On Sun, Apr 20, 2008 at 5:50 PM, Jared M. Spool <[EMAIL PROTECTED]> wrote:
> Paul Nuschke wrote: > > Eyetracking lets you see where people are looking in real time. > > Yes. But just because you know where someone looks or doesn't look > doesn't mean you know anything about what they see, what they wanted > to see, and what they didn't see. It's not clear to me how one > interprets the "they gazed at this point on the screen for 400 ms" > information. Was that good? Was that bad? Imagine that a user needs to click on a link to go somewhere. If she fixates on the link and don't click it, then that's pretty good evidence that she did not understand the link. We know that people see things through their peripheral vision, such > as the scroll bar, so that's not recorded by the eye tracker. That > means we can't even assume that when someone doesn't gaze at a spot > that it wasn't seen. True, but that's a good thing. You can't read or see fine details in your peripheral vision, so even if you notice something it doesn't mean that you looked at it enough to understand what it contained (unless the important details were very big). In the example above, even if the user noticed that a link existed, if she did not attend to it, then she would not have been able to read it. Show me a study that shows that N separate evaluators looked at the > same eye tracking data and came away with the same conclusions and > I'll change my mind. That some data does not make sense is not a phenomenon unique to eyetracking. I've seen plenty of different interpretations of statistics as well. > >One analogy I find useful, in terms of understanding what the > participant is doing/thinking, is that having eyetracking versus not > having eyetracking is like testing in person versus testing remotely. > > You lost me there. > In remote testing, you loose voice quality and you don't see mannerisms, facial expressions, etc. In "in person" testing, you have gestures and facial expressions, and voice inflections. In eyetracking, you add the ability to see where they are looking. You lose something too, though, in your testing methodology, but that's another e-mail thread. Paul ________________________________________________________________ Welcome to the Interaction Design Association (IxDA)! To post to this list ....... [EMAIL PROTECTED] Unsubscribe ................ http://www.ixda.org/unsubscribe List Guidelines ............ http://www.ixda.org/guidelines List Help .................. http://www.ixda.org/help
