download pdf version

Introduction Pilots use an array of instruments to monitor aircraft performance and to detect if and when performance is not meeting expectations. There are, however, no gauges to monitor the most critical function in the system—the pilot’s mental performance— until now. Lockheed Martin Advanced Technology Laboratories (LM ATL) has developed the Sensor-based Mental Assessment in Real Time (SMART) tool to monitor an individual’s mental state. SMART Technology Based on both internal and DoD-funded research, the SMART system uses task independent and user-independent algorithms to convert physiological data into multiple discrete gauges that display mental activity. These mental state gauges —they currently include cognitive workload, distraction, engagement, and drowsiness— provide insight into an individual’s mental processing in a manner that has previously been unavailable. Previous methods for assessing mental state are either subjective and/or intrusive and produce infrequent measurements. In contrast, SMART provides second-by-second objective measures of mental state without interrupting performance. SMART also allows post event analysis of the operator’s mental state in the performance of a task. SMART includes a common timeline that allows assessors to view data from multiple gauges on a common graph so that they can investigate patterns in the operator’s mental state. Applications SMART works in any domain where it is useful to continuously monitor or generate a log of mental state changes. SMART provides an objective assessment during the natural flow of tasks being performed. Two domains in which SMART can provide immediate benefits are: (1) human factors engineering and (2) education and training. Human factors engineers can use SMART data to test alternate interface designs during task performance. For example SMART provides: (1) views of multiple, real-time streams of sensor data for interface evaluations;(2) guidance for improving existing user interfaces by comparing mental state effects of new versus baseline designs; and (3) objective, application-independent measures of subjects’ mental states —including cognitive workload— during tasks. For training, data collected by SMART: (1) allows instructors to see if a task requires less mental workload with practice; (2) allows trainers to pinpoint tasks that may require additional engagement; and (3) enables teachers to monitor students’ levels of distraction during training sessions. In the future, SMART technologies may be used to monitor an operator’s performance during critical functions, such as battlestation management or air traffic control. For future applications, the necessary sensors can be incorporated into a helmet or headset. Connecting an operator to an “instrument panel” could be as simple as putting on a headset.