When similarity and causality compete in category-based property generalization

Area of Science:

• Cognitive Psychology
• Concept Learning
• Causal Inference

Background:

• Category-based generalization is a fundamental cognitive process.
• Similarity and typicality traditionally influence how people generalize properties.
• The role of causal explanations in this process is less understood.

Purpose of the Study:

• To investigate how causal explanations affect category-based generalization of novel properties.
• To examine the interaction between causal explanations and similarity-based generalization.
• To determine if category-based property generalization follows principles of causal inference.

Main Methods:

• Five experiments were conducted.
• Participants generalized novel properties attributed to existing category features via causal explanations.
• The influence of typicality, diversity, and similarity on generalization was tested with and without causal explanations.

Main Results:

• Causal explanations largely eliminated the effects of typicality, diversity, and similarity on generalization.
• Generalizations showed properties of causal reasoning, where effects were judged more prevalent when their causes were prevalent.
• Novel properties were generalized based on causal links rather than solely on similarity.

Conclusions:

• Category-based property generalization is frequently driven by causal inference.
• Causal explanations override traditional similarity-based heuristics in generalization.
• Understanding generalization requires considering causal reasoning mechanisms.