Rethinking fast and slow based on a critique of reaction-time reverse inference.

Area of Science:

• Cognitive Psychology
• Neuroscience
• Behavioral Economics

Background:

• Dual-process theories propose intuitive vs. deliberative decision-making.
• Reaction time (RT) is often used to infer intuitive choices in research.
• Reverse inference from behavioral data to mental processes is methodologically challenging.

Purpose of the Study:

• To investigate whether reaction time (RT) differences reliably indicate intuitive choices.
• To assess the validity of using RT as a proxy for intuitive decision-making.
• To examine the influence of choice discriminability on RT-selfishness correlations.

Main Methods:

• Analysis of two datasets from value-based choice experiments.
• Controlling for discriminability (strength of preference) in choice options.
• Utilizing variations of a standard value-based choice paradigm.

Main Results:

• No systematic RT differences were found between choice types when discriminability was controlled.
• Correlations between RT and selfishness could be predictably replicated, eliminated, or reversed.
• Findings challenge the interpretation of RT as a direct measure of intuitive preference.

Conclusions:

• RT differences alone are insufficient evidence for dual-process accounts of decision-making.
• The use of RT in reverse inference for mental processes requires careful consideration of confounding factors like discriminability.
• Methodological rigor is essential when inferring cognitive processes from behavioral measures.