Partition-edit-count: naive extensional reasoning in judgment of conditional probability

Area of Science:

• Cognitive Psychology
• Decision Science
• Probability Theory

Background:

• Understanding how individuals assess conditional probabilities is crucial for cognitive science.
• Existing models often assume rational processing, but human judgment can be influenced by subjective factors.

Purpose of the Study:

• To investigate the cognitive processes underlying human evaluation of conditional probabilities.
• To propose and test a novel model, the partition-edit-count model, for probability judgment.

Main Methods:

• Experimental studies involving conditional probability problems.
• Manipulation of irrelevant information, problem wording, and event grouping.
• Protocol analysis to support the proposed cognitive model.

Main Results:

• Participants' probability judgments were significantly influenced by irrelevant information, wording variations, and event grouping.
• These findings align with the predictions of the partition-edit-count model.
• The model was extended to account for judgments under uncertainty where events are not interchangeable.

Conclusions:

• The partition-edit-count model provides a robust explanation for subjective conditional probability evaluation.
• Cognitive biases and subjective interpretations play a significant role in probability assessment.
• This research offers insights into human reasoning under uncertainty and chance.