Confirmation Bias through Selective Overweighting of Choice-Consistent Evidence

Area of Science:

• Cognitive psychology
• Neuroscience
• Decision science

Background:

• Human judgment often deviates from available information, a phenomenon linked to cognitive biases.
• Confirmation bias, where initial decisions influence subsequent judgment, is often explained by dissonance suppression.

Purpose of the Study:

• To investigate the underlying mechanism of how choices bias subsequent information processing.
• To test the hypothesis that choices selectively alter the weighting of new evidence, similar to selective attention.

Main Methods:

• A novel psychophysical task involving random dot motion stimuli and sequential motion-direction judgments.
• Participants performed a categorical discrimination after the first stimulus and a continuous estimation after the second.
• A computational model was used to test choice-dependent selective gain modulation against alternative mechanisms.

Main Results:

• Participant sensitivity to the second stimulus was selectively enhanced when it was consistent with their initial choice.
• The observed effect was better explained by choice-dependent selective gain modulation than by alternative models.
• This choice-dependent gain modulation was replicated in a task involving numerical value averaging.

Conclusions:

• Intermittent choices direct selective attention during the evaluation of subsequent evidence.
• Decision-making and selective attention exhibit a recurrent interplay, potentially mediated by decision-related neural feedback.