How we acquire and use visual information to guide movement

大脑必须决定身体在各种情况下的运动地点、时间和方式。虽然计算机现在可以掌握国际象棋和围棋等棋类游戏，但运动的更大复杂性解释了为什么两足机器人仍然无法与人类行走相提并论。正是这种复杂性挑战了大脑的运动决策能力。想象一下，在拥挤的购物中心穿行，或者在森林里徒步旅行。在这里，每一刻要决定往哪里走，走哪条路是至关重要的。这个研究项目的长期目标是了解大脑如何获取和使用视觉信息来做出关于运动的即时决定。大脑可以将眼睛移动到环境中感兴趣的点上，以收集有用的视觉信息。这些目光转移的位置和时间与正在进行的运动行为密切相关，因此可以深入了解大脑如何做出决定。什么因素影响我们分配凝视的地点、时间和时间？凝视在确保准确的脚位、避开障碍物以及选择安全到达目的地的路径方面发挥了怎样的作用？为了解决这些问题，我们将确定完成当前行为所需信息的感知价值，由混乱，不断变化和不可预测的世界性质所产生的环境不确定性，先前的经验以及与行动相关的成本如何影响大脑分配凝视的方式。我们将在实验中使用常见的行走任务，包括必须在不同的地形上行走或绕过障碍物。目标导向、视觉引导的行走是研究决策的绝佳模型，因为这是一种自然行为，涉及从一个地方到另一个地方的一系列决策。重要的是，实验将首先在实验室中进行（并模拟现实环境），然后移到实验室外更自然的环境中。我们将使用最先进的移动眼动追踪器来量化凝视行为，并使用高科技动作捕捉设备来量化运动。这一研究项目将推进关于大脑如何决定往哪里看和往哪里走的基础知识。这项工作还将有助于改进现有的或开发新的凝视行为和运动控制的计算模型。洞察引导凝视的因素对视觉引导机器人的设计、移动辅助设备（如轮椅等凝视控制设备）和营销（包括商店货架上的产品植入）都有影响。此外，这将促进注视训练干预措施的发展，教人们在执行不同动作时何时何地注视。这些设备可用于改善眼疾患者的活动能力，训练运动员，以及更好地培训从事各种工作的工人。

The brain must decide where, when, and how the body is to move under a variety of situations. Although computers can now master board games like chess and Go, the greater complexity of movement explains why bipedal robots still do not match human walking. It is this complexity that challenges the brain's motor decision ability. Imagine navigating a crowded shopping mall or hiking in the forest. Here, the decisions of where to step and which path to walk at each moment are critical. The long-term goal of this program of research is to understand how the brain acquires and uses visual information to make moment-to-moment decisions about movement. The brain can move the eyes to points of interest in the environment to gather useful visual information. The location and timing of these gaze shifts are intimately linked with ongoing motor behaviour and thus gives insight into how the brain makes decisions. What factors influence where, when, and for how long we allocate gaze? How does gaze play a role in ensuring accurate foot placement, that obstacles are avoided, and in selecting the path to take to safely reach our intended destination? To address these questions, we will determine how the perceived value of the information needed to complete the current behaviour, the environmental uncertainty created by the cluttered, ever-changing, and unpredictable nature of the world, prior experience, and the cost associated with an action affect the way the brain allocates gaze. We will use common walking tasks in our experiments, which include having to step on different terrain or walk around obstacles. Goal-directed, visually guided walking is an excellent model to study decision-making, because it is a natural behaviour that involves a sequence of decisions to get from one place to another. Importantly, experiments will first occur in the lab (and simulate realistic environments) and then move outside of the lab to more natural settings. We will use a state-of-the-art mobile eye tracker to quantify gaze behaviour and high-tech motion capture equipment to quantify movement. This program of research will advance fundamental knowledge on how the brain decides where to look and where to walk. This work will also help to refine existing, or develop new, computational models of gaze behaviour and motor control. Insight into the factors guiding gaze has implications for the design of visually guided robots, mobility aids (e.g., gaze-controlled devices like wheelchairs), and marketing (including product placement on store shelves). In addition, this will facilitate the development of gaze training interventions that teach people where and when to look when performing different actions. These can be used to improve the mobility of people with eye disease, to teach athletes, and to better train workers in a variety of jobs.