Eye Tracking, Heat Maps and Pamela Anderson
Tags: eye tracking, heat map
Eye tracking, heat maps and Pamela Anderson. What do they have in common?
Just to spice up your Friday evening, take a look at the picture below, for no more than 5 seconds. Keep in mind what’s the first spots you looked at. Then, scroll down and see what does the computer perceives as “hot spots”.
Pamela Anderson
Now, let’s see where are the hot spots perceived by computer (the heat map of the picture) (thanks to the guys at www.feng-gui.com)
Pamela Anderson Computer Predicted Heat Map
Kind of strange, isn’t it? Now, don’t be ashamed, where did you look at?
What I am trying to say is that a computer is a machine, and even if it can detect skin and faces, smiles or eyes within pictures, it is still not accurate (yet). However, the Feng GUI tool is doing a much better job at predicting the heat maps for websites (or at least for ours). This a 1024 x 768 , above the fold screenshot, heat map prediction for our home page.
Pitstopmedia's Home Page 1024x768 above the fold
Heat Map Prediction
If you are designing your website and you want to take out a little bit of guessing from the design process by using the help of eye tracking tool, you may want to take into consideration of the following:
- human eyes (the direction they point to) and smiles (white teeth smile) also play an important role on where your visitors will look on the page. A picture with eyes looking straight at users will attract most of the eye movements , while the same image but with the eyes of the “hero” pointing to a product will “heat up” the product image
- a website seen at 800×600 resolution will have a different heat map compared to the same website seen at 1024 x 768 (what’s above the fold). You need to design for the bigger chunk of your audience
- foveal and periphal vision also affects the eye movement
- banner blindness also has some influence
- animation will change the hot spots
Pitstop Media is a Vancouver based conversion rate optimization company which offers results oriented landing page testing and conversion rate optimization services. Our A/B and multivariate tests have helped of companies increase conversion rates by as much as 75%. Let us increase your conversion rates, too!
No related posts.
Paul Olyslager said:
Aug 30, 09 at 12:21 amGreat post, personally I think heatmaps should be based on actual tracking. Every person looks at images in a different way, of course we have similarities. But what about the people who read from right to left, they have a completely different perspective on the design.
I’ve made a small list with free alternatives for Crazy Egg, which you can find at http://www.paulolyslager.com.
TraiaN said:
Sep 01, 09 at 10:02 amThanks Paul. The collection of heat map tools is in bookmarks now.
James Breeze said:
Jan 24, 10 at 3:25 pmHey, The Feng Gui heat map has no accuracy whatsoever.
A ‘normal’ person would look at the 2 eyes, mouth and the curve of the breast fat.
That makes more sense than the image above and it s scientifically proven.
TraiaN said:
Jan 29, 10 at 5:50 pmJames, do you have any links for studies on how we look at humans?
Peter Hartzbech said:
Jul 28, 10 at 3:06 amDear Traian,
Yes, we just did a test in less than 10 secs in our office with the same image with Attention Tool and then results are very precise, I will uploas a gaze replay of it later today to youtube and also some images of the output to Flickr.
Check out website http://www.attentiontool.com for a lot of different outputs on humans.
Thanks,
Peter
Peter Hartzbech said:
Jul 28, 10 at 3:07 amSorry, I meant 10 minutes :) Exposure time per respondent was 6 seconds. You do not need more to see clear results :)
Peter
TraiaN said:
Jul 28, 10 at 12:40 pmpeter, I am waiting for the gaze replay and images.
James Breeze said:
Jul 28, 10 at 4:06 pmHi,
Back on after a little break. Peter mentioned it to me.
Here’s an article abstract on how people look at humans. And another in the right hand column.
http://www.ncbi.nlm.nih.gov/pubmed/20169468
You’ll need to access the article separately.
Cheers,
James
TraiaN said:
Jul 28, 10 at 4:10 pmthat’s interesting :)”However, variations in breast size had no significant effect on eye-tracking measures (initial visual fixations, number of fixations, and dwell times). The majority of initial fixations during eye-tracking tests were on the areolae.”