For several decades now, techniques have existed for tracking the gaze of a person. Until recently, most applications of eye tracking have been in psychological research for probing into subjects' perceptual or cognitive processes, for example when driving in traffic or reading text (Wierda & Maring 1993, Just & Carpenter 1976, Just & Carpenter 1980, Rayner & McConkie 1975, Graf & Krueger 1989), or for examining the process by which the eye-movements are determined. Especially, research into human visual search strategies has based itself on tracking subjects' gaze during "target object" search tasks (e.g. Scott & Findlay 1993), but also evaluation of computer displays has been based upon recordings of fixation patterns.
Initially, the techniques were only usable for laboratory experiments: the eye tracking equipment was large and expensive, and required the user's head to be fixed, either in a frame or by using a bite-bar. Some eye tracking techniques using contact lenses (cf. section 2.3) are so obtrusive that they cannot be used for extensive periods, and are thus useless outside the laboratory. In these laboratory experiments, data would typically be collected and analysed off-line, i.e., after the subject had finished the experimental task.
This state of affairs has changed, though; military research first developed head-up displays (displays integrated in the windscreens of aircraft, so that instrumental data is displayed "on top of" the surrounding flight scene) combined with eye tracking for guiding the missile system, thus freeing the pilot's hands for other tasks (Smyth et al. 1994, Cleveland & Cleveland 1992). This naturally required on-line processing of the tracking data, and this processing was not aimed at probing the pilot's perceptual or cognitive processes, but rather letting the pilot use the eyes as an extra manipulation channel. Later, the on-line processing of eye-gaze tracking data was extended to user-interfaces for non-military purposes.
The techniques are still expensive, but less so now, and the equipment has
been reduced in size. Some techniques have been made totally unobtrusive
and now allow for (small) head movements
(cf. sections 2.1.3 and 2.1.4
![[*]](http://www.diku.dk/images/latex2html//cross_ref_motif.gif)
), making them usable for
enhancing the user-interfaces for disabled people-especially
quadriplegics (cf. section 4.3.2).
While we find that the use of eye controlled word processors to aid disabled people in their daily life is an important development of eye-gaze tracking technology, our main interest lies in applications for the general community. If this is to have an effect, it must naturally also be based on on-line processing of the tracking data, to provide a real-time interface response to the user's eye movements. Thus, "[w]e seek ...to use eye tracking to provide an effective input method for a user whose hand might be resting on a mouse, who is fully able and available to operate the mouse, but who will find the eye movement based interface better, that is, faster, more convenient, or more natural" (Jacob 1995).