The Role of Visual Attention in Multiple Object Tracking

许多关键任务，如驾驶和驾驶飞机，依赖于跟踪多个不断变化的位置、速度和轨迹的运动物体。多目标跟踪（MOT）的实验室研究要求观察者跟踪嵌入在一组相同的运动干扰物中的少量运动目标物体。令人惊讶的是，这些研究表明，人们在开始犯错之前只能记住大约四个物体。在忽略干扰物的情况下，人们如何将注意力分散到被跟踪的物体上？为什么分散的注意力仅限于四个物体？研究者建议通过在MOT过程中在目标和干扰物上呈现探针闪烁，并使用基于脑电图的测量方法记录大脑活动来测量视觉注意力来研究这些问题。利用该方法进行的初步研究表明，成功的跟踪性能取决于被跟踪目标的增强和干扰物的抑制。拟议的研究将进一步揭示为什么跟踪仅限于少数对象，以及为什么当注意力超出能力时发生错误。多目标跟踪（MOT）任务捕获了许多关键的现实世界任务的一个重要方面，例如空中交通管制，其中错误可能导致灾难性后果。2009年，空中交通管制员记录了1869次“操作失误”，其中333次属于严重错误。这些数字被认为是实际错误数量的一小部分，因为大多数错误都没有报告。造成这些错误的至少部分原因可能在于跟踪大约四个以上对象的能力有限。有什么干预措施可以提高这种关键能力吗？重要的是，一些研究表明，可以通过训练来增加跟踪对象的数量。例如，玩电子游戏的人的这一比例更高。相反，最近的报告表明，由于精神病理（精神分裂症）和正常衰老，可追踪的物体数量存在实质性缺陷。了解导致这些影响的基本大脑机制是最终能够设计训练方法以提高跟踪能力并减少环境对错误的影响的最佳途径。最后，从基础科学的角度来看，理解MOT有助于我们更广泛地理解视觉注意的机制。目前许多关于注意力的理论都是基于要求观察者短暂地关注一个固定的位置或物体的研究。这些理论在解释长时间对多个移动物体保持注意力的能力方面做得很差，而这种注意力在处理实验室之外的动态视觉世界时可能很常见。

Many critical tasks such as driving and piloting a plane depend on tracking multiple moving objects with continuously changing positions, speeds, and trajectories. Laboratory studies of multiple object tracking (MOT) require observers to keep track of a small number of moving target objects embedded in a set of identical moving distractors. Surprisingly, these studies show that people can only keep track of about four objects before they begin to make errors. How do people divide their attention between tracked objects while ignoring distractors and why is divided attention limited to four objects? The investigator proposes to study these questions by presenting probe flashes on targets and distractors during MOT and measuring visual attention by recording brain activity using EEG-based measures. Preliminary studies using this approach show that successful tracking performance depends on both enhancement of tracked targets and suppression of distractors. The proposed studies should further reveal why tracking is limited to a small number of objects as well as illuminating why errors occur when attentional capacity is exceeded. The multiple object tracking (MOT) task captures an important aspect of many critical, real-world tasks such as air-traffic control where errors can have catastrophic consequences. In 2009, there were 1,869 "operational errors" recorded for air traffic controllers including 333 in the serious category. These numbers are considered to be a small fraction of the actual number of errors as the majority go unreported. At least part of the reason for these errors may lie in the limited ability to track more than about four objects. Are there interventions that might improve this critical ability? Importantly, several studies show that the number of tracked objects can be increased through training. For example, it is higher in people who play video games. Conversely, recent reports indicate substantial deficits in the number of objects than can be tracked due to both psychopathology (schizophrenia) and normal aging. Understanding the basic brain mechanisms responsible for these effects is the best approach to ultimately being able to design training methods for improving tracking ability as well as reducing environmental contributions to errors. Finally, from a basic science perspective, understanding MOT should help us understand the mechanisms of visual attention more broadly. Many current theories of attention are based on studies requiring observers to briefly attend to a single stationary location or object. These theories do a poor job of explaining the ability to sustain attention on multiple moving objects for extended periods of time, and this kind of attention may be commonplace in dealing with the dynamic visual world outside of the laboratory.