Testing the limits of optimality: the effect of base rates in the Monty Hall dilemma

Area of Science:

• Cognitive Science
• Animal Behavior
• Probability Theory

Background:

• The Monty Hall dilemma is a classic probability puzzle involving three doors and a prize.
• It tests decision-making under uncertainty, with a counterintuitive optimal strategy.

Purpose of the Study:

• To compare how pigeons and humans learn and perform the optimal strategy in variants of the Monty Hall dilemma.
• To investigate differences in probability learning and strategic decision-making between species.

Main Methods:

• Pigeons and humans participated in two variants of the Monty Hall dilemma with varying probabilities behind doors.
• Performance was tracked over training to analyze strategy development and reinforcement rates.

Main Results:

• Pigeons learned to approximate the optimal strategy: choosing the lowest-probability door and then switching.
• Humans failed to adopt the optimal strategy, not preferring low-probability options or consistently switching.
• Pigeons demonstrated a learned sequence for maximum reinforcement, while humans explored varied responses.

Conclusions:

• Pigeons exhibit superior probability learning and adaptive strategy formation compared to humans in this dilemma.
• Human performance suggests a potential over-reliance on complex strategy seeking, hindering optimal probabilistic decision-making.