Framework for Eye Tracking Patterns and Usability Problems: Pt 2
In a recent post, we started to review new research being done at the Centre for HCI Design at City University in London. Two researchers, Ehmke and Wilson, set out to establish a framework by which to observe and analyze any correlations between eye tracking and what a person is thinking as they’re maneuvering around a website during a usability study. While both the Internet and usability studies have grown immensely in popularity over the last decade, we have yet to see an established connection between these two aspects – or the “eye-mind hypothesis,” Just and Carpenter’s theory that assumes a user is thinking about what they’re looking at at any one moment.
While we assume this to be true, we don’t have much else to go by in way of interpretation, and the fact that the human eye darts and rests (saccades and fixations) so quickly, it’s a very difficult thing to measure and analyze. There’s certainly a lot of money being made by businesses offering professional usability studies, heat maps, and site evaluation. But if you are paying high prices for these services, you want to know it’s working, and Ehmke and Wilson argue that just assuming the correlations are true is not enough.
In their paper, Ehmke and Wilson lay out a typical interpretation scheme for a usability study: Long fixations imply interest or confusion. Back-track saccades could be interpreted as possibly confusion. Not looking at elements of a page is indicative of a page not working. Scanning behavior rather than reading, or saccades and fixations not in a left-to-right pattern of sweeps mean a user is looking for something. Back and forth between two objects? The user must be trying to make a choice, perhaps a comparison. The first place a user looks might mean something that drew their attention, and the last place – maybe they lost interest… why?
You get the point. But what’s so frustrating about these evaluations is that that’s exactly what they are: evaluations. Subjective findings about how a user saw the site at that moment in time. Often, findings like these are noted during the test and then discussed with the user afterwards. Or a user is asked to provide retrospective protocols cued by a replay of their eye tracking data to stimulate their original feelings and thoughts, which they can then explain. This method is called Post-Experience Eye-Tracked Protocol, or PEEP, which assumes that concurrent think-aloud protocols can be incomplete and distracting for the participant. Perhaps there’s something going unnoticed in the more subconscious cognitive processes.
We’ll stop here and continue with more about their process in the next post.
Identifying Web Usability Problems from Eye-Tracking Data
Related articles:
- Framework for Eye Tracking Patterns and Usability Problems: Part 1
- Framework for Eye Tracking Patterns and Usability Problems: Pt 3
- Framework for Eye Tracking Patterns and Usability Problems: Pt 4
- Shocking Revelation: Eye Tracking Has Problems
- Eye Tracking and F Patterns: Recurring Theme in Web Usability
- Is Eye Tracking for Usability Studies Worth the Trouble?
- The Latest in Eye Tracking Web Usability Research pt2
- Mastering Eye Tracking Web Usability Metrics
- Eye Tracking and Usability: Which Metrics Are Valuable?
- Moodle Using Eye Tracking to Study Usability and eLearning