A Brief History of Eye-Tracking

Today, eye-tracking is used heavily by marketing groups to craft effective designs in advertising, and by usability researchers to define the optimum user experience. This technology is anything but new though. In fact, eye-tracking goes all the way back to the 1800's.

Ever notice how some websites seem to just flow, while others feel harder to navigate? There are many contributing factors in building usable web sites and applications, but one of the most interesting areas of research is eye-tracking.

Today, eye-tracking is used heavily by marketing groups to craft effective designs in advertising and by usability researchers to define the optimum user experience. This technology is anything but new though. In fact, eye-tracking goes all the way back to the 1800s.

Eye-tracking origins

In 1879, Louis Emile Javal noticed that people do not read smoothly across a page, but rather pause on some words while moving quickly through others.

Edmund Huey later built a device that was used to track eye movement in reading, and he published many of his findings in The Psychology and Pedagogy of Reading, still considered a profound work in the study of reading.

Huey’s device was quite intrusive however, and required readers to wear lenses that only had a small opening with a pointer attached to it. The device allowed Huey to observe where a reader was looking while reading, and he was able to study which words a reader would pause on.

Later developments in eye tracking technology arrived when Charles H. Judd developed the eye movement camera, a non-intrusive eye-tracking device that recorded motions of eyes on film allowing detailed study of eye motion. Along with Judd, Guy Thomas Buswell studied and analyzed eye movements in reading within different ages and levels of schooling. Buswell’s studies led to many leaps in the field of education and literacy, and also secured him a seat in the “Reading Hall of Fame.” He died recently at the remarkable age of 103, in 1994.

In 1931, Earl, James, and Carl Taylor created the Ophthalmograph and Metronoscope, devices used to record the movement of eyes while reading and tools that trained people to read more effectively. Essentially, it was understood that reading was not simply a smooth motion over words just as Louis Javal had suggested. Instead, a reader scans several words, pauses a moment to comprehend them, and then scans again (each scan is called a hop, and each pause is a fixation). The Ophthalmograph was used to measure a readers hops and fixations. Efficient readers would have a steady rhythm of hops and stares.

For readers that hadn’t developed an efficient reading style, the Metronoscope was used to establish this steady rhythm: a reader would be shown 1-3 words at a time through this special machine, and as they finish reading a set of words, a new set is shown. As a reader became used to this pattern, the machine would speed up, helping the user read more quickly by eliminating regressions and establishing a steady flow of words.

For many years, eye-tracking was used as a tool in education research and by medical researchers and physicians more than anywhere else because it was so expensive. It wasn’t until later in the century that it became more accessible to a new niche: marketers.

Eye-tracking used for marketers and designers

In the 1980’s, marketing groups really began using eye-tracking to measure the effectiveness of ads in magazines. Eye-tracking was able to determine what parts of a magazine page were seen, which elements of the page were actually read, and how much time was spent on each part. Previously, voice stress analysis and galvanic skin response tests were used to evaluate the effectiveness of ads (both forms of lie detection, that haven’t exactly been proven to work).

Eye-tracking research continued and led to better understanding in how our eyes and minds cooperate in digesting literature, problems, and images. However, it wasn’t really until the late 80’s and early 90’s that eye-tracking began to distinguish important differences in print and screen design.

In 1990, Gallup Applied Science’s eye-tracking system was used on NFL analyst Joe Theismann, and on average fans in the viewing of professional football games to determine what parts of the game the typical watcher missed. These devices filmed the user’s eye, and a computer would track where his eye followed the screen. After watching, a cursor marked on the film where the viewers were looking.

In the late 1990s, organizations including one of the world’s largest advertising and marketing agency networks EURO RSCG began using eye-tracking technology to measure and study reactions to information on the World Wide Web. For a large number of web designers up until this point, it was assumed that web design should be fashioned off of print and newspaper design.

In 2006, British behavioral consultancy research firm Bunnyfoot researched in-game advertising using eye-tracking and physiological data. The study examined how effective advertising was in video games in virtual worlds with digital billboards. Until recently, this has hardly been considered, but with the high demand of video games in today’s entertainment market, in-game advertising may become more prominent in the future (total sales in 2009 come close to $20 billion, compared to $10 billion in revenue from the box office).

From 2001 till present day, Tobii Technology has been developing eye-tracking technology that both allows disabled users to control devices using only their eyes, as well as helps designers understand how users view websites. One of their most recent product-launches includes the Tobii T60 XL Eye Tracker, the first widescreen eye-tracker. Like many of their other products, the Tobii T60 appears much like a usual computer monitor with built-in sensors that track eye movement and user reactions to different stimuli.

Eye-tracking and you

Today, eye-tracking is widely used in the scientific community, marketing, and in usability studies. This technology has seemingly become far more popular in the past decade than any other time in history, and is heavily used in developing effective advertising campaigns and usable websites. Access to this technology remains far beyond the average Joe however, with eye-tracking hardware often priced in the tens of thousands of dollars.

Jakob Nielsen, a pioneer in web usability, has authored a book Eye Tracking Web Usability which analyzes “…1.5 million instances where users look at Web sites to understand how the human eyes interact with design.” While this may not offer all the advantages of testing a self-developed web application with your own eye-tracking software, it absolutely will help reaffirm and reveal best practices based on many other online examples.

Also, while it does not replace eye-tracking by any means, click heatmaps do offer some useful feedback on websites, and are cheap to deploy and analyze.

There is work being done to reduce the cost of expensive eye-tracking equipment, and hopefully in the near future independent developers and designers will be able to cheaply conduct their own tests. Until then, following the research of others will continue to provide invaluable insight to a century-old area of study.

About the Author

David Leggett

David Leggett is a designer, developer, and builder of things. He currently resides as Director of Marketing and Design at Python Safety.

Related Articles

26 Comments

  • AzterikMedia Reply

    Great article. I had heard of eye-tracking but was not aware of how long it had actually been around. This was an excellent read, thank you!

  • Todd Toler Reply

    Unfortunately this piece doesn’t touch on the epistemological split within the usability community on the validity of eye-tracking. It is hugely controversial. I recommend this great thread from the IxDA discussion boards as further reading on the topic.

  • DesignLovr Reply

    Great and interesting article!
    Lots of stunning facts I didn’t know about yet, especially how long eye-tracking actually dates back.

    I’m pretty convinced that at some point in the future eye-tracking can be achieved on regular PCs with webcams…

    • David Leggett Reply

      Interesting that you say that, there actually have been some efforts using ordinary webcams (the cheap $20-40 kind), and open source software for DIY eye-tracking hardware. Obviously, it isn’t up to par with developers like Tobii, but it’s promising to say the least.

      Ran across this gem in my research: http://hci.stanford.edu/cstr/reports/2006-08.pdf Might be of interest!

    • David Leggett Reply

      Wonderful Robert.

      “Many Users find thinking aloud difficult and it makes them feel uncomfortable.” —Nielsen, 1993

      You make some really great points. Hopefully you’ll have time to add to the post with Kara’s thoughts sometime. We’d also love to have you speak your mind here on the Booth sometime if you’re interested.

      I’ve got some reading material cut out for me regarding PEEP, but could you recommend any other resources for learning more about best practices in Eye-Tracking?

  • Alex Reply

    Actually eye-tracking was developed in the late 1990’s by a company called DSW,(gave us those colorful hazmat suits for Intel) prior to EURO RSCG acquiring them. So they should have gotten credit in this article. I should know I spent many nights in that I-lab.

    • David Leggett Reply

      Very interesting Jacob, I’ve never seen this before. I wonder how accurate a mouse cursor would be (in being aligned with eye placement)in a normal environment. I saw that ClickTale makes reference that the application is about 80% accurate, which I would have never guessed. Anywhere we can find more information on this, and how viable it is?

    • Neil Reply

      The research the clicktale site refers to states;

      “Of the regions that a mouse cursor visited, 84% of them were also visited by an eye gaze. Furthermore, among the regions that the eye gaze didn’t visit, 88% of them were not visited by the mouse cursor, either.”

      It is hardly remarkable to say that users look where they are going to place their cursor. This does not mean however, that users move the cursor round the screen as they look at different parts any more than you need to point at everything you look at.

      Some of the most useful data from eyetracking is from the initial eye movements when the user is deciding what to do before they move the cursor and engage with the page. Click heat maps and tools like click tale can be useful but they aren’t an alternative to eye tracking.

  • BigM75 Reply

    cool article, more pictures please

  • Scott Reply

    Great article! Any plans to follow up with one that discusses findings on eye-tracking for the web/interactive community?

  • Eye test Reply

    Eye tracking has been around for decades but has only become commercially available over the past few years. It’s been heralded as the next big thing in user research, ‘the closest thing to mind reading’!

  • leeander Reply

    Great Article.

    what is missing?
    with a generic title like “A brief history of eye-tracking”, talking about E-T as input device, it is part of the story.

    I don’t know how relevant is E-T in the usability field in term of social impact. Perhaps a good eCommerce could save my life, you know… :)

    meanwhile I am sure IAble is “breaking the silence” supporting the communication problems of people suffering from low movement disfunction, or seriously compromised neuromotor abilities.
    http://www.srlabs.it/en/iable.html

    ok, now you will think that I am trying to get you attention and my comment is an ADV. No guys. Believe me when I say that: designing a GUI based on E-T as input device is an exciting challenge. A new “xerox” park where you “mouse” can’t be visualized on the screen, and point and click have to be re-invented.

    about adv…
    IMHO perhaps “the end” with the last book of our preferred Guru it could be perceived as an adv.

    sorry for a so long comment and thank you for your article. :)

  • Rob Willox Reply

    Neil is correct! ClickTale is no substitute for the real thing just as pepsi is not substitute for coke but it is a very good substitute all the same.

    What ClickTale say is that their mouse movement heatmap correlates reasonable closely enough with controlled eye-tracking studies to offer valuable insights to those site owners not able to justify the cost of full E-T studies.

    In the end it is horses for courses and if ClickTale can provide actionable data to base testable optimisation opportunities then user interaction can be incrementally enhanced and improved with conversions and profitability increased.

  • viona esen Reply

    What is the market size of eye-tracking device?

  • Josh Reply

    I worked in the eye tracking industry for many years. Many of the observations made here are poorly considered.

    First, the notion that “What a person is looking at is what that person is thinking about” is rarely correct, and has been proven so for many environments. There is, in fact, a distinction in the research field between “point of gaze” (POG) and “focus of attention” (FOA); that is: “what a person is looking at” vs. “what a person is focusing his/her visual attention on”. In other words, we do have periphery vision, and we do use it…often.

    With carefully controlled research methods, these two regions of the visual landscape can be teased apart. But too often, poorly trained eye tracking researchers fail to take this into consideration, and wind up with wildly ignorant conclusions.

    Most marketing firms, for example, who have applied our eye tracking equipment do so without any trained scientists (psychology or otherwise) to set up properly controlled experiments. Their results are as meaningless as a monkey’s script on a typewriter.

    Don’t get me wrong, I love eye tracking technology. See this relatively recent (and tangentially related) post for potentially cool new applications:

    http://www.popsci.com/diy/article/2010-04/winscape-makes-world-outside-your-window-whatever-you-want-it-be

    Keep in mind, that eye tracking and head tracking are very closely related–or not at all, for some applications.

    But regardless, people need to stop pretending they know how to do science just because they have a $25,000 eye tracker at their disposal. This is very complicated stuff.

  • Bond!tsch Reply

    good Article and very GOOD Pictures! Pls more

  • drummer Reply

    Great site and very nice info! Thanks!

  • alain Reply

    But regardless, people need to stop pretending they know how to do science just because they have a $25,000 eye tracker at their disposal. This is very complicated stuff.

Leave a Comment on This Article