A number of approaches are being pursued around the world to devise human- computer communication mechanisms that do not require mechanical involvement of the user's limbs. An interface capable of efficient operation under that constraint would be very valuable to individuals with severe motor disabilities (Spinal Cord Injury (SCI) and Amyotrophic Lateral Sclerosis (ALS) patients, etc.) The availability of such an efficient means of interaction with personal computers has the potential of enhancing the living standards of these individuals, by empowering them to participate in the increasing exchange of communication through computer networks. We continue to witness the proliferation of software accessible through Graphic User Interfaces (GUIs) for the most diverse applications, and the explosion of customer services available through World-Wide Web (WWW) browsers. Accordingly, the development of a truly practical "Hands-Off' Human-Computer Interface (HOHCI), could restore some areas of activity in the lives of these individuals to normality.
Although advances in the area of speech recognition as a means for communication from the user to the computer continue to be made, we believe that HOHCI systems that adhere to the point-and-click paradigm underlying current GUIs will be more readily applicable to this generation of personal computers. In line with this vision, we have pursued the development and/or application of three fundamentally different mechanisms for computer cursor control: The first is the "Brain-Computer Interface" (BCI) system, based on the classification of Electroencephalogram (EEG) patterns voluntarily generated by the user. The second type is a HOHCI driven by Electromyogram (EMG) patterns resulting from voluntary movement of facial muscles by the subject.____________________