Non-Optimality in the Diagnosis of Dynamic System States
Abstract
This research note discusses various sources of non-optimality in the diagnosis of a dynamic system's state, looking at them within the context of a military flight scenario. Subjects examined integrated cues which varied in their informational worth under different conditions of information load and clue salience. Actual responses were correlated with an optimal response function, as well as with seven non-optimal response functions, modeled on the basis of filtering, heuristics, and salience biases. Sequential updating strategies were also analyzed. Results from the two studies indicated that the optimal response function provided the best fit to the data. The imposition of time stress produced a slight bias in favor of processing more salient display locations. A significant performance decrement occurred in secondary task conditions, manifest in trend toward conservatism in judgement, but no biases in display sampling. Analysis of sequential updating strategies also suggested that hypothesis updating was somewhat conservative. Keywords: Dual task performance, Decision making, Anchoring, Information processing, Information load, Heuristics, Judgement, Optimality, Salience; Displays; Stress.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1988
- Accession Number
- ADA202462
Entities
People
- Barbara Barnett
- Christopher Dow Wickens
Organizations
- University of Illinois Urbana–Champaign