. Thompson-Schill - US grants

This project will examine the cortical systems that support semantic memory, the human ability to retrieve long-term knowledge in a selective manner. Recent advances in brain imaging have provided a noninvasive method for studying cognitive processes, including semantic memory, in healthy volunteers. These advanced methods, known collectively as functional neuroimaging, have identified prefrontal cortex as the region of the brain that may be specialized for the retrieval of knowledge from semantic memory. This project is based on the hypothesis that prefrontal cortex is involved in the selection of information among competing alternatives and not in retrieval of information from semantic memory, per se. According to this selection hypothesis, which was based on an analysis of the literature and on some preliminary findings, prefrontal cortex acts as a dynamic filter that biases the retrieval of relevant information, when needed, from other parts of the brain. Some of the experiments will test this hypothesis in patients with focal brain damage, in order to allow inferences to be drawn about the necessity of prefrontal cortex for normal retrieval of semantic information. The first goal of this project will be to establish a database of patients who have been identified on the basis of neuroanatomical criteria, in order to characterize brain-behavior relations by studying patients with relatively focal lesions. The database will contain both a precise anatomical description of the lesion, based on an MRI scan, and the results of a comprehensive behavioral neurological mental-status exam designed to assess a variety of cognitive capacities. This important database of focal brain lesions may prove extremely useful in the current and in future neuropsychological studies. The second goal of this project will be to test hypotheses concerning the role of prefrontal cortex in semantic memory. These experiments will examine the effects of competition on the ability of patients with prefrontal damage to retrieve information from semantic memory. In addition, these studies will extend the selection hypothesis to a different kind of memory, short-term, or working memory, in order to examine the homogeneity of prefrontal function (or alternatively, its functional-anatomical specialization) across a wider range of domains. These studies will challenge newly developed claims about the role of prefrontal cortex in semantic memory and will exemplify the way in which functional neuroimaging and more traditional neuropsychological methods can be integrated to provide converging evidence about brain-behavior relationships. This POWRE award will significantly enhance the career development of this promising young investigator at a pivotal career junction, as she makes the transition to a tenure-track faculty member and independent scientist and will contribute to her increased visibility in the research community and to her subsequent academic advancement in her field.

Semantic memory refers to knowledge about objects, facts, concepts, and words and their meanings. Interest in the neural bases of semantic memory comes from observations of patients with striking impairments in semantic memory despite otherwise intact cognitive functioning. Such impairments are observed most commonly in degenerative diseases: semantic memory impairments are one of the earliest and most prominent features of Alzheimer's disease, and loss of semantic knowledge is the defining feature of a syndrome known as "semantic dementia", associated with Pick's disease. Impairments of semantic memory have also been observed in patients with herpes simplex encephalitis, posterior cerebral artery infarctions, hypoxic-ischemic injury, and tumor resection. These observations have led to speculation about the neural bases of semantic memory, aided by the development of modern neuroimaging techniques for studying normal cognition. These two methods (lesion studies and imaging studies) have identified two regions that may be specialized for semantic memory: left temporal cortex and left prefrontal cortex. Based on an analysis of the literature and some preliminary findings, we hypothesize that prefrontal and temporal cortex play quite distinct roles in semantic memory; and that prefrontal cortex is necessary for the selection of competing information, and not for semantic retrieval per se. The proposed project has four major goals: (1) to demonstrate a dissociation between the roles of prefrontal cortex and temporal cortex in semantic memory; (2) to describe in more precise terms, both cognitively and neurally, the nature of semantic repetition effects in temporal cortex; (3) to develop further the hypothesis that prefrontal cortex subserves the selection of competing information and to distinguish this hypothesis from competing alternatives; (4) to provide converging evidence as to the neural bases of semantic memory with complementary imaging and lesion methodologies.

[unreadable] DESCRIPTION (provided by applicant): In the past few decades, cognitive neuroscience has experienced an explosion in research investigating the functions of human prefrontal cortex (PFC). One hypothesis that we have championed is that a function of PFC, perhaps specific to the posterior part of the left, inferior frontal gyrus (LIFG), is the selection of information from competing alternatives. In the current proposal, we aim to explore the relation between this putative selection mechanism and the language deficits that have often been associated with brain damage to this region. For more than a century, it has been known that lesions to the left frontal lobe produce linguistic deficits that are characteristically described as "nonfluent," ranging from no language output to truncated, agrammatical phrases. In recent years, however, it has been possible to move beyond broad anatomical correlations with a clinical description of aphasia syndromes, based in large part on the advent of structural and functional neuroimaging techniques. The experiments in this proposal are designed to take advantage of both of these techniques, in order to better describe structure-function relations in PFC, and to understand the link between these functions and the clinical picture of nonfluent aphasia that is associated with damage to prefrontal regions. The major aims of this line of research are as follows: (1) to further develop a unified theory of the functions of PFC; (2) to relate this theory to the clinical syndromes associated with damage to this region of cortex; (3) to clarify inconsistencies regarding symptomatology and lesion localization of nonfluent aphasias with our theoretical approach. The proposal is divided into four sections. In Section I (Verbal Fluency) we will examine the effects of competition and set size on semantic and phonemic fluency. In Section II (Picture Naming) we will examine competition and context effects on lexical retrieval. In Section III (Lexical Ambiguity) we will examine the inhibition of multiple meanings during the resolution of ambiguous words. In Section IV (Syntactic Processing) we will examine the effects of syntactic complexity and ambiguity on sentence comprehension.

What brain regions participate in the regulation of emotion? How do they interact? How does the ability to think about and describe one's emotional experience or mood in a specific and detailed way affect activity in these regions as well as the relative success of regulating an emotional response or mood? With National Science Foundation funding, Dr. Sharon L. Thompson-Schill will direct graduate student research in a study examining the role of specific brain regions (the amygdala, ventromedial prefrontal cortex, and dorsolateral prefrontal cortex) during the conscious voluntary regulation of emotional responses to both neutral and unpleasant, negative pictures in healthy human participants. Functional magnetic resonance imaging (fMRI) provides a measure of neural activity. Galvanic skin response (GSR), a peripheral psychophysiological measure, and self-reported measures of current emotional state are collected during fMRI scanning to confirm and measure the degree to which the emotion regulation manipulation is successful. Relations between neural activity and the physiological and self-report measures are tested. Emotion differentiation, or the degree to which emotional experience is parsed in a discrete, differentiated fashion, is assessed using a 14-day diary protocol with handheld computers in study participants after completion of the fMRI procedure. Greater emotion differentiation is hypothesized to provide additional emotion knowledge pertaining to the cause, context, bodily sensations, appropriate expressions, and action sequences associated with enhancing or reducing the emotional experience. Therefore, among individuals who experience negative affect at a similar intensity, those who differentiate their negative emotional experience into specific discrete emotions such as "fear" or "anger" are hypothesized to be better able to regulate their negative affect than those who simply describe a "bad feeling." Analysis of the relation between individual differences in emotion differentiation and ability to regulate emotion as indexed by the self-report, GSR, and fMRI data tests whether knowing exactly how one feels is associated with greater ability to modulate that feeling in a goal-appropriate manner. This study builds upon both (a) the investigators' previous finding that maintenance of an emotional response to unpleasant pictures is associated with a prolonged increase in amygdalar activity and (b) reported changes in amygdalar and prefrontal cortical activity during suppression of negative emotion, because both enhancement and suppression of negative emotion are examined in the same participants and is compared to a condition when no regulation is performed. Thus, brain regions displaying activity changes related to affective state changes can be dissociated from those displaying changes related to attempt to modulate affective state, providing information pertaining to the function of the circuitry components.

Broader Impacts. This study tests whether greater differentiation of one's emotional experience on a daily basis is associated with greater ability to regulate emotional responses in a laboratory setting. Development of a model of the neurocircuitry of emotion regulation is critical for understanding normal emotional processes such as how the brain perceives and responds to affective stimuli and how it recognizes the actual experience of the emotion, as well as for developing appropriate treatments for psychological disorders such as depression and anxiety that are characterized by abnormal regulatory processes. In addition, it may also increase understanding of how the brain performs other self-regulatory processes. This project provides valuable graduate student training in the acquisition and analysis of self-report, psychophysiological, and neuroimaging data through its integration of both cognitive neuroscience and social psychological methods.

[unreadable] DESCRIPTION (provided by applicant): Investigations of the formation, representation, and organization of concepts have been central to the field of cognitive psychology over the past several decades. One dominant theory of semantic memory (Allport, 1985) describes object concepts as distributed mental representations implemented in functionally and physically distinct attribute-domains that correspond to different sensory or motor domains of which they are also a part. In fact, according to Allport, these modules are the very same areas of the brain that are dedicated to processing sensorimotor information. One such attribute-domain that has been discussed extensively in the cognitive and cognitive neuroscience literatures is the domain of visual knowledge. This attribute-domain is the topic of investigation in the current proposal. Specifically, we aim (1) to characterize visual knowledge attribute-domains and, in particular, to examine subdivisions in conceptual representations about object appearance that parallel subdivisions in visual perception; (2) to explore variation in visual knowledge retrieval across different concepts and categories; (3) to investigate the link between visual knowledge and visual perception; and, (4) to relate individual differences in visual knowledge attribute-domains to variation in behavior and experience. The proposed experiments make use of a variety of methodologies, including behavioral experiments in normal subjects, congenitally blind adults, and brain-damaged patients, and neuroimaging experiments in both normal and congenitally blind subjects. [unreadable] [unreadable]

Assumptions about learning styles and cognitive abilities inform the nation's STEM educational curricula and standards. The proposal is aimed at addressing fundamental unanswered questions about the construct, including the neural correlates of individual differences in visual, verbal, and spatial cognitive styles. The PI will examine whether cognitive styles reflect learning and reasoning strategies rather than abilities, which has important implications for the way in which individuals? needs are met in classrooms and other learning environments. Through a series of experiments, the PI will study the behavioral and neural correlates of cognitive styles as well as its effects on memory and attention. These experiments will address such specific questions as: Do cognitive styles represent orthogonal dimensions rather than opposing ends of a spectrum? What neural mechanisms underlie these styles? Does general fluid intelligence relate to an ability to switch flexibly between styles as dictated by task context? Will those who have a propensity towards a particular cognitive style show better memory retention for information presented in their preferred modality? Is successful encoding correlated with activity in predicted brain regions? At the time of encoding, does one tend to convert information presented in a non-preferred modality into a preferred modality? The project will culminate in series of experimental interventions design to study the effectiveness of training participants to adopt a new cognitive style and apply it to an untrained task. The promise of this project is that it will help synthesize and make more coherent current theories and findings on cognitive and learning styles and, supported by solid scientific methodology, lay the foundation for improved pedagogical techniques.