Path: utzoo!utgpu!news-server.csri.toronto.edu!neat.cs.toronto.edu!mgreen
Newsgroups: comp.human-factors
From: mgreen@cs.toronto.edu (Marc Green)
Subject: re: Eye Movement Trackers
Message-ID: <91Jun14.160659edt.6227@neat.cs.toronto.edu>
Organization: Department of Computer Science, University of Toronto
Date: 14 Jun 91 20:07:16 GMT
Lines: 52


>From: asylvain@felix.UUCP (Alvin "the Chipmunk" Sylvain)

>How about tossing that old mouse into the trash can, and construct a
>device which looks into your eyes while you're reading the screen, and
>can actually track the precise location of what you're looking at?

>Rather than "clicking" a mouse, you just touch a button when you want to
>select the word or screen-button that's in your current "gaze."  Natur-
>ally, a cursor follows your gaze around the screen, and touching the
>button flashes the cursor (or inverts the screen button, whatever) for
>immediate feedback.

>Such devices already exist, but they require physical contact with the
>eyeball.  This is undesireable (at least to me!)

>This may sound rather like science-fiction, but it's probably possible.

>Don't forget, if anybody _does_ invent such a thing, you heard it here
>first!

You obviously don't know much about the world of eye trackers. There
are numerous eye trackers which do not make contact with the eye. The
king of them all is the SRI double Purkinje tracker which detects
changes in the position of the Purkinje images in the the eye.
(Purkinje images are create by the different refractive indices of
different ocular media. All you need is about $60K and you have one of
your very own. There are also cheaper methods, glasses that bounce
infrared beams of the eye, EOG's which measure activity in the
ocular-motor muscles, etc. These have problems with accuracy and
reliability and many operate only in the horizontal plane.

All trackers require careful and frequent calibration.  Further, It is
not easy to tell where a person is looking, even if you know the
position of the pupil; there is a big difference between knowing the
position of the eye and the locus of gaze. People also make many
involuntary eye movements, so a tracker would unintentional cause
actions to occur. I don't think that eye trackers will ever become
popular. Just too many problems.

Marc Green
Trent University