User:Lewiscatton/HILAS Eye Tracking

=Eye Tracking= Eye tracking is monitoring where persons are focussing their eyes, or identifying the Area of Interest (AoI) that they are focussing on.

Eye measures comprise a number of variables that all can be measured and recorded with an eye tracker. The measures can be used, like psychophysiological measures as indicators of fatigue, mental- or visual workload, operator scanning strategy and also as indicators of attention. Eye tracking can be used as stand alone HF tool, however, it is strongly recommended to measure according to the “converging evidence” principle where the eye tracker data is combined with data from other sensors and sources to clarify the interpretation of the data.

Many derived measures can be deduced from the raw eye tracker data (Harris et al., 1986; Stern et al., 1984; Stern & Kelly, 1984; Stern, 1994; Goldstein et al., 1985; Wilson et al., 1987, 1993). In order to be able to cognitively process visual information a person must fixate at that point for a prolonged period of time and within a small visual angle. It depends on a number of factors how long that period should be and what the magnitude of that visual angle should be. In other studies where eye tracker data was used, this ranged from 100 to 200 ms. and 1 to 1.5 degrees.

Since an eye tracker records raw eye positions and movements, these must be recalculated into fixations and dwells in order to give them meaning for researchers. The process from raw eye tracker data to meaningful dwell time is illustrated below.

=Output= The eye trackers that were used in the HILAS project have two main data output formats. The first is video based. In order to create that, a second video camera is mounted on the subject’s head next to the eye tracker equipment. This camera films everything that is visible in front of the subject. After a calibration procedure the position and orientation of the subject’s eye is superimposed on that video camera image as cross hairs that indicate precisely where the subject’s eyes are aimed at. That video based output format is intuitive to interpret.

The other output format is actually the combined output of an eye tracker and a head tracker. This combination is referred to as a gaze tracker. The head tracker determines the position of the eye in a coordinate system in which also the AoIs (for example cockpit displays) are described. From the centre of the eye the system determines a so called eye line of gaze. When the eye line of gaze crosses an AoI in the coordinate system, the eye is aimed at that AoI. When this happens for a prolonged period of time the person might be looking at that AoI. The system can create a file in which time of day, x and y coordinate where the eye line of gaze crossed the AoI, pupil diameter, and some other parameters can be recorded. That file is less intuitive to interpret than the video output format. But it is very efficient to analyse it in software packages like Excel or SPSS.

There are several measures that can be derived from raw eye data, which are summarised in the table below. This table also includes the typical field of application for each measure.

=Methodology= In the HILAS experiments head mounted video based eye trackers were used. Video based refers to the fact that these systems have a infra red sensitive camera to film the eye for data input. The eye tracker that was used in HILAS is the Applied Science Laboratories (ASL) 501 with Ascension Technologies optical head tracker. Two trackers were placed in the cockpit (one for each pilot). The pilots wore a headband on which the optronics were mounted.

A trained operator is needed to install install and calibrated the equipment for each subject. The data are stored during the experiments and the quality of the data can be verified both during the experiment, and more detailed post experiment. After that the data analysis can be performed. Even though the equipment is not difficult to switch on and the equipment records data relatively easy, it is recommended to apply experienced staff during the entire eye tracking process, from preparation to data analysis, in order to get good quality data and interpretations.

=References=
 * HILAS Home Page
 * Flight Deck Technology Strand