ERIC Number: ED314015
Record Type: RIE
Publication Date: 1989-Mar
Application of Intelligent Tutoring Technology to an Apparently Mechanical Task.
The increasing automation of many occupations leads to jobs that involve understanding and monitoring the operation of complex computer systems. One case is PATRIOT, an air defense surface-to-air missile system deployed by the U.S. Army. Radar information is processed and presented to the operators in highly abstract form. The system identifies aircraft as friendly or hostile based on flight patterns and radio signals. The operator watches the system make life-or-death decisions about aircraft which are well out of visual range. He or she must understand what is happening during the few minutes that a track takes to traverse the radar's area of coverage and be prepared to override the system in cases of local exceptions to the decision procedure, intelligence from other sources, or overriding commands from higher echelons. Formative evaluation research is reported on INCOFT, an intelligent instructional system developed to train soldiers to monitor the automatic identification processes used by PATRIOT. The research indicates a significant way in which attempting to teach the apparently mechanical task of matching the PATRIOT identification algorithm is not an entirely mechanical process. Students do not see the task as understanding an algorithm but rather as determining the identity of the aircraft picked up on their radar. Simple heuristics are often preferred over the effortful mental arithmetic even when very few aircraft are involved. An initial step in training is to convey the notion that there is an algorithm on which these decisions must be based. Understanding and using the algorithm requires tolerating uncertainty as information on aircraft slowly accumulates. (5 references) (GL)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A
Note: Paper presented at the Annual Meeting of the American Educational Research Association (San Francisco, CA, March 27-31, 1989).