Implicit learning mediates base rate acquisition in perceptual categorization.

Area of Science:

• Cognitive Psychology
• Perceptual Learning
• Machine Learning

Background:

• Base-rate neglect is a common cognitive bias.
• Implicit learning may play a role in how base-rate information influences category knowledge.
• The COVIS theory of categorization posits an implicit learning system.

Purpose of the Study:

• To investigate whether implicit learning mediates the influence of base-rates on category knowledge acquired through direct experience.
• To examine how different training conditions affect base-rate influence.
• To test the role of implicit learning in base-rate sensitivity.

Main Methods:

• Two experiments were conducted using simple perceptual categories with unequal base-rates.
• Experiment 1 involved response training versus observational training.
• Experiment 2 involved immediate versus delayed feedback during response training.
• Signal detection theory (β values) and decision-bound modeling were used for analysis.

Main Results:

• Base-rate influence was weaker under conditions that disrupt implicit learning (observational training, delayed feedback).
• This effect was more pronounced when category discriminability was low.
• Disrupting implicit learning attenuated the base-rate effect.

Conclusions:

• Implicit learning underlies the influence of base-rates on category knowledge acquired through direct experience.
• The implicit learning system within the COVIS theory may be involved in developing sensitivity to category base-rates.
• Findings support the role of implicit processes in base-rate utilization during category learning.