Decisions reduce sensitivity to subsequent information.

Area of Science:

• Cognitive Psychology
• Neuroscience
• Decision Science

Background:

• Behavioral studies show categorical choices alter beliefs and create confirmation bias.
• Choice-dependent biases suggest limitations in human rationality.
• Computational mechanisms underlying these biases are not well understood.

Purpose of the Study:

• To investigate if choice-induced biases extend to lower-level decisions.
• To quantitatively dissect how choices affect evidence accumulation using novel psychophysical protocols.
• To elucidate the computational mechanisms behind choice-induced biases.

Main Methods:

• Developed novel psychophysical protocols within sequential-sampling models.
• Tested biases in abstract numerical evidence accumulation (Experiment 1).
• Tested biases in low-level perceptual evidence accumulation (Experiment 2).

Main Results:

• Found robust choice-induced biases in evidence accumulation for both numerical and perceptual tasks.
• Biases deteriorated numerical estimations and reduced decision revision likelihood.
• Computational modeling indicated choices reduce sensitivity to evidence via gain modulation.

Conclusions:

• Categorical choices fundamentally alter the evidence accumulation mechanism itself.
• Decision-makers become less sensitive to new information after a choice.
• These findings challenge existing models of decision-making and rationality.