A METHOD FOR SYSTEM TASK ANALYSIS USING STATISTICAL DECISION THEORY.
Abstract
For system tasks involving detection, a method is presented by which statistical decision theory may be used to derive limiting conditions for adequate operator performance from results obtained in an abstract laboratory task. The limiting conditions may then serve as a basis for making design decisions regarding functions allocation and for specifying operating rules and requirements. The method is illustrated with a collision-prediction task, where two visually displayed objects move on converging paths, in which previous research suggests that performance depends on the ability to detect a nonzero rate of change in the relative bearing between the two objects. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 12, 1965
- Accession Number
- AD0621999
Entities
People
- Alvin S. Cooperband
- Lawrence T. Alexander
Organizations
- System Development Corporation