当人们阅读网页上的信息,他们不会统一分发注意:一些内容超过其他内容。实际上,如果你曾经跑过或看results from an eyetracking study那you will see that certain areas of the screen tend to receive more gazes than others.

Is it good or bad if people glance at a piece of content more than at others? Are they confused by it or are they engaged with it? In eyetracking research, we can tell the difference. We observe that people spend more time on a certain page segment in three circumstances:

1.详尽审查。During exhaustive reviewthe eyes are drawn to the same area of the page repeatedly. In other words, the person looks at an area, looks elsewhere (maybe on the same page or even on another page), and then returns to looking at the original area. In more extreme cases this back-and-forth is repeated multiple times.

在我们的2009年“eyetracking.Web Usability” book we described exhaustive review as “unconstructive combing of pages and menus.” Exhaustive review happens when people are confused because the page content or the page UI violates users’ expectations. Only when users think that the site is the best place to do their task do we see exhaustive review; otherwise, if they have doubts about the site and feel confused, they simply leave.

In these situations, users’ confusion reflects an “I can’t believe it’s not there” feeling. People are so convinced that the content should be in an area, that they return to it multiple times because they believe they must have made and error and missed it.

2.期望的探索。Sometimes people look at the same content multiple times not because they are confused, but because they enjoy it or it has many nuances that are relevant to them. Desired exploration occurs when users refer to the same piece of content multiple times because they are interested and highly engaged with it

For example, imagine a person using a website to change the filter in her lawnmower. She will move attention back and forth between the physical lawnmower and the instructions containing a diagram of the equipment, to make sure that she is on the right track as she is doing the task. Looking at the instructions multiple times is necessary to complete the task.

Or, imagine a traveler who is going to Hungary for the first time and planning to sightsee. He finds a filmstrip of images of Budapest and scrolls through them slowly. Then he revisits some of the images again, because they intrigue him, and he wants to absorb them better. He is enthusiastic, not confused.



In this article we focus on exhaustive review.



  1. Like information is expected to appear together。人们assume that similar content and UI elements will be placed together on the page.
  2. 常见的网络设计模式以及大多数其他网站的设计设置用户期望对于UI元素和内容将出现的位置。


我记得目睹了2005年的第一个深刻审查的例子。用户正试图找到乔治布雷特的哪项运动和位置。(我迷上了1982-1992大约1982-1992的美国联赛Slugger。作为致敬,有时他会在我的身上露面可用性 - 测试任务。)


She first looked in the main content area, a grey rectangle with text, positioned in the upper left, just below the navigation. Here was his full name, that he batted right and threw left, weighed 200 pounds and stood at 6 feet tall. It included important dates, like his birthday, major-league debut, and when he was inducted into the Baseball Hall of Fame. But, this grey rectangle did not include his field position.


BaseBall -Reference.com:凝视情节中的蓝色点显示一个参与者的固定,按时间顺序编号。即使在仔细阅读灰色区域中的所有信息之后,她也回到了几次,寻找她期望在那里找到的答案而不是。


  1. 上下文。用户是否棒球迷,considered that a player’s position would be one of the most basic pieces of information about the player. All basic information about the player was consolidated in that grey rectangle, so the field position should have been there too.
  2. 此外,该矩形在页面内容区域的左上角,通常是一个网页布局中的优先级点人们首先注意到并仔细参加

You can almost hear users saying, “It必须be here,” when they return to the original section during exhaustive review.


参与者正在比较Vespa网站上的2个踏板车模型。他读了所有的Fuel and Speeddetails of each model — fuel consumption, top speed, engine, tank capacity. He then moved to the方面area, scanned, and scrolled down further. The list of links seemed to indicate that he was no longer looking at details about these 2 Vespa models. He scrolled to the end and when he saw the social icons, he immediately scrolled up to the top of the page and said,Hmm, where’s the price?

这re, he tried tapping one of the models, presumably to get more details about it, but nothing happened. He scrolled and scanned the titles of the same sections of the page again, and read the menu below more closely; then scrolled to the end of the page, shook his head “no,” and scrolled to the top of the page yet again. Once more he scanned the details, even though he had already looked at these sections fully, and said,我期待有价格。

这top half of a page (left) and the bottom half of the same page (right) on the mobile version of Vespa.com compares two Vespa models.


这个视频是通过规范记录的ial eyetracking glasses worn by the user. The red circle represents where he is looking. When he shakes his head “no,” this appears as a vigorous panning of the video image in the recording. (In most browsers, hover over the video to display the controls if they're not already visible.)


这re are no true, written web design standards, but there are some common locations and designs for certain elements. Let’s consider one of these now with a little experiment:找到Niemanbetx官方网站手机版lsen Norman Group的全球导航。

你马上看看页面的顶部吗?如果你这样做是因为桌面上的主要导航传统上是一个令人惊讶的是horizontal list of links across the top of the page。On mobile, too, the main navigation usually appears at the top, although it is often collapsed under a按钮(具有图标,文本或两者)。

我们通过200多个用户和眼镜技术进行了该实验的版本。我们测试了40多个桌面网站和asked people to locate various UI elements。We then launched the site and observed where people looked for them. To prevent bias, we gave each user a random subset of tasks, and also randomized the order in which people had to do these tasks.

position of the company logowas one of the elements we studied. A logo is present on most web pages, and is usually located in the upper left corner in desktop designs. Consider 4 of the sites we studied: National Education Association, Dell, the Florida Keys & Key West, and Kohler. The first two of these displayed the logo in the upper left, and the others showed itcentered at the top




On the NEA site, the red heat concentrated on the logo shows that people found the logo without looking in multiple places. 30 users located it within just a fraction of a second.



On the Florida Keys site, several people looked first to the far upper left, at the map icon. Then their eyes moved to the temperature — the closest bit of content, which appeared still in the upper left area of the page. Next, they went back to the map label. That is an example of exhaustive review, because, even though users had already determined that the item in the far upper left was not the logo, when the second-best candidate for the logo was also not the logo, they looked at the first element again.

Next, participants moved their eyes to the small grid of social icons to the right of the temperature, and, when they determined that it wasn’t the logo, some went back to the temperature once more, in the hopes that it was the logo even though they had already seen that it said81度。Finally, participants looked at the centered text,Florida Keys and Key West,并在他们确定他们找到了徽标时停了在那里。


在科勒网站上,参与者倾向于朝向左上角的眼睛KOHLER Worldwide图标和下拉,以及BATHROOMmenu. Some alternated looking at each of these items a few times (also an example of exhaustive review), before eventually moving their gazes to the right and settling on the tagline科勒的大胆外观在中心。检查此元素后,不确定它是否可以被视为“徽标”,一些参与者再次返回左侧的一些项目。




Avoiding exhaustive review means了解用户想要的东西and where they expect to find it. Following generally accepted UX design and research practices is the best way to combat exhaustive review.





You can see exhaustive review in action in眼镜研究。And, if your分析数据或可用性测试show that people spend a long time on your page, it may be because they are engaged with the content or because they are puzzled by it. To help determine if exhaustive review is the culprit, ask yourself these questions:

  • Have we positioned something in a non-standard place?
  • 我们省略了信息用户可能期望找到的信息吗?
  • Have we placed highly-related content in different places?
  • 内容放置是否反映了内容重要性?
  • 用户说(在可用性测试,反馈表格或支持通信中)他们期待或正在寻找他们找不到的东西?

如果答案不确定,请考虑思考 - 大声笑,适度的可用性测试(遥控器或人)或者以自己的方式开始眼镜可用性研究

For more about the behaviors discussed in this article, consult our “eyetracking.Web Usability“书和”人们如何阅读网络:眼镜证据” report.