Cognitive bias

From Systems Analysis Wiki
Jump to navigation Jump to search

Cognitive biases are systematic errors, inaccuracies, and contradictions that can arise in the process of perception, evaluation, and selection of alternatives by a decision-maker (DM). These biases are caused by the peculiarities of human cognition, perceptual limitations, informational uncertainty, emotional state, and other psychological factors.

Within the framework of decision theory, cognitive biases are considered deviations from the rational model of behavior, where a DM's preferences become inconsistent, contradictory, or demonstrably erroneous from the standpoint of logic and formal selection procedures.

Causes of Cognitive Biases

According to Petrovsky, cognitive biases in a DM's behavior can be caused by:

  • Complexity of the situation being analyzed — high cognitive load on attention and thinking.
  • Lack or unreliability of information — incomplete or distorted data.
  • Time constraints — the need to make decisions under tight deadlines.
  • Cognitive limitations — the inability of the DM to consider all aspects of a situation or construct a logically consistent model.
  • Individual predispositions — over- or underestimated evaluations, a propensity for risk-taking, or, conversely, excessive caution.
  • Inattention, fatigue, or overload — factors that reduce the quality of analysis.

Such biases are characteristic not only of non-professional decisions but also of the behavior of experienced specialists, especially under conditions of high uncertainty and pressure.

Manifestations of Biases

  1. Inconsistency of preferences — The DM may change their opinion with a minor change in the problem's conditions, violating the transitivity of judgments.
  2. Contradictory evaluations — An option previously considered the best may be rejected without objective reasons.
  3. Perceptual limitations — The DM ignores some relevant information or assigns excessive importance to insignificant features.
  4. Context-dependent choice — The same option can be perceived differently depending on how the problem is framed or the order in which alternatives are presented.
  5. Susceptibility to external factors — The choice depends on the current state, the authority of the information source, group pressure, etc.

Approaches to Identifying and Mitigating Biases

In decision theory, the necessity of identifying and mitigating cognitive biases is recognized during the stage of constructing and refining the preference model. The following methods are used for this purpose:

  • Interactive methods for checking judgment consistency — repeated comparisons, logical checks for transitivity and completeness.
  • Supporting the DM in formulating and clarifying preferences — applying verbal methods and expert procedures.
  • Using verification procedures — for example, duplicating questions, cross-checking, and logical control of answers.
  • Involving consultants and analysts — for external quality control and to ensure the consistency of expressed preferences.

Significance for Theory and Practice

Understanding the nature and mechanisms of cognitive biases allows for:

  • Improving the quality of decisions made.
  • Adapting selection methods to the actual conditions of human cognition.
  • Developing more flexible and robust models oriented toward subjective yet rationally structured behavior.