A quantum probability framework for human probabilistic inference.

Area of Science:

• Cognitive Science
• Psychology
• Decision Making

Background:

• Human inference judgments vary widely, sometimes aligning with Bayesian (normative) principles and other times deviating.
• Existing models struggle to unify both Bayesian and non-Bayesian influences in human reasoning.

Purpose of the Study:

• To propose a unified explanation for human inference that integrates both Bayesian and non-Bayesian influences.
• To introduce a novel framework using quantum probability theory to model mental representations in inference.

Main Methods:

• Developed a hierarchical model of mental representations, ranging from fully quantum to fully classical.
• Investigated how assumptions about compatibility (joint event representation) change across this hierarchy.
• Conducted three experiments to test the model's explanatory power.

Main Results:

• The quantum probability model successfully explains five key phenomena in human inference: order effects, reciprocity, memorylessness, violations of the Markov condition, and antidiscounting.
• No existing theory could account for all five phenomena simultaneously.
• Model demonstrated that classical representations improve with task familiarity and individual cognitive style (Cognitive Reflection Test) influences representation.

Conclusions:

• Quantum probability theory offers a unified framework for understanding the diverse landscape of human inference.
• The proposed hierarchy of mental representations provides a flexible account of both normative and non-normative judgments.
• Task familiarity and individual cognitive differences shape the transition between quantum and classical representational states.