Mechanisms of Learning a Visual Discrimination

描述(由申请者提供)：通过训练提高视觉任务的成绩。我们的长期目标是了解导致这种形式的知觉学习的神经变化。我们使用的任务需要对随机点刺激中微弱运动信号的方向做出决定。经过训练，猴子和人类一样，学会了更准确、更快速地解读这些运动信号。我们将确定这些性能收益的神经底物。在猴子身上，有几种机制有助于决策的形成。位于中间颞区(MT)的神经元代表用于执行任务的运动信息。当眼球运动指示做出决定时，眼球运动回路似乎会随着时间的推移积累这种运动信息来形成决定。我们的三个具体目标将确定这些机制的变化，因为绩效随着培训的提高而提高。目标1将确定大脑表示感觉刺激的方式的变化。我们将记录MT神经元，以测试它们对运动的敏感性是否随着性能的提高而变化。目标2将确定大脑如何解释运动信息以形成行为反应的变化。我们将通过电刺激额部眼场来唤起眼球跳动。这些被激发的眼跳对发展眼球运动指令很敏感，因此将被用来测试这些指令中所代表的积累过程如何随着训练而变化。目标3将确定大脑计算决策方式的变化。我们将记录来自顶叶外侧区域的神经元，这些神经元代表了受训猴子的决策形成和眼动反应的形成。我们将确定这些不同的神经计算在训练过程中是如何联系在一起的。从长远来看，这些研究将有助于将系统水平的电生理学与可塑性和学习研究联系起来。这些联系将有助于开发计算工具来治疗学习障碍和影响大脑处理和解释信息的能力的临床障碍(例如精神病、痴呆症和失认症)。

DESCRIPTION (provided by applicant): Performance on visual tasks improves with training. Our long-term goal is to understand the neural changes that give rise to this form of perceptual learning. We use a task that requires a decision about the direction of weak motion signals in a random-dot stimulus. With training, monkeys, like humans, learn to interpret these motion signals more accurately and quickly. We will identify the neural substrate of these performance gains. In monkeys, several mechanisms contribute to decision formation. Neurons in the middle temporal area (MT) represent the motion information used to perform the task. When the decision is indicated with an eye movement, oculomotor circuits appear to accumulate this motion information over time to form the decision. Our three Specific Aims will identify changes in these mechanisms as performance improves with training. Aim 1 will identify changes in how the brain represents the sensory stimulus. We will record from MT neurons to test whether their sensitivity to motion changes as performance improves. Aim 2 will identify changes in how the brain interprets motion information to form the behavioral response. We will evoke saccadic eye movements with electrical microstimulation of the frontal eye field. These evoked saccades are sensitive to developing oculomotor commands and thus will be used to test how the accumulation process represented in these commands changes with training. Aim 3 will identify changes in how the brain computes the decision. We will record from neurons in the lateral intraparietal area that represent formation of the decision and formation of the eye-movement response in a trained monkey. We will determine how these different neural computations become linked over the course of training. In the long term, these studies will help connect systems-level electrophysiology with the study of plasticity and learning. These connections will aid in the development of computational tools to treat learning disabilities and clinical disorders (e.g., psychosis, dementia and agnosia) that affect the brain's ability to process and interpret information.