Introduction
One of the trends in studies of human behavior is a growing interest in devices for measuring eye movement and point of gaze. In addition to issues related to the eye-tracking technology itself, problems arise when using an eye-tracking system as input to video analysis programs such as The Observer Video-Pro. Eye tracking systems, such as iView, generate a stream of video images showing the visual scene seen by the test subject with a crosshair indicating the point of gaze. In order to obtain data suitable for quantitative analysis, the video signal must be processed frame-by-frame and transitions of the point-of-gaze between zones of interest must be scored. The resulting series of events can be subjected to standard statistical methods to obtain frequency, duration and sequential data. Getting from visual scene to statistical analysis, with several hardware and software interfaces in between, is not trivial. Similar problems face workers in various diverse fields, including:

• Human factors and usability research on air traffic control systems, car dashboards, aircraft cockpits, mobile phones and other equipment with a user interface
• Usability testing software and web sites
• Studies on the behavior of people in supermarkets, museums, etc.