The Cognitive Neuroscience of Human Category Learning

分类是对不同类别或类别中的对象或事件做出不同的反应。这种至关重要的技能使人们能够接近食物或朋友，避免毒素或陷阱。最近的证据表明，人类的类别学习是由多个，定性不同的学习系统介导的，现在已经知道了很多关于这些系统的神经生物学基础。大多数证据来自专门设计为只加载在一个系统上的人工任务。因此，几乎没有人知道这些不同的系统如何相互作用，以及它们各自的贡献是如何协调的。这一点尤其重要，因为在大多数现实生活中的类别学习情况下，许多系统似乎都有贡献。真实的生活和大多数实验室研究的类别学习之间的另一个关键区别是，在真实的生活中，人们成为某些类别的专家。专业知识如何发展的问题尤为重要，因为有充分的证据表明，调节新类别学习的神经机制和途径与调节高度学习的类别表征的神经结构不同。这个项目是一个研究项目的延续，该项目为多个系统提供了大量证据，并发现了组成系统的许多独特属性。目前的项目有两个目标。第一个是要了解在不同的分类学习系统中学习是如何协调的，第二个是要了解分类专业知识是如何发展的。这些问题将使用许多方法来解决，包括传统的认知实验，对亨廷顿病患者的研究，功能磁共振成像实验和神经计算模型。这些结果将提高对人类基本技能的理解，更好地了解各种不同神经系统疾病导致的认知变化，并建议改进复杂分类任务的训练程序（例如，教放射科医生在X射线中发现肿瘤）。

To categorize is to respond differently to objects or events in separate classes or categories. This vitally important skill allows people to approach food or friend and to avoid toxin or trap. Recent evidence suggests that human category learning is mediated by multiple, qualitatively distinct learning systems, and much is now known about the neurobiology that underlies these systems. Most of this evidence comes from artificial tasks that were specially designed to load on only one system. As a result, almost nothing is known about how these various systems interact and about how their separate contributions are coordinated. This is especially important because it seems likely that a number of systems would contribute in most real-life category-learning situations. Another critical difference between real life and most laboratory studies of category learning is that in real life people become experts at certain types of categorization. The question of how expertise develops is especially important because there is good evidence that the neural mechanisms and pathways that mediate the learning of new categories are different from the neural structures that mediate the representation of highly learned categories. This project is a continuation of a research program that provided much of the evidence for multiple systems, and that discovered many unique properties of the component systems. The current project has two goals. The first is to understand how learning in the various category-learning systems is coordinated and the second is to understand how categorization expertise develops. These problems will be attacked using a number of approaches, including traditional cognitive experiments, studies with Huntington's disease patients, fMRI experiments, and neuro-computational modeling. The results will improve understanding of a basic human skill, lead to better insights into the cognitive changes that result from a variety of different neurological disorders, and suggest improvements in training procedures for complex categorization tasks (e.g., teaching radiologists to find tumors in x-rays).