Discovering a Decision Maker’s Mental Model with Instance-Based Cognitive Mining:

Abstract

The paper "Discovering a Decision Maker’s Mental Model with Instance-Based Cognitive Mining" by David and Natalie Steiger introduces a novel approach to externalizing tacit knowledge—specifically, the mental models that guide decision-making.

Core Concepts:

• Mental Models: These are internal, often subconscious frameworks individuals use to interpret information and make decisions. They include assumptions, beliefs, and heuristics shaped by experience.
• The Challenge: Mental models are tacit and difficult to articulate, making them hard to analyze, share, or improve upon—yet they critically influence organizational decisions.
• Instance-Based Cognitive Mining (IBCM): This technique uses inductive learning algorithms to analyze multiple decision instances. By examining how a decision maker responds to various scenarios, the system infers a polynomial representation of their mental model.
• Benefits:
  • Makes implicit decision logic explicit.
  • Enables comparison between individuals’ models.
  • Supports training, knowledge transfer, and improved decision support systems.
• Application: The method was tested in a decision support context, showing promise in capturing and refining how individuals weigh factors and make judgments.