Neural Mechanisms for the Construction of Subjective Value and Preferences

主观价值和偏好构建的神经机制

• 批准号: 7883451
• 负责人: John Clithero
• 金额: $ 2.87万
• 依托单位: DUKE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2009
• 资助国家: 美国
• 起止时间: 2009-05-01 至 2011-04-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7883451
• 关键词: Affect; Animals; Anxiety; Area; Attenuated; Behavior; Behavioral; Brain; Brain region; Clinical; Cognitive; Complex; Corpus striatum structure; Data; Decision Making; Depressed mood; Disease; Electrophysiology (science); Environment; Evaluation; Experimental Designs; Food; Functional Magnetic Resonance Imaging; Gambling; Goals; Human; Image Analysis; Individual; Lead; Learning; Literature; Machine Learning; Mental Depression; Mental disorders; Neurobiology; Neurosciences; Nicotine; Outcome; Pain; Pharmaceutical Preparations; Phase; Physiological; Play; Population; Primates; Process; Proxy; Psychology; Research; Research Proposals; Rewards; Risk; Role; Signal Transduction; Smoker; Sorting - Cell Movement; Specificity; Staging; Structure; Sum; System; Techniques; Testing; Uncertainty; Ventral Striatum; Work; addiction; base; deprivation; improved; neuromechanism; novel; pleasure; preference; relating to nervous system; research study; response; reward processing

DESCRIPTION (provided by applicant): Adaptive decision making relies on the ability to evaluate and integrate potential rewards. Data from primate electrophysiology and human functional magnetic resonance imaging (fMRI) studies implicate key structures including the orbitofrontal cortex (OFC) and striatum in valuation and reward-driven learning. However, how value signals from distinct rewards are integrated into a decision value remains relatively unknown. My research proposal will investigate the neural underpinnings of value computation and construction. In particular, I will conduct several experiments that will lead to a better understanding of these regions' separate contributions and functional interactions. First, I will delineate contributions of the striatum to making valuations involving only positive information, only negative information, or both (Specific Aim 1). Second, I investigate the influence of the striatum on hierarchical value computations, as necessary for complex real-world decisions (Specific Aim 2). Third, I will identify a causal role of the striatum in the manipulation of valuation processes (Specific Aim 3). Despite the large fMRI literature involving the striatum and other reward-processing regions of the brain, these aims have been unaddressed with standard fMRI analyses, which are generally limited in spatial accuracy and ill-suited to make strong conclusions regarding relationships between brain regions. My proposal will employ analytic techniques drawn from the machine- learning and biophysical literatures that are well-equipped to predict individual behavior, identify fine-scale differences in brain computational and cognitive processes, and draw conclusions about causal relationships between brain regions. These analytic techniques, combined with novel experimental designs, will allow me to make a significant contribution to an important area of the neuroscience literature. This proposal focuses on subjective value construction, deficits of which are endemic in clinical disorders. Previous research demonstrates that depressed or drug-deprived individuals differentially process positive and negative material and rewards; the research proposed here will tie those results to value computation and decisions involving risk. Additionally, a better understanding of how individuals make complex choices should prove useful for treatment of mental health disorders like depression or anxiety.

描述（由申请人提供）：自适应决策依赖于评估和整合潜在奖励的能力。来自灵长类动物电生理学和人类功能磁共振成像（fMRI）研究的数据暗示了关键结构，包括轨道额皮层（OFC）和纹状体在估值和奖励驱动的学习中。但是，如何将来自不同奖励的价值信号集成到决策值中仍然相对未知。我的研究建议将研究价值计算和构造的神经基础。特别是，我将进行几项实验，以更好地理解这些区域的单独贡献和功能相互作用。首先，我将描述纹状体对仅涉及正面信息的估值，仅负面信息或两者（特定目标1）的贡献。其次，我研究了纹状体对复杂现实世界决策所必需的层次值计算的影响（特定目标2）。第三，我将确定纹状体在处理估值过程中的因果作用（特定目标3）。尽管涉及大脑的纹状体和其他奖励加工区域的FMRI文献大量文献，但这些目标对标准的fMRI分析尚未得到解决，这通常在空间准确性上受到限制，并且不适合对大脑区域之间的关系得出强烈的结论。我的建议将采用从机器学习和生物物理文献中得出的分析技术，这些文献具有良好的能力来预测个人行为，确定大脑计算和认知过程中的精细尺度差异，并得出有关大脑区域之间因果关系的结论。这些分析技术以及新型的实验设计将使我能够为神经科学文献的重要领域做出重大贡献。该提案的重点是主观价值构建，其缺陷在临床疾病中是地方性的。先前的研究表明，抑郁或剥夺药物的个体会差异化处理正面和负面的物质和奖励。这里提出的研究将把这些结果与涉及风险的计算和决策重视。此外，对个人如何做出复杂选择的更好理解应被证明可用于治疗抑郁症或焦虑等精神疾病。