The road not taken: selective sampling and persisting inaccurate impressions

Area of Science:

• Cognitive Psychology
• Decision-Making Science
• Behavioral Economics

Background:

• Individuals often learn through experience, favoring choices perceived to yield optimal outcomes.
• Prior research indicates a tendency to consistently select perceived optimal choices, which can bias evidence and perpetuate inaccurate beliefs.
• The influence of the learning context's emotional valence (positive vs. negative) on this bias persistence remains underexplored.

Purpose of the Study:

• To investigate whether the persistence of choice bias in experiential learning is modulated by the valence of the learning context.
• To determine if negative learning contexts promote the maintenance of inaccurate beliefs compared to positive contexts.

Main Methods:

• Participants engaged in a two-armed bandit task with identical options.
• Initial evidence was manipulated to create a belief in one option's superiority.
• Experiment 1 used financial gains as a reinforcer (rewarding context); Experiment 2 used the omission of an aversive event (aversive context).

Main Results:

• Behavioral measures in Experiment 2 (aversive context) indicated a lasting bias.
• Explicit contingency estimates in the aversive context were also biased.
• No significant bias persistence was observed in Experiment 1 (rewarding context).

Conclusions:

• Outcome valence significantly modulates the persistence of bias in experiential learning.
• Negative learning contexts can foster the maintenance of irrational beliefs.
• Findings highlight the importance of context in shaping belief updating and decision-making processes.