Attention and sensory integration in active vision of moving objects

为了探索动态的视觉场景，人类--和所有其他灵长类动物一样--结合了快速的视线转移(扫视)和对运动物体的平稳连续跟踪(追逐)。当物体在扫视过程中继续移动时，将目光转移到移动的物体会给视觉系统带来挑战。为了将眼睛对准目标，在准备扫视时需要考虑这一运动。此外，在每次扫视过程中，场景的其他相关部分(本身可能在移动)会从根本上改变它们在眼球后部的位置。事实上，参与处理场景这些部分的许多视觉大脑区域继承了这种视网膜组织。那么，我们如何跟踪移动物体的位置，并--面对眼球运动--持续关注场景中当前与行为相关的部分？在这个项目中，我们将研究如何估计眼球运动目标的运动，以拦截它们或跟踪它们的位置。我们将使用凝视跟踪、视觉心理物理学和计算建模的组合来实现三个目标。(1)我们将建立过程的动力学，使视觉系统能够在运动准备期间估计和纳入移动对象的速度。(2)我们将致力于更好地了解视觉注意在这一过程中的作用。(3)我们将揭示预测过程是否以及如何更新视觉注意，以跟踪与行为相关的、跨眼睛运动的移动对象。最终目标是在功能水平上理解主动视觉处理，并发展一种连贯的理论，即当我们在一个动态的世界中移动时，感知和行动的连续性。

To explore dynamic visual scenes, humans--as all other primates--combine fast gaze shifts (saccades) and smooth continuous tracking (pursuit) of objects in motion. Shifting gaze to a moving object poses challenges to the visual system, as the object continues to move during the saccade. To land the eyes on target, this motion needs to be taken into account during the preparation of the saccade. Moreover, during each saccade, other relevant parts of the scene (that may themselves be moving) radically change their position on the back of the eye. Indeed, many visual brain areas, that are involved in processing these parts of the scene, inherit this retinotopic organization. So how do we keep track of the locations of moving objects and--in the face of eye movements--continuously pay attention to those parts of the scene that are currently relevant for behavior?In this project, we will investigate how we estimate the motion of the targets of eye movements, to intercept them or keep track of their locations. We will use a combination of gaze tracking, visual psychophysics, and computational modeling to achieve three goals. (1) We will establish the dynamics of the processes that enable the visual system to estimate and incorporate the velocity of moving objects during movement preparation. (2) We will aim at a better understanding of the role of visual attention in this process. (3) We shall uncover if and how predictive processes update visual attention to track behaviorally relevant, moving objects across eye movements. The ultimate goal is to understand active visual processing at a functional level and to develop a coherent theory of the continuity of perception and action as we move through a dynamic world.