Geography and eye tracking, prospects and possibilities

As I’m coming to the end of my stay in Melbourne, I’m going to take the opportunity to reflect on some ideas I’ve been having about the possibilities for geographers presented by eye tracking technology.
 
In collaboration with Jodi Sita from the Australian Catholic University, I’ve written a grant proposal while I’ve been here which seeks to use eye tracking glasses to examine how cyclists engage with urban spaces.  These glasses overlay a record of where you’re pointing your eyes onto a video recorded from a front facing camera mounted at eye level.  These devices used to be a little Heath Robinson, but are now very slickly packaged with a similar weight and appearance to sports sunglasses, making them suitable for field-based use.
 
Meeting with a couple of eye tracking specialists while I’ve been here has given me the chance to think through some ideas around how geographers can engage with these technologies.  The majority of work in eye tracking is lab-based, asking participants to look at a screen and recording how long they spend looking at different elements of images displayed on the screen.  Again, this used to be a technology that was very expensive and complex, but which has tumbled in price and rocketed in usability over the last few years.
 
Jodi has just published an edited collection looking at what eye tracking can bring to film and TV studies.  I had a fascinating conversation with Angela Ndalianis of Swinburne University of Technology about her collaboration with Jodi examining some of the assumptions made by film studies scholars about how audiences watch movies – what elements filmmakers intend to draw the eye vs. the parts of the screen that audiences actually look at.  A fascinating opportunity to debunk some long-held theories and confirm others.  Jodi has also run an amazing project examining how people respond to green space, showing them film of walks through parks to see what elements they pay attention to in the landscape.
 
In conversation with Adrian Dyer at RMIT, he revealed his concerns that a great many studies being undertaken with eyetracking these days (not including those described above) are insufficiently rigorous and overly descriptive.  This is doubtless a fair point.  But there’s something about this moment with the technology that allows for research to emerge that might not meet the standards of rigour of conventional approaches to eye tracking, but which can nonetheless give new insights into a variety of different areas.
 
One can buy a basic screen-based eye tracker for less than £200, although this unfortunately lacks the specialist software that allows you to do the automated analysis.  That specialist software gets quite expensive quite quickly – about €2500 for a year’s subscription – but allows you to start identifying how much time people spend examining different elements within an image. 
 
At high levels of sophistication, eye tracking can give insights into people’s decisionmaking.  Market researchers use this to determine things like how the design of packaging can make people more or less likely to buy a product (there’s a group at Monash working on this).  One of the things I intend to do when I get back to the UK is see if I can hack together something relatively crude to allow students to do basic analysis using a cheap gaming eye tracker.  From a scientific point of view, this would be entirely without rigour, but for demonstrating the principles of how one could start to use this technology, I think it could be very useful.  Once something usable falls below £1000, however, I can see that there would be serious interest among geographers in what this technology could do.  Scholars working on, for example, place, mobilities, landscape etc. could gain some fascinating new insights working with these techniques.