Our instincts have taught us rules for and responses to emotional cues from other humans. This often unconscious form of communication has not made the leap into our technology. This results in a problem when people express emotions to a computer that cannot recognize the emotions. The proposed solution is a device that can begin to record some of these signals in a natural way [Picard, 1995].
Computers currently lack emotional intelligence, especially awareness of others emotions. There is evidence [Reeves and Nass, 1996] that despite this fact, humans interact with their computer as if it was capable of understanding and responding to emotion signals. Threatening, pounding on, coaxing, patting on the monitor, these are behaviors exhibited by computer users to express to their machine something of the user's emotional state. This is a natural form of communication that is wasted on the computer because it can not detect or interpret this input. The more frustrated the user becomes with the equipment, the more frustrating it becomes that the computer is not receiving this drastic form of expression. Because the computer can not detect or use the expression information, it can not take action to try to rectify the current state of tension.
One step in enabling computers to recognize the emotional cues from the
user is to study and understand how the autonomic system behaves for
various emotional situations. The specific autonomic response signals
being recorded are chosen for the availability of non invasive bio
sensors that can be used in conjunction with a wearable
computer for real time portable signal acquisition [Thad Starner, et al., 1997]. Such
measurements as blood volume (BVP), heart rate (EKG), galvanic skin
conductance (SC), and respiratory rate are commonly used in emotion
research experiments. For these particular signals characteristic
patterns have been found which correlate with different self reported
emotional states. The most widely accepted axes for the categorization
of emotions are valence, the discrimination between positive and
negative experiences, and arousal, the intensity with which the emotion
is experienced. These two axes have been widely accepted in many
diverse theories and research studies.