Eye tracking visualizations can communicate important aspects of visual behavior clearly and with great impact. It can give you quick overviews and can greatly shorten your time to finding insights. On this page we have gathered the most common types of eye tracking visualizations.
Table of Contents
Heatmaps show the distribution of gaze, and can reveal the focus and attention of multiple users simultaneously. Regions of high gaze durations in heatmaps aren’t always directly tied to user interest, and could imply other things such as confusion. This is heavily contextual but allows you to pinpoint irregularities or patterns that you can then for example ask the users about in more detail about.
One of the biggest values from heatmaps is that you can quickly get an overview of an entire test set of users, and quickly compare that to other heatmaps of other iterations of the design.
Keep in mind that aggregated heatmaps can be difficult to analyze for dynamic scenarios where for example objects can be moved and interacted with.
A converse of the heatmap is a notion called perception mapping. This is where you only fill in the details of a scene as the user perceived them in their fovea region – adding color, texture, contrast in a physio-realistic manner. This gives you immediate insights into the user focus, and when aggregated across many users, can give you an average summary of how your environment is perceived overall.
This does not account for your peripheral vision, and even though you see less details here, it still makes up the biggest portion of your field of view.
Object-based Color Coding
Color coding individual objects can be great in order to get an overview of large sets of aggregated data to, for example quickly see what objects in a scenario draws the most attention.
To not interfere with the design of an object, a color-coded box around the object can be visualized instead of the object itself.
Gaze plots show the location, order and time spent looking at locations. The below example show gaze progression on a product box, where the areas looked at are numbered and the size of the circles indicate relative duration spent looking there.
Gaze plots are useful when you want to understand the gaze progression of users, for example in product and user interface design.
When a user spends time looking back and forth between two or more objects, it can be assumed that the user is making (or is attempting to make) some correlation between those objects. The more back-and-forth, the stronger the attempt at mental correlation.
Like many quantitative data points, this should be disambiguated by using more data collection, for example questioning the user about what relationship they were making while performing the action.
Being able to replay what a user looked at in a complete, replayable context enables testers to “walk in their users shoes” and get some understanding from their perspective.
This also enables asking users to clarify the actions they performed at different points during the recording without having to ask and disturb them during the study. It’s also easier for the user to remember and explain their thought process during a specific part of the study if they can watch the replayed visualization.