Mechanisms of Perceptual Learning in Primary Visual Cortex

描述：在不同的大脑区域中实现了许多类型的学习，但不同形式的学习背后的基本机制可能非常相似。学习的最广泛的研究形式之一是视觉系统中的感知学习，它被定义为在感官任务的训练中表现的永久改善。训练提高了感知简单和复杂视觉属性的能力，比如用随机点立体图定义的深度，刺激的弹出，以及通过x光机可视化的炸弹轮廓。研究视觉系统中的感知学习有助于理解学习机制，因为它被认为涉及视觉处理的早期阶段，在这个阶段对单个单元和皮层结构的特性了解最多。此外，知觉学习范式被用于视觉障碍（如弱视）的治疗以及脑损伤后的重组和行为恢复，强调了理解其潜在机制的重要性。了解学习机制的先决条件是表征单个神经元的反应特性及其相互作用在学习过程中如何变化。实现这一点的一个主要障碍是无法在学习过程中长期记录同一个体神经元的情况。现在，我们第一次绕过了这一限制，开发了一种方法，使用长期植入的四极阵列，使我们能够记录连续数天或数周清醒行为的猕猴的相同神经元。我们计划在一个定向辨别任务中研究知觉学习与作为练习的结果在刺激的定向辨别中提高表现的关系。这种模式下的学习是特定于训练过的视网膜位置的，因此被认为反映了初级视觉皮层（V1）等早期视觉区域的可塑性。我们将在猕猴的V1区研究这一过程。首先，我们将测量在没有训练期间的几天和几周内的可变方向调整函数（具体目标1）。确定V1中基线方向图的稳定性是研究学习如何修改它们的重要第一步。此外，一群神经元的活动所包含的信息也取决于这些神经元之间的配对相互作用，以及它们各自的属性。一个很好的类比是，一个团队的表现不仅取决于其个人成员的能力，还取决于玩家之间的互动方式。因此，为了量化学习前后神经回路的信息含量，还必须测量神经元之间的相关性强度（Specific Aim 2）。在Specific Aim 3中，我们将在动物接受定向识别任务训练时，从V1的相同神经元长期记录。这是非常令人兴奋的，因为我们现在第一次能够记录学习过程中的单个神经元，并描述神经回路中相关的变化。学习的最广泛研究形式之一是视觉系统中的感知学习，它被定义为通过大量感官任务的广泛训练而永久提高表现的结果。知觉学习范式被用于弱视等视觉障碍的治疗，使用学习范式在增强脑损伤后的重组和行为恢复方面具有巨大的潜力，这凸显了理解学习机制的重要性。我们将研究知觉学习，同时在学习过程中对相同的神经元进行长期记录。

DESCRIPTION: There are many types of learning realized in different brain areas but the basic mechanisms underlying different forms of learning may be quite similar. One of the most widely studied forms of learning is perceptual learning in the visual system, defined as the permanent improvement in performance as a result of training on a sensory task. Training improves the ability to perceive simple as well as complex visual attributes such as depth defined with a random-dot stereogram, pop-out of a stimulus, and bomb outlines visualized though x-ray machines. Studying perceptual learning in the visual system is advantageous for understanding mechanisms of learning since it is thought to involve early stages of visual processing where the most is known about the properties of single units and cortical architecture. In addition, perceptual learning paradigms are used in the treatment of visual disorders (such as amblyopia) and in the enhancement of reorganization and behavioral recovery after brain injury, underscoring the importance of understanding the underlying mechanisms. A prerequisite for understanding learning mechanisms is to characterize how the response properties of individual neurons as well as their interactions change during the course of learning. A major impediment for realizing this has been the inability to record from the same individual neurons chronically, in vivo, during the course of learning. We now, for the first time, bypassed this limitation by developing a method using chronically implanted tetrode arrays that allows us to record from the same neurons across multiple consecutive days and weeks in awake, behaving macaques. We plan to study perceptual learning in an orientation discrimination task associated with improved performance in the discrimination of the orientation of a stimulus as a result of practice. Learning in this paradigm is specific to the trained retinal position and is therefore thought to reflect plasticity in early visual areas like the primary visual cortex (V1). We will study this process in area V1 of the macaque. First, we will measure how variable orientation tuning functions are across days and weeks during periods of no training (Specific Aim 1). Determining the stability of baseline orientation maps in V1 is an essential first step for studying how learning can modify them. Moreover, the information contained in the activity of a population of neurons also depends on the pair wise interactions between these neurons in addition to their individual properties. A good analogy is the performance of a team which does not only depend on the capabilities of its individual members but also in the way players interact with each other. Therefore, in order to quantify the information content of neural circuits before and after learning it is essential to also measure the strength of correlations between the neurons (Specific Aim 2). In Specific Aim 3, we will record chronically from the same neurons in V1 while animals are being trained in an orientation discrimination task. This is very exciting since for the first time we are now able to record from individual neurons during the course of learning and characterize the associated changes in neural circuits. PUBLIC HEALTH RELEVANCE One of the most widely studied forms of learning is perceptual learning in the visual system, defined as the permanent improvement in performance as a result of extensive training on numerous sensory tasks. Perceptual learning paradigms are used in the treatment of visual disorders such as amblyopia and using learning paradigms has great potential in the enhancement of reorganization and behavioral recovery after brain injury, underscoring the importance of understanding learning mechanisms. We will study perceptual learning while recording from the same neurons chronically during the course of learning.