Tuesday, January 15, 2008

Cheap eyetracking using mouse tracking

Googlers Kerry Rodden and Xin Fu wrote a SIGIR 2007 workshop paper, "Exploring How Mouse Movements Relate to Eye Movements on Web Search Results Pages" (PDF, pp. 33-36) that has a clever idea.

Eyetracking studies are really expensive, but mouse tracking can be done just by dropping some Javascript into a page. How close can we approximate the results from an costly eyetracking study just using cheap and easily implemented mouse tracking?

From the paper:
Eye tracking can provide insights into users’ behaviour while using the search results page, but eye tracking equipment is expensive and can only be used for studies where the user is physically present. The equipment also requires calibration, adding overhead to studies.

In contrast, the coordinates of mouse movements on a web page can be collected accurately and easily, in a way that is transparent to the user. This means that it can be used in studies involving a number of participants working simultaneously, or remotely by client-side implementations – greatly increasing the volume and variety of data available.

To capture mouse movements, we used ... a piece of Javascript code at the top of every Google search results page visited. This Javascript code captured the user’s mouse coordinates (and the ID of the HTML element the mouse was over) at 100 millisecond intervals, and submitted the gathered data, with timestamps, into a MySQL database every 2 seconds (or when the user left the Google search results page).
The work is just exploratory -- concluding only that mouse tracking "definitely [shows] potential as a way to estimate which results ... the user has considered before deciding where to click" -- but does have interesting results on the relationship between mouse and eye movements. It also reports on common mouse moving patterns they saw searchers do when considering on which search result to click.

6 comments:

AndyEd said...

We review the literature on eye-mouse synch in: INSTRUMENTING THE DYNAMIC WEB Edmonds, A., White, R. , Morris, D., Drucker, S., Journal of Web Engineering, Vol. 6, No. 3, 2007. http://research.microsoft.com/%7Esdrucker/papers/EdmondsJWE2007.pdf

It's interesting that mouse-eye synch is more common in pages where a click is likely to happen. If the page is all text, with no links, there strategy of keeping the mouse near the eye is less relevant. See also http://alwaysbetesting.com/abtest/index.cfm/2007/4/29/Eye-Tracking-vs-Mouse-Tracking

David Pennock said...

Here is another low-cost eye (ok, head) tracking hack using a Nintendo Wii remote:

http://www.cs.cmu.edu/~johnny/projects/wii/

"Using the infrared camera in the Wii remote and a head mounted sensor bar (two IR LEDs), you can accurately track the location of your head..."

Greg Linden said...

Thanks, Andy. I appreciate the link to that Microsoft Research paper. It was an interesting read.

As the Google paper says, what differentiates their work from the past work comparing eye and mouse tracking is their exclusive focus on web search, but I agree that your paper and some of the previous work like the Chen et al. 2001 paper is also quite interesting.

By the way, I really like this line from your paper: "Discriminating between a lack of usefulness and a lack of discovery is a key challenge." That nicely captures the value in mouse and eye tracking, that you can get information about whether the reason that no one clicks on something is because they never saw it.

Greg Linden said...

Thanks, Dave! That Wii demo is great! Very clever!

Unknown said...

Dear Greg,

You and your blog readers may be interested in a research I am performing with a group of colleagues where we propose to log keyboard activities, mouse tracking data, and \ or eye tracking results as a measure of user effort to be used for evaluating GUI Usability.

Our hypothesis is that GUI Usability, that is operability, understandability, and learnability of the GUI as well as the satisfaction level of the interaction with the GUI, is an “inverse function” of user effort.

The following two publications lay down the details of our hypothesis and describe a set of experiments we performed this summer to assess the hypothesis. Two papers that shed more light on our research and results have been submitted.

We are currently working on expanding our research. We get a lot of traction with software engineers. But, interestingly we face emotional and non rational resistance from the cognitive usability community. Are they concerned about their livelihood? Maybe
Cheers,

Dan Tamir

dt19@txstate.edu

D. E. Tamir, O. V. Komogortsev, and C. J. Mueller, “An Effort and Time Based Measure of Usability”, 6th Workshop on Software Quality, 30th International Conference on Software Engineering, Leipzig, Germany, May 2008.

C. J. Mueller, O. V. Komogortsev, D. E. Tamir and L. Feldman, “An Effort-Based Approach to Measuring Software Usability,” Technical report, TXSTATE-CS-TR-2008-9 , Texas State University 2008.

Pete said...

Have you tried mouseflow.com ? If not, you really should. Free playback of your visitors mouse and key actions, heatmaps and a lot more :)