A-Level计算机考点：眼动追踪原理与应用 | A-Level Computer Science: Eye Tracking Principles and Applications

A-Level计算机考点：眼动追踪原理与应用

眼动追踪（Eye Tracking）是A-Level计算机科学人机交互（HCI）模块的重要考点，在心理学研究方法中同样高频出现。它通过测量眼睛的注视位置与运动轨迹，帮助研究者判断用户在看什么、按什么顺序看、每处停留多久。很多同学觉得这个考点零散，其实只要抓住"原理、指标、应用、评价"四条主线就能全部串起来。本文逐条拆解，并在最后给出考试答题模板。

Eye tracking is an important topic in the human-computer interaction (HCI) module of A-Level Computer Science and a frequent question in psychology research methods. By measuring the position and movement of the gaze, it helps researchers determine what a user looks at, in what order, and for how long. Many students find this topic scattered and hard to revise, but the whole area can be organised around four threads: principles, metrics, applications and evaluation. This article unpacks each thread and finishes with an exam answer template.

一、什么是眼动追踪？| What Is Eye Tracking?

眼动追踪技术记录眼睛注视点的位置和移动过程，用来推断人的注意力分配与认知加工。它的基本假设是"所见即所关注"：眼睛注视的地方，通常就是大脑正在处理的信息。一套系统通常由眼动仪硬件和数据分析软件组成，输出注视坐标、注视时长、眼跳路径等原始数据。理解这个定义要注意两点：第一，眼动追踪测的是眼球运动，而不是直接测大脑活动；第二，它输出的是一连串时间序列数据，需要经过统计汇总才有意义。考试中的定义题常要求写出眼动追踪的目的，即测量并记录眼球运动，用于分析注意力分布与认知过程。

Eye tracking is a technology that records the position and movement of the point of gaze, allowing researchers to infer where a person's attention is directed and how information is being processed. Its core assumption is that what you look at is what you attend to: the eyes usually fix on the information the brain is currently handling. A typical system combines tracker hardware with analysis software, producing raw data such as gaze coordinates, fixation durations and saccade paths. Two points matter for the definition. First, eye tracking measures eye movements, not brain activity directly. Second, it outputs time-series data that only become meaningful after statistical aggregation. A definition question usually asks you to state the purpose of eye tracking, which is to measure and record eye movements in order to analyse attention distribution and cognitive processing.

二、眼睛运动的三种基本类型 | Three Basic Types of Eye Movement

眼球运动大致分为三类，这是选择题的高频考点。第一类是注视（fixation），指眼睛在某个位置停留约200到300毫秒，期间大脑持续读取信息，注视时长可以反映加工难度。第二类是眼跳（saccade），指注视点快速跳跃到下一个位置，速度极快，但跳跃过程中几乎不获取新信息，所以眼跳本身代表搜索或转移。第三类是平滑追踪（smooth pursuit），指眼睛平滑地跟随一个移动目标，比如追踪飞行的球。三者的区别要会对比：注视是"停下来读"，眼跳是"跳过去找"，平滑追踪是"跟着看"。考试中还可能出现"回视"（regression），即视线回到已读区域，通常说明内容有歧义或需要复核。

Eye movements fall into three broad categories, and this distinction is a frequent multiple-choice point. The first is fixation, where the eye stays on a point for roughly 200 to 300 milliseconds while the brain reads the information; fixation duration can reflect processing difficulty. The second is saccade, a rapid jump from one fixation point to the next; the movement is extremely fast, but almost no new information is acquired during the jump, so a saccade represents searching or shifting attention. The third is smooth pursuit, where the eye follows a moving target smoothly, for example tracking a ball in flight. You should be able to contrast the three: fixation means stopping to read, saccade means jumping to search, and smooth pursuit means following along. Exams may also mention regression, where the gaze returns to previously read regions, which usually signals ambiguity or the need to re-check.

三、主流眼动仪的测量原理 | How Mainstream Eye Trackers Work

目前最主流的是基于视频的眼动仪（video-based eye tracker）。它用红外光源照射眼睛，摄像头以每秒数十到数百帧的速度拍摄，再通过瞳孔中心与角膜反射点（Purkinje图像）之间的相对位置计算出注视方向，这种方法称为瞳孔-角膜反射法（pupil-corneal reflection）。它的优点是精度高、非接触、对被试干扰小，缺点是设备昂贵、对头部位置敏感。另一种是眼电图法（EOG），把电极贴在眼周皮肤上，测量角膜与视网膜之间存在的电位差，电位差随眼球转动而改变。EOG设备轻便、成本低，但精度较低，容易受眨眼和面部肌肉活动干扰。考试常要求对比这两种原理，答题时按"原理、精度、成本、干扰"四个维度展开即可拿满步骤分。

The most common device today is the video-based eye tracker. It illuminates the eye with an infrared light source while a camera captures frames at tens or hundreds of frames per second. The gaze direction is computed from the relative position between the pupil centre and the corneal reflection point, known as the pupil-corneal reflection method. Its advantages are high accuracy, non-contact measurement and little disturbance to the participant; its disadvantages are high cost and sensitivity to head position. An alternative is electrooculography, or EOG, which places electrodes on the skin around the eye and measures the potential difference between the cornea and the retina, a voltage that changes as the eye rotates. EOG devices are lightweight and cheap, but less accurate and easily disturbed by blinks and facial muscle activity. Exams often ask you to compare these two principles; answering along four dimensions, principle, accuracy, cost and interference, secures full method marks.

四、眼动数据中的关键指标 | Key Metrics in Eye-Tracking Data

实验输出中常见的指标有五个，每一个都有对应的认知含义。注视时长（fixation duration）反映加工难度，时长越长说明内容越难理解或越吸引人。首次注视时间（time to first fixation）衡量一个元素吸引注意的速度，越短说明越显眼。总注视次数（fixation count）反映兴趣程度或搜索效率，次数过多可能意味着页面难以理解。回视（regression）指视线折回已读区域，通常表示产生歧义或需要核对。扫描路径（scanpath）是注视点的先后连线，展示浏览顺序。答题时必须把指标和含义一一对应，比如"较长的平均注视时长意味着被试对该区域进行了更深的认知加工"，这样写才能得分。

Five metrics appear regularly in eye-tracking output, and each carries a cognitive meaning. Fixation duration reflects processing difficulty: the longer the duration, the harder or more engaging the content. Time to first fixation measures how quickly an element attracts attention; a shorter time means the element is more salient. Fixation count reflects the level of interest or the efficiency of a search; an unusually high count may indicate that the page is hard to understand. Regression refers to the gaze moving back to previously read regions, usually suggesting ambiguity or the need to verify. The scanpath is the ordered sequence of fixations and reveals the browsing order. When answering, you must link each metric to its meaning, for example a longer average fixation duration indicates deeper cognitive processing of that region, and writing the link explicitly earns the marks.

五、热点图与扫描路径：数据如何可视化 | Heatmaps and Scanpaths: How Data Is Visualised

眼动数据汇总后通常用两种图呈现。热点图（heatmap）用颜色深浅表示注视时长或次数的分布：红色代表注视最密集的区域，黄色次之，绿色和蓝色代表低密度区域，一眼就能看出页面上最吸睛的位置。扫描路径图（scanpath）展示单个被试的浏览过程，用圆圈表示注视点，圆圈越大注视越久，用线段连接注视顺序，能清楚看到"先看标题，再看图片，最后看按钮"这样的路径。考试中常给出一张热点图，要求描述注意力分布并解释原因。答题结构建议为：先描述颜色分布特征，再联系界面元素（标题、图片、导航、按钮）说明为什么该区域吸引注意，最后指出对设计改进的启示。

Aggregated eye-tracking data are usually presented in two types of chart. A heatmap uses colour intensity to show the distribution of fixation duration or count: red marks the most densely fixated regions, yellow is intermediate, and green and blue mark low-density areas, so the most eye-catching parts of a page can be seen at a glance. A scanpath shows the browsing process of a single participant: circles represent fixation points, larger circles mean longer fixations, and connecting lines show the order, making it clear whether the user looked at the heading first, then the image, and finally the button. Exams often present a heatmap and ask you to describe the attention distribution and explain it. A good answer structure is: first describe the colour pattern, then relate it to interface elements such as headings, images, navigation and buttons, and finally state what the finding implies for design improvement.

六、眼动追踪在网站可用性测试中的应用 | Eye Tracking in Website Usability Testing

可用性测试（usability testing）是眼动追踪最成熟的应用之一。设计师让目标用户在眼动仪前完成真实任务，比如查找某件商品、填写注册表单或比较两个套餐，同时记录他们的注视数据。大量实验得出几条经典结论：用户浏览网页多呈F形阅读模式，前几行和每行开头获得的注视最多；导航栏位置和命名直接影响查找效率；弹窗广告能吸引短暂注视，但会打断任务、增加回视次数。这些证据帮助设计师决定标题字号、图片位置、按钮颜色和广告投放方式。考试问"眼动追踪如何改进网页设计"时，答题要点是三步：发现问题区域、验证设计假设、用数据支持改版决策，每一步都要结合具体指标举例。

Usability testing is one of the most mature applications of eye tracking. Designers ask target users to complete realistic tasks in front of the tracker, such as finding a product, filling in a registration form or comparing two packages, while their gaze data are recorded. Classic findings from many experiments include: users read web pages in an F-shaped pattern, with the first lines and the beginning of each line receiving the most fixations; the position and wording of the navigation bar directly affect search efficiency; and pop-up adverts attract brief attention but interrupt the task and increase regressions. This evidence helps designers decide heading sizes, image positions, button colours and advertisement placement. When asked how eye tracking improves web design, structure the answer in three steps: identify problem areas, validate design assumptions, and support redesign decisions with data, giving a concrete metric as an example at every step.

七、视线交互与无障碍设计 | Gaze Interaction and Accessibility Design

眼动追踪不仅能"看"，还能作为输入手段，这就是视线交互（gaze interaction）。用户通过注视屏幕上的虚拟键盘或按钮来操作电脑，为渐冻症（ALS）等运动障碍患者提供了重要的沟通通道，让他们能够打字、发邮件甚至上网。常用的触发机制是"注视停留"（dwell）：视线在目标上停留超过设定时间（如500毫秒）就视为点击，从而避免误操作。这类系统还广泛用于科研、军事模拟和游戏领域。考试可能会问视线交互的局限，答题要点包括：眼睛无法像手指一样精确点击微小目标；长时间注视会造成视觉疲劳；还有一个经典问题叫"米达斯接触"（Midas touch），即目光扫过即触发，用户无法在不激活功能的情况下自由浏览，因此通常需要配合眨眼或按键确认。

Eye tracking can not only observe but also serve as an input method; this is gaze interaction. Users operate a computer by looking at an on-screen virtual keyboard or buttons, which provides a vital communication channel for people with motor disabilities such as ALS, allowing them to type, send emails and even browse the web. The common trigger mechanism is dwell: if the gaze stays on a target longer than a set time, such as 500 milliseconds, a click is generated, which avoids accidental activations. Such systems are also widely used in research, military simulation and gaming. Exams may ask about the limitations of gaze interaction. Key points include: the eye cannot click tiny targets as precisely as a finger; prolonged staring causes visual fatigue; and there is the classic Midas touch problem, where every glance triggers a command, so users cannot browse freely without activating functions. Gaze input therefore usually needs to be confirmed with a blink or a key press.

八、更多应用领域：驾驶、广告与医疗 | More Applications: Driving, Advertising and Medicine

在汽车行业，眼动仪用于驾驶员注意力监测：系统检测视线长时间偏离道路或眨眼频率异常，判断驾驶员疲劳或分心，并及时发出警报，这是高级驾驶辅助系统（ADAS）的组成部分。在广告与营销领域，企业用热点图评估海报、包装和货架陈列的视觉吸引力，研究"注意力经济"，比如判断顾客在货架上先看哪个品牌。在医疗领域，眼动追踪辅助诊断阅读障碍、自闭症谱系障碍和帕金森病，也被用于手术训练反馈，帮助医生改进操作习惯。在心理学研究中，它被用来探索阅读过程、决策机制和广告记忆。考试选择题常问"下列哪项不属于眼动追踪的应用"，所以要把五个领域记牢：驾驶安全、广告评估、医疗诊断、人机交互、心理学研究。

In the automotive industry, eye trackers are used for driver attention monitoring: the system detects that the gaze has left the road for too long or that the blink rate is abnormal, judges that the driver is tired or distracted, and raises an alarm in time; this is part of advanced driver assistance systems, or ADAS. In advertising and marketing, companies use heatmaps to evaluate the visual appeal of posters, packaging and shelf displays, studying the attention economy, for example which brand a customer looks at first on a shelf. In medicine, eye tracking supports the diagnosis of dyslexia, autism spectrum disorder and Parkinson's disease, and provides feedback in surgical training to help surgeons improve their habits. In psychology research, it is used to explore reading processes, decision making and advertising memory. Multiple-choice questions often ask which option is not an application of eye tracking, so memorise the five areas: driving safety, advertising evaluation, medical diagnosis, human-computer interaction and psychological research.

九、眼动追踪的优点与局限 | Strengths and Limitations of Eye Tracking

评价题是这类考点的压轴题型。优点方面：第一，数据客观，不受被试口头报告偏差的影响，被试无法轻易伪装自己的目光；第二，时间分辨率高，能捕捉毫秒级的注意变化，这是问卷无法做到的；第三，可以研究无法用语言表达的群体，比如婴幼儿和严重运动障碍患者。局限方面：第一，设备昂贵，实验通常在受控实验室进行，生态效度有限；第二，"看"不等于"注意"，存在"视而不见"（looking without seeing）的反例，比如开车时目光在前方却想着别的事；第三，个体差异大，瞳孔大小、眼镜反光都会影响数据质量，需要较大的样本量。答题时按"优点两条、局限两条、各配一个例子"的结构组织，最容易拿高分。

Evaluation questions are the climax of this topic. On the strengths side: first, the data are objective and free from the bias of verbal self-report, because participants cannot easily fake their gaze; second, the temporal resolution is high enough to capture attention changes within milliseconds, which questionnaires cannot achieve; third, it can study groups that cannot express themselves in language, such as infants and people with severe motor disabilities. On the limitations side: first, the equipment is expensive and experiments usually take place in a controlled laboratory, which limits ecological validity; second, looking does not equal attending, and there are counterexamples of looking without seeing, for example a driver whose gaze is on the road while the mind is elsewhere; third, individual differences are large, and pupil size or reflections from glasses can degrade data quality, so larger samples are needed. Organising the answer as two strengths, two limitations, each with an example, is the most reliable route to high marks.

十、A-Level考试常见题型与答题模板 | Common Question Types and Answer Template

这个考点在考试中主要有四类题型。第一类是定义题：写出眼动追踪的目的，并列举两种眼球运动类型。第二类是原理题：解释瞳孔-角膜反射法如何工作，要写出红外光源、摄像头、瞳孔中心与角膜反射点的相对位置这几个关键词。第三类是图表分析题：根据热点图或扫描路径描述注意力分布，先描述再解释再给启示。第四类是评价题：讨论方法在某场景中的优缺点。通用答题模板是"术语定义加展开解释，每个论点配一个例子"。评价题则从客观性、时间精度、成本、生态效度四个维度展开。考前建议完成两套包含HCI或研究方法内容的真题，把标准答案的关键用词背熟，考试时直接套用模板。

Four question types dominate this topic in the exam. The first is definition: state the purpose of eye tracking and name two types of eye movement. The second is principle: explain how the pupil-corneal reflection method works, and make sure to include the infrared light source, the camera and the relative position between the pupil centre and the corneal reflection point. The third is data analysis: describe the attention distribution from a heatmap or scanpath, using the structure of describe, explain and then implication. The fourth is evaluation: discuss the advantages and disadvantages of the method in a given scenario. The general answer template is to state the term, expand the explanation, and support every point with an example. For evaluation questions, expand along four dimensions: objectivity, temporal precision, cost and ecological validity. Before the exam, complete two past papers that cover HCI or research methods and memorise the key wording of the model answers, so you can apply the template directly in the real exam.

十一、考点串联：原理、指标与应用一条线 | Connecting the Dots: Principles, Metrics and Applications

最后把整个考点串成一条主线：硬件测量（瞳孔-角膜反射法）产生数据（注视、眼跳、回视），数据汇总成可视化（热点图、扫描路径），可视化支撑应用（可用性测试、视线交互、驾驶监测），应用又暴露局限（成本高、生态效度有限、视而不见）。复习时抓住四个关键词：原理、指标、应用、评价。每复习一个应用场景，就自问一遍：它用了哪些指标？有什么优点和局限？这样遇到评价题就不会漏点。本文从定义讲到答题模板，希望帮你把眼动追踪这个考点彻底拿下。如果你想针对薄弱环节进行一对一辅导，欢迎联系文末的老师，免费帮你规划复习路线。

Finally, connect the whole topic into a single line: hardware measurement, the pupil-corneal reflection method, produces data, fixation, saccade and regression; the data are aggregated into visualisations, heatmaps and scanpaths; the visualisations support applications, usability testing, gaze interaction and driver monitoring; and the applications reveal limitations, high cost, limited ecological validity and looking without seeing. When revising, remember four keywords: principles, metrics, applications and evaluation. For every application scenario, ask yourself which metrics it uses and what its strengths and limitations are. This habit prevents you from missing points in evaluation questions. From the definition to the answer template, this article aims to help you master eye tracking completely. If you would like one-to-one tuition on your weak areas, feel free to contact the teacher at the end of the article for a free revision plan.

需要专业A-Level辅导？
16621398022
同微信 | 免费咨询课程规划
访问 aleveler.com 了解更多

更多咨询请联系16621398022（同微信）