Cognitive heuristics and feedback in a dynamic decision environment

Article Abstract:

The performance of heuristics in a complex dynamic setting, characterized by repeated decisions with feedback, is investigated. A simulated task resembling medical decision problems is described, computer models of decision strategies are developed, and selected task characteristics are systematically varied and their influence on performance is evaluated. Results show that the dynamic characteristics of feedback-related aspects of the task are a major determinant of when heuristics perform well or badly, and have a greater influence than task characteristics such as symptom diagnosticity and disease base-rates. The results also provide insights about the costs and benefits of various cognitive heuristics; for example, random acquisition strategies often did no better than the random benchmark.

Author: Kleinmuntz, Don N.
Decision-making, Decision making, Management, Comparative analysis, Health services administration, Diagnosis

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA


A standard measure of risk and risk-value models

Article Abstract:

The problems of risk measurement and risk-value tradeoffs are considered from a general and fundamental perspective. A general measure of risk is developed based on the converted expected utility of normalized lotteries with zero-expected values. Called a 'standard measure of risk,' it possesses a number of attractive characteristics that capture intuitively attractive conceptions of risk. It includes risk measures proposed in earlier studies as special cases, and also serves as a preference-based and unified approach to risk research. The measure permits explicit or implicit application in an expected utility model since it is compatible with the measure of expected utility. Finally, a standard risk-value model is proposed.

Author: Dyer, James S., Jia, Jianmin
Models, Risk assessment, Measurement, Risk (Economics)

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA


The reliability of subjective probabilities obtained through decomposition

Article Abstract:

Decomposition procedures can be used to improve the consistency of subjective probability encoding. The random error associated with decomposition estimates is described as a function of characteristics of the component assessments. A psychometric model is used to describe the random error. Decomposition and direct assessment are compared in terms of the percent change in measurement error that is due to the use of decomposition.

Author: Ravinder, H.V., Kleinmuntz, Don N., Dyer, James S.
Research, Management science, Probabilities, Probability theory, Decomposition (Mathematics)

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.