Eye Tracking 101: A User Researcher’s Guide
It’s not just for optometrists anymore.
As part of our UX 101 education series, where we discuss the different types of studies and research methodologies you can use with the UserZoom platform, we’d like to introduce our readers to the exciting world of eye tracking!
Before we jump into the article we want to make it clear that eye tracking is not available as a study or question type on the UserZoom platform. This article is simply about introducing the research methodology.
Okay, let’s jump right in!
What is eye tracking?
Eye tracking measures the location and duration of a user’s gaze on stimuli (web page, computer screen, mobile app, etc.), and can provide valuable insight into users’ viewing behaviors. This can unearth usability problems in which elements are being missed.
More broadly, eye tracking is used by UX researchers in various user research methods such as A/B testing and moderated task-based usability testing to see through a user’s eyes and track/record what viewers are looking at and for how long. It also reveals how eyes move and search for information on the screen, which can help identify whether users are attracted to specific content or features.
We should point out that most eye tracking studies are conducted as moderated studies in an eye tracking lab under controlled light conditions. However, with eye-tracking technology like portable eye trackers it is possible to collect eye movements outside of a lab environment.
What are typical use cases for eye tracking?
Eye tracking has a broad range of applications and has been used by a number of researchers in UX research, psychology, and neuromarketing to study user behavior. Tracking people’s eye movements can help UX researchers understand the factors that may impact the usability of an interface by detecting where users focus their attention and how they look for information on the interface.
Eye tracking can be added to a moderated usability study at any time in the product lifecycle/design cycle to obtain objective feedback from users’ viewing behavior. Anytime you need to see exactly what a user is looking at (or not looking at) in real-time during navigation, eye tracking will help.
How does eye tracking work?
Most commercially available eye tracking systems today capture eye movements using an unobtrusive method known as video-based corneal reflection. Eye trackers include infrared light sources and cameras. The light sources send light to the eyes and the exact location of the eyes are measured from the corneal reflection of the eyes on the screen (Fig.1).
It is for these reasons that eye tracking studies are usually conducted in eye tracking labs with controlled light condition; however, portable eye tracking devices could be used to conduct studies outside of the laboratory environment as well.
Keep in mind that eye trackers must be calibrated for each viewer, therefore, each participant needs to complete the calibration process before the start of the eye-tracking study. During the calibration process the eye tracker learns how certain coordinates on the stimulus (e.g., certain coordinates on a computer screen) correspond to a viewer’s eye position. The calibration process allows the eye tracker to associate the viewer’s pupil data with known locations on the stimulus.
Eye tracking devices collect eye movements with different sampling rates from 30 Hz to 1000 Hz. The higher the sampling frequency of the eye trackers the higher the accuracy of eye tracking. Eye trackers with higher sampling rates (>300) are used for more in-depth psychological eye-tracking studies (e.g., to study eye movements in reading.)
When should you use eye tracking?
Here are essential facts to consider while deciding on whether or not eye tracking is the right approach for your research goals.
- Can get very accurate readings of how users are viewing your stimuli.
- Is more informative than a heatmap from a click test as you can actually measure the amount of time users spend viewing and where they are drawn to first.
- Eye tracking studies can be time intensive, as a typical moderated eye tracking study may take at least 30 to 45 minutes.
- Recruiting participants for an eye-tracking study could be cumbersome and expensive. Some participants do not like the idea of their eyes being tracked.
- Eye trackers are quite sensitive instruments and can have difficulty tracking participants who have bifocal or trifocal glasses. This narrows down the pull of participants especially for studies that focus on older adults’ eye tracking.
What results do you get?
Typical metrics obtained from eye tracking software include duration and location of gaze on the screen (fixations), the trajectory path that eyes travel from one fixation to another (saccades), coordinates of the eyes on the screen (x,y, and z coordinates), and the size of the pupil of the eyes (Fig.2).
The results obtained from an eye-tracking software depends on the eye tracking software though they are typically categorized as the following:
- Raw eye movement metrics: Commonly these are fixation, saccades, pupil dilation, and the coordination of the gaze on the screen.
- Heatmaps: These are visualizations that show the general distribution of gaze points. Heat maps represent where the visitor concentrated their gaze and how long they gazed at a given point. They display viewing patterns as color-coded pictures (See Fig. 3). Generally, a color scale moving from green to red indicates the duration of gaze. Thus, a red spot over an area of your page might indicate that a participant, or group of participants, focused on this part of a page for a longer period of time.
- Gaze Plots or fixation/saccade patterns: These are visualizations that show the location of the fixations made as well as the sequence in which they were made. Gaze plot also allows us to look at the gaze behavior of groups of people in relation to similar stimuli.
Tips for analyzing your results
In an eye-tracking study the size of the exported data for each participant depends on the sampling frequency of the eye tracking device and the duration of the study. To save processing time in large-scale eye tracking studies it is especially important to choose the right metrics that relate to your research questions.
We also wanted to quote two great pieces of advice from William Albert & Thomas Tullis’ book Measuring the User Experience.
“Only look at time data when the participant is engaged with the task. Do not include any time data when the participant is debriefing about her experience and still being tracked.”
“Be careful when analyzing eye movements on dynamic websites. ”
Finally, if you do decide to run an eye tracking study keep in mind that think-out-loud (TOL) and self-reporting surveys after the task complement eye tracking quite well. This is because combining objective eye tracking data with subjective self-report user’s responses can be beneficial in understanding the user’s conscious or unconscious behavior.
This concludes our introduction to eye tracking. Thank you for reading and don’t forget to check out the rest of our UX 101 education series to help you on your way!
John is a User Experience Researcher with 10+ years of practical experience in a broad range of quantitative and qualitative user research, analysis, and design for software and web products. John is focused on helping companies learn how to make their web and mobile products easier to use.
Mina is a UX Researcher at Google. Previously she was on the Professional Services team at UserZoom in San Jose, CA. Mina has her PhD in Information Systems with a concentration on Innovation with User Experience. In her spare time Mina enjoys playing guitar and doing yoga.