When learning order affects sensitivity to base rates: challenges for theories of causal learning

Area of Science:

• Cognitive Psychology
• Causal Inference

Background:

• Understanding how people learn causal relationships is crucial in cognitive psychology.
• Base-rate neglect is a common bias in human judgment, where prior probabilities are ignored.

Purpose of the Study:

• To investigate if the direction of learning (predictive vs. diagnostic) affects base-rate use in causal judgments.
• To examine how scenario complexity influences base-rate incorporation in diagnostic reasoning.

Main Methods:

• Three experiments were conducted comparing predictive learning (cause-to-effect) and diagnostic learning (effect-to-cause).
• Participants made diagnostic judgments after learning causal structures of varying complexity.
• Base-rate recall was assessed to rule out encoding failures.

Main Results:

• Learners successfully used base rates in simple causal scenarios, irrespective of learning direction.
• In complex scenarios, base-rate use was observed only after diagnostic learning.
• Predictive learning led to significant base-rate neglect in complex scenarios, despite adequate base-rate encoding.

Conclusions:

• The direction of causal learning significantly impacts base-rate use, particularly in complex situations.
• Diagnostic learning facilitates base-rate integration more effectively than predictive learning under increased complexity.
• Findings challenge existing theories of causal learning and judgment, highlighting the role of learning direction and complexity.