Reference Dependence in Bayesian Reasoning: Value Selection Bias, Congruence Effects, and Response Prompt...

Area of Science:

• Cognitive Psychology
• Decision Science
• Medical Diagnostics

Background:

• Diagnostic reasoning is crucial in many fields, yet susceptible to cognitive biases.
• Reference dependence, where judgments are influenced by presented values, can distort decision-making.

Purpose of the Study:

• To investigate how reference dependence, specifically value selection bias and congruence effects, impacts diagnostic reasoning.
• To analyze errors in simplified Bayesian reasoning problems influenced by problem structure and information salience.

Main Methods:

• Two studies evaluated accuracy and error types in simplified Bayesian problems.
• Analyses considered congruence, information amount, computational needs, and numerical abilities.

Main Results:

• A strong value selection bias was observed, with incorrect answers favoring provided values.
• A consistent error involved using the superordinate value (total sample size) instead of reference class values, weakening congruence effects.
• Experiment 2 showed introducing new samples increased reliance on the superordinate value.

Conclusions:

• Reference dependence significantly influences diagnostic reasoning, particularly through the salience of information like superordinate values.
• Cognitive biases, such as superordinate preference, can impair accurate diagnostic reasoning, highlighting the need for awareness and mitigation strategies.
• Higher numerical skills generally correlate with improved accuracy in diagnostic tasks.