Human Failure Detection and Resource Allocaton in Dynamic Environments.
Abstract
Three major areas of research are summarized. (1) Processing Resource allocation between tasks in dynamic environments. Results indicate that these abilities are distinctly limited. Rather than effectively allocating in response to task demand increases, the operator appears to temporarily 'expand' the capability of available processing resources. (2) Failure detection in dynamic systems. Human operators are found to be better at detecting failures (step transitions n the order of a dynamic system) when they are active participants in controlling those systems, than when they are passive monitors of the system under autopilot control. Some reasons for this difference are discussed, while research explores the role of training, and information processing channels in the detection process. (3) The attentional demands of failure detection. A model of human processing resources is described which partitions these resources into structure-specific reservoirs related to stages of processing. Task interference patterns between failure detection in the controlling vs. the autopilot monitoring mode, and between two qualitatively different loading tasks are shown to be consistent with this structure-specific resource view. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1979
- Accession Number
- ADA069821
Entities
People
- Christopher Dow Wickens
Organizations
- University of Illinois Urbana–Champaign