On Aug 14, 2009, at 7:48 AM, Joshua Porter wrote:
Interesting take from Google on their use of eye trackers:
"In addition to search research, we also use eye-tracking to study
the usability of other products, such as Google News and Image
Search. For these products, eye-tracking helps us answer questions,
such as "Is the 'Top Stories' link discoverable on the left of the
Google News page?" or "How do the users typically scan the image
results — in rows, in columns or in some other way?"
Eye-tracking gives us valuable information about our users' focus of
attention — information that would be very hard to come by any other
way and that we can use to improve the design of our products.
However, in our ongoing quest to make our products more useful,
usable, and enjoyable, we always complement our eye-tracking studies
with other methods, such as interviews, field studies and live
experiments."
More here:
http://googleblog.blogspot.com/2009/02/eye-tracking-studies-more-than-meets.html
Ok, may this warrants a more serious response.
The problem with Google & everyone else's use of eye tracking is that
it requires a leap of faith from observation to inference.
We can see the observation clearly and most of the time, we can agree
on them. An observation is that the user's gaze was recorded on a
specific x/y coordinate for a specific time period. Another
observation might be that the device didn't record any gaze fixations
on a different x/y coordinate.
We might also observe that the first x/y coordinate matches up with a
link to a news story. The second x/y coordinate matches up with an
advertisement.
So, we could conclude that the fixation of the user was on the news
because they were interested in it. And that they didn't look at the
ad because they weren't interested.
But that conclusion could be very flawed. Assuming we can account for
any calibration errors in the device (where the x/y coordinates didn't
actually match the news link or ad -- a frequent occurrence in state-
of-the-art eye tracking systems), we still don't know the brain
activity behind the gaze fixations.
Maybe they stared at the news link because they were completely
baffled by the headline? Maybe they didn't realize it was a news
headline and thought it was something else?
Maybe they actually saw the ad in a quick, transitive glance that was
too fast for the eye tracker to pick up? Maybe they registered the ad
out of there peripheral vision, beyond that of the foveal focus
region? (Many eye trackers won't show an experienced user's eyes
moving to scroll bar even though they move their mouse there to
scroll. It seems they acquire the scroll bar with peripheral vision,
keeping their focus on the items of interest on the screen.)
Jumping too quickly from observation to inference is the #1 cause of
design problems. We assume things without eliminating other
possibilities, make assumptions, and run with them. Spending a little
more time to test our inferences, to ensure we've properly qualified
them and eliminated alternative explanations can save a lot of energy
and downstream problems.
(I've written about this in an article called "The Road to
Recommendation": http://www.uie.com/events/roadshow/articles/recommendation/)
So, here's the problem with eye trackers: Every inference must be
tested without the eye tracker. As the folks from Google say:
we always complement our eye-tracking studies with other methods,
such as interviews, field studies and live experiments.
Fact is, had they started with the other methods, they wouldn't have
discovered anything new in the eye tracker. And the other methods are
cheaper, more efficient, and more beneficial.
There is one advantage to eye tracking hardware. On a recent visit to
the Googleplex, I asked about their usage their and this observation/
inference problem. They agreed with me, but told me about the "real"
reason they use the devices.
It turns out, the engineers and developers are more likely to attend
usability tests when the eye tracker is in use. In the few labs they
have that aren't outfitted with the devices, the engineers and
developers rarely attend. They line up to watch eye tracking tests.
For that purpose, the device may have some value. But so does good
chinese food. I've found that a quality catering job is much more cost
effective than mucking with the toys. (At Google, that might not work
so well, since they have four-star chefs in their cafeteria -- hard to
top that with catering.)
That's my more serious response. It pisses off eye-tracking
aficionados world-wide. I'm good with that.
Jared
Jared M. Spool
User Interface Engineering
510 Turnpike St., Suite 102, North Andover, MA 01845
e: jsp...@uie.com p: +1 978 327 5561
http://uie.com Blog: http://uie.com/brainsparks Twitter: @jmspool
________________________________________________________________
Welcome to the Interaction Design Association (IxDA)!
To post to this list ....... disc...@ixda.org
Unsubscribe ................ http://www.ixda.org/unsubscribe
List Guidelines ............ http://www.ixda.org/guidelines
List Help .................. http://www.ixda.org/help
