The Role of Feedback to Facilitate Category Learning

描述（由申请人提供）：类别学习是指导日常感知、行动和决策的必要条件。类别学习和分类过程是快速的，在健康成人中可以很容易地进行（Homa & Cultice, 1984）。类别学习的一种亚型——原型扭曲，在外显记忆严重受损的人群（例如，失忆症患者，Reber等人，1998年）和内隐记忆受损的患者（例如，帕金森病，Knowlton等人，1996年）中是完整的。然而，我们实验室最近的发现[Little et al.， 2004]表明，在原型扭曲学习过程中，与学习相关的大脑激活变化的神经元基础暗示着一个比以前报道的更广泛的网络[Aizenstein et al., 2000；Reber et al., 1998]。此外，不仅学习相关的变化跨分布式网络，而且这些区域内的激活取决于学习的阶段和成功[Little & Thulbom，正在审查中]。这导致了一系列关于疾病影响的问题，以及在其他行为衰退的情况下，网络的哪些组成部分被保留以允许完整的学习。对这些问题的回答可能会让我们深入了解补偿病态过程的网络的可塑性。我们的长期目标是定义和阐明在神经系统疾病存在时允许成功原型学习的认知过程（Keri et al., 2001; Squire & Knowlton, 1995; Knowlton et al., 1996）。了解在存在严重记忆缺陷的情况下允许成功学习的冗余过程，可以更好地定义和指导补救和康复方案。我们的主要目标是定义反馈在成功的原型扭曲学习中的作用，以及反过来那些损害学习速度的条件。为了做到这一点，我们将定义反馈在原型扭曲学习中的作用，同时操纵任务难度（类别数量，类别之间的物理相似性）并保持反馈时间不变。其次，我们的目标是在保持任务难度不变的情况下，定义有用反馈的时间边界。最后，我们建议收集试点数据功能MRI数据，在原型扭曲学习响应反馈的条件下，前额叶皮层受到影响。

DESCRIPTION (provided by applicant): Category learning is required for guiding everyday perception, action, and decision making. Category learning and the process of categorization is rapid and can be carried out with apparent ease in healthy adults (Homa & Cultice, 1984). One subtype of category learning, prototype-distortion, is intact in populations with severely impaired explicit memory (e.g., amnesiacs: Reber et al., 1998) as well as in patients with impaired implicit memory (e.g., Parkinson's disease: Knowlton et al., 1996). However, recent findings from our laboratory [Little et al., 2004] demonstrate that the neuronal underpinnings for learning related changes in brain activation during prototype-distortion learning implicate a much broader network than previously reported [Aizenstein et al., 2000; Reber et al., 1998]. Further, not only are the learning related changes across a distributed network but the activation within these regions is dependent upon the stage and success of learning [Little & Thulbom, under review]. This leads to a series of questions about the effects of disease and what components of the network are spared to allow intact learning in the presence of other behavioral decrements. Answers to such questions may provide insight into the plasticity of the network that compensate for diseased process. Our long range goal is to define and elucidate the cognitive processes that allow for successful prototype learning in the presence of neurological disease (Keri et al., 2001; Squire & Knowlton, 1995; Knowlton et al., 1996). Knowledge of the redundant processes that allow successful learning in the presence of severe memory deficits may allow better defined and directed programs for remediation and rehabilitation. Our primary objective is to define the role of feedback in successful prototype-distortion learning and conversely those conditions that impair the rate of learning. To do this we will define the role of feedback during prototype-distortion learning while manipulating task difficulty (number of categories, physical similarity between categories) and holding the timing of feedback constant. Secondly, we aim to define the temporal boundaries for the presentation of useful feedback while holding task difficulty constant. Finally, we propose to collect pilot data functional MRI data, on the conditions under which prefrontal cortex is implicated during prototype-distortion learning in response to feedback.