Operator State Classification Developer's Toolkit

Period of Performance: 04/24/1998 - 04/24/2000


Phase 2 SBIR

Recipient Firm

SAM Technology, Inc.
San Francisco, CA 94108
Principal Investigator

Research Topics


Mental Overload, fatigue, boredom and lapses in situational awareness, medication effects and side effects, illness, and other factors can impair performance in operators of complex systems. A reliable means of detecting and monitoring operator mental state would allow an intelligent system to alert the operator in the case of mental lapses, or to adaptively automate a subset of control functions in the case of operator overload. Physiological and neurophysiological signals such as eye movements, the electroencaphalogram, heart rate, and respiration can be used to classify mental states. Laboratory research has established that such measures can be used to identify which of several cognitive tasks an operator is engaged in, the level of cognitive load an operator is experiencing, and whether he or she is fatigued or otherwise impaired. Improved technology for psychophysiological research is needed to develop such measures to the point where they can be used in operational settings. In Phase I of this project we designed and began implementation of an "Operator State Classifier Developer's Toolkit" that extends the functionality of our commercial MANSCAN research system for this application. This proposal describes our progress and details our Phase II plan for fully implementing, testing, and commercializing this technology.