Effective use of eye-tracking in UX research

@steve_denning @uservision hello@uservision.co.uk
Stephen Denning
Director of User Experience
“Is what you see what you get?”
Effective use of eye-tracking in UX research

How we read
When we read, we don’t focus on
every word.
We only focus on approx. 30-60% of
words.
The more technical or complex the
language, the higher the percentage
of words we focus on.

1879 - Louis Emile Javal
Observation of reading behaviour
Early 1900s - Edmund Burke Huey
Physical tracking of eye movements
1922-37 - Buswell and Judd
First photographic recording of eye movement
1980s - Adoption in marketing
What were consumers looking at?
1990s - Early web
Treated as electronic versions of print media
2000s > Present - Use in usability/UX research…

@steve_denning @uservision hello@uservision.co.uk Credits: Tobii AB
“Eye-tracking is a methodology
that helps researchers
understand visual attention.”
(Romano Bergstrom & Schall)

@steve_denning @uservision hello@uservision.co.uk Credits: EyeGaze
Pupil centre corneal reflection
(PCCR)

Which visual content attracts attention
How quickly visual content is noticed
Which visual content is ignored/overlooked
In which order the visual elements are noticed
How one visual layout compares with another
Eye-tracking can tell us…
Credits: iMotions

If somebody finds content engaging
If a website is easy to use
If content is understood
If users like/dislike a design
If products/services meets users’ needs
If someone is more/less likely to convert
Why people are looking at what they are looking at!
Eye-tracking (alone) cannot tell us…

foveal
para-foveal
peripheralperipheral

@steve_denning @uservision hello@uservision.co.uk Credits: MeasuringU
The problem is…
We can often look at things
without ‘seeing’ them (i.e. gazing
without focusing)
We can also ‘see’ things without
looking directly at them (e.g.
banner blindness) or things that
may not even be there (e.g.
gestalt principles of
completion/closure)

Therefore, eyetracking has traditionally
been seen as…
Expensive
Complex
Ambiguous

Is it worth
the effort?

1. Ask the right questions
2. Choose the right equipment
3. Gather high-quality data
4. Communicate effectively
Effective use of eye-tracking

1. Ask the right questions

What do we want to find out?
Is eye-tracking necessary for this?
What would eye-tracking bring to the party?
What disadvantages could eye-tracking bring?
Is using an eye-tracker practical?
Should we use eye-tracking?

What is the stimulus?
What is the research question?
What data will answer that question?
What type of task will be used?
Credits: iMotions
How should we use eye-tracking?

What is the stimulus?
What is the research question?
What data will answer that question?
What type of task will be used?
Credits: iMotions
How should we use eye-tracking?
Areas of
Interest
Statistics

@steve_denning @uservision hello@uservision.co.uk Credits: Salesforce.com

Do users notice the “Buy
Now” call to action?
Do users notice the
payment method on offer?
Sensitivity Selectivity

Statistic Insight
Time to first fixation If long, then not visually obvious (in comparison to
surrounding elements)
Fixation duration Longer fixations can indicate interest or confusion
Number of fixations More fixations can indicate confusion over purpose,
difficulty understanding content or increased comparison
with other elements
Or, carefully read text
Percentage fixated How many participants are noticing elements
Time from first fixation to click Longer delays can indicate confusion, lack of clarity, or
difficulty in making a decision
Credits: Romano Bergstrom & Schall

@steve_denning @uservision hello@uservision.co.uk Credits: Alfred Yarbus
In 1965, Alfred Yarbus…
…ran an experiment to look at
how the task that someone is
presented with, will affect their
gaze.

@steve_denning @uservision hello@uservision.co.uk Credits: Alfred Yarbus

2. Choose the right equipment

Tower mount (Fixed-head)
Remote (head-free)
Wearable (head-free)

Device Output Freedom Data quality Analysis
Tower mount
(head-fixed)
Fixed pixel None Best Easier
Remote
(head-free)
Fixed pixel Better Lower Easier
Wearable
(head-free)
Point on scene
camera image
Complete Lowest Difficult

3. Gather high-quality data

@steve_denning @uservision hello@uservision.co.uk Credits: Freepik.com
Participant factors… heavy prescriptions, glasses frames, eye makeup…

Margin of error… remains
consistent across stimulus
sizes, so smaller stimulus
areas prove less accurate

Concurrent think-aloud (CTA)
Vs
Retrospective think-aloud (RTA)
Vs
Hybrid
Consider the
style of moderation…

Consider the
“head box”…

Consider the
calibration…

Consider the
calibration…
Small variances in the
calibration results can
result in significant margins
of error on the stimulus…

4. Communicate effectively

Heatmap Gaze plot Data chart
Most engaging Most useful

For best
results,
combine with
moderated
usability
testing…

Always encourage
observation…

So, to summarise…
„ Eye-tracking can provide useful data about what
people are looking at on a screen
„ It cannot give us qualitative feedback about the UX
„ Eye-tracking studies can offer benefits when they are
approached in a robust manner.
„ We must:
• Ask the right questions
• Choose the right equipment
• Gather high-quality data
• Communicate the results effectively
Stephen Denning
UX Director | User Vision
hello@uservision.co.uk
www.uservision.co.uk
@uservision
@steve_denning

Effective use of eye-tracking in UX research

Recommended

More Related Content

What's hot (20)

Similar to Effective use of eye-tracking in UX research (20)

More from Stephen Denning (8)

Recently uploaded (20)

Effective use of eye-tracking in UX research