Function of Fixational Instability During Natural Viewing

项目摘要 人的眼睛永远不会休息。凝视重定向通常发生2-3次/秒，分隔小的， 不断的眼球运动。乍一看，眼球运动的功能似乎很明显：它们是必要的 以将感兴趣的对象带到并保持在视网膜的最高敏锐度区域（foveola）内。然而，在这方面， 一个压倒性的证据，部分来自我们国家卫生研究院资助的研究，表明这一观点是 简单化，并且通过将空间场景重新格式化为视网膜上的时空刺激， 提供基本的视觉功能，而不仅仅是定位小凹。在这里，我们测试几个新的假设 关于不太明显但同样重要的作用，三种主要的眼睛运动：扫视，追求， fixational drift（在fixation期间持续发生的眼抖动）。研究策略包括评估 眼球运动对视网膜输入的影响，以及由此产生的神经编码、感知和 控制这些实验依赖于最先进的人眼运动的高分辨率测量， 视在视网膜刺激的控制所有的实验都得到了视觉的数学模型的支持。 输入信号及其编码结果的神经建模。目的1关注生理变化 扫视和注视漂移之间的关系刻板印象，扫视引起的瞬变之后是刻板印象，但 视网膜图像的明显的类似布朗抖动的周期。我们的初步分析显示， 将自然场景的能量重新打包成不同的时空格式，循环变化的放大 和视网膜神经元的时间敏感带宽内的光谱分布。预测结果 是眼动肌驱动的动态视觉敏感性，歧视，和形式的知觉在自然后， 我们将量化和测试的扫视数字。目标2专注于扫视本身。我们预测， 扫视的结果是，视觉输入的频谱密度在瞬间有效地驱动视网膜神经元。 起始被均衡到取决于扫视幅度的截止空间频率。这种效果进一步 限制了视觉动力学，并意味着在早期fifth视觉编码取决于振幅的 前眼跳。它还表明，视觉系统可以根据任务利用这种调整。我们将 测试这些预测隔离的贡献眼跳瞬变在各种低和高层次的视觉 任务目标3进一步概括了这些想法。在我们的建模工作的基础上，它研究了固定漂移是否 也可以调整，以调整视觉灵敏度的任务要求，以及是否追求之间的交替 运动和“追赶”眼跳在视觉跟踪中起着类似于眼跳/漂移周期的作用 静态目标。我们所有的假设都得到了初步数据的支持。据我们所知，他们完全是 新的，其中任何一个的确认将有广泛的影响，了解设计原则， 视觉系统，可能的眼部神经眼科疾病的贡献，和发展， 康复策略和假肢。

PROJECT SUMMARY Human eyes are never at rest. Gaze redirections normally occur 2-3 times/second, separating periods of small, incessant eye movements. At first glance, the function of eye movements seems obvious: they are necessary to bring and maintain the object of interest within the foveola, the highest acuity region of the retina. However, an overwhelming body of evidence, in part coming from our NIH-funded research, indicates that this view is simplistic and that, by reformatting a spatial scene into a spatiotemporal stimulus on the retina, eye movements serve fundamental visual functions beyond just orienting the foveola. Here we test several new hypotheses concerning less-obvious but equally critical roles for three main kinds of eye movements: saccades, pursuit, and fixational drift (the eye jitter that continually occurs during fixation). The research strategy consists of evaluating the effect of eye movements on the retinal input and the resulting consequences for neural coding, perception, and control. The experiments rely on state-of-the-art high-resolution measurements of human eye movements and gaze-contingent control of retinal stimulation. All experiments are supported by mathematical modeling of visual input signals and neural modeling of their encoding consequences. Aim 1 focuses on the physiological alternation between saccades and fixational drift. Stereotyped, saccade-induced transients are followed by stereotyped, but distinct, periods of Brownian-like jitter of the retinal image. Our preliminary analyses show that this alternation repackages the energy of natural scenes into different spatiotemporal formats, cyclically varying amplification and spectral distribution within the temporal sensitivity bandwidth of retinal neurons. The predicted outcomes are oculomotor-driven dynamics of visual sensitivity, discrimination, and form perception during natural post- saccadic fixation, which we will quantify and test. Aim 2 focuses on the saccades themselves. We predict that, as a consequence of a saccade, the spectral density of the visual input effective in driving retinal neurons at fixation onset is equalized up to a cut-off spatial frequency that depends on the saccade amplitude. This effect further constrains visual dynamics and implies that visual coding during early fixation depends on the amplitude of the preceding saccade. It also suggests that the visual system can exploit this tuning according to the task. We will test these predictions by isolating the contributions of saccade transients in a variety of low- and high-level visual tasks. Aim 3 further generalizes these ideas. Building on our modeling work, it examines whether fixational drift can also be adjusted to tune visual sensitivity to the task demands, and whether the alternation between pursuit movements and “catch-up” saccades during visual tracking plays a role similar to that of saccade/drift cycle for static targets. All our hypotheses are supported by preliminary data. They are, to our knowledge, entirely novel, and confirmation of any of them will have broad implications for understanding the design principles of the visual system, possible oculomotor contributions to neuro-ophthalmologic disorders, and the development of rehabilitative strategies and prostheses.