Eye Movements and Visual Perception

描述（由申请人提供）：在快速眼球运动过程中，我们不会意识到视网膜图像的运动，也不会在每次眼球运动后感知到世界在不同的地方。这表明，在将视网膜输入转化为视觉感知时，大脑以某种方式纠正了眼睛的运动。该项目结合了猴子的细胞方法、行为研究和人类的功能成像，以研究眼动存在时感知稳定性的神经机制。我们提出并测试了感知稳定性的双重机制理论的具体实施。在这个理论中，一种机制是利用眼睛位置信息将以眼睛为中心的视网膜信息转化为稳定的、以世界为中心的信息。当快速眼球运动正在进行时，这种转换被认为是不完美的。因此，另一种被称为扫视抑制的机制被用来减弱视觉感知，并隐藏在扫视过程中坐标变换的缺陷。我们的假设是，这两种机制都在早期视觉皮层区域实现。正如其他人所指出的那样，在这些区域中存在的眼睛位置信号原则上提供了执行所需坐标变换所需的信息。然而，足够信息的存在并不一定意味着信号实际上用于坐标变换。在我们的第一个具体目标中，我们测试了假设的一个强预测，即眼睛位置信号的错误会导致感知错误。我们将使用单细胞记录来测试在扫视时间前后，真实眼位与早期视觉区域（V1， MT）的眼位信号之间是否存在不匹配。至关重要的是，这种不匹配应该与已知发生在扫视周围的定位感知错误相匹配。我们的第二个目标是测量这些神经元在扫视时视觉反应的变化。我们的假设首先预测，这些变化解释了为什么定位和检测视觉刺激的错误发生在眼球运动的颞部附近。其次，我们将测试神经反应的随机变化是否与动物行为反应的随机变化相关。这些行为和电生理实验将为我们的假设提供有力的证据。这个项目的意义在于，它将增强我们对解决视觉感知基本问题的神经机制的理解。这将有助于制定治疗方案的视觉神经障碍和创伤和疾病后的康复。

DESCRIPTION (provided by applicant): During rapid eye movements, we are not aware of the motion of the retinal image, nor do we perceive the world to be in a different place after every eye movement. This shows that, somehow, the brain corrects for the movement of the eyes when translating the retinal input into a visual percept. This project combines cellular approaches in monkeys, with behavioral studies, and functional imaging in humans to investigate the neural mechanisms underlying perceptual stability in the presence of eye movements. We propose and test a specific implementation of the dual mechanism theory of perceptual stability. In this theory, one mechanism uses eye-position information to transform eye-centered retinal information into stable, world-centered information. This transform is assumed to be imperfect while rapid eye movements are underway. Another mechanism -called saccadic suppression- is therefore invoked to blunt visual perception and hide imperfections in the coordinate transform during saccades. Our hypothesis is that both mechanisms are implemented in early visual cortical areas. As others have pointed out, the presence of eye-position signals in those areas in principle provides the information needed to perform the required coordinate transform. The presence of sufficient information, however, does not necessarily mean that the signal is actually used for the coordinate transform. In our first specific aim we test a strong prediction of the hypothesis, namely that errors in the eye-position signal should cause errors in perception. We will use single cell recordings to test whether, around the time of saccades, there is a mismatch between the true eye-position and the eye-position signals in early visual areas (V1, MT). Crucially, this mismatch should match the perceptual errors in localization that are known to occur around saccades. Our second aim is to measure changes in the visual response of these neurons around the time of a saccade. Our hypothesis predicts first, that these changes explain why errors in localization and detection of visual stimuli occur in the temporal vicinity of eye movements. Second, we will test whether random trial-by-trial variations in the neural response are correlated with trial-by-trial variation in the behavioral response of the animal. Together these behavioral and electrophysiological experiments would provide strong evidence for our hypothesis. The significance of this project is that it will enhance our understanding of the neural mechanisms that solve a fundamental problem in visual perception. This will help to develop treatment programs for neurological disorders of vision and rehabilitation after trauma and disease.