Ockham's razor cuts to the root: Simplicity in causal explanation

Area of Science:

• Cognitive Psychology
• Decision Making
• Causal Inference

Background:

• Simpler causal explanations are generally preferred.
• However, how people assess simplicity and the consequences of this preference are not well understood.
• Two metrics for causal explanation simplicity are node simplicity and root simplicity.

Purpose of the Study:

• To investigate how individuals assess simplicity in causal explanations.
• To determine the consequences of simplicity preferences on judgment and memory.
• To differentiate between node simplicity and root simplicity as metrics for explanation preference.

Main Methods:

• Four experiments were conducted to evaluate explanatory preferences.
• Participants' preferences were assessed in relation to node simplicity (number of causes) and root simplicity (number of unexplained causes).
• The influence of probabilistic evidence and root cause strength was examined.

Main Results:

• Explanatory preferences were found to align with root simplicity, not node simplicity.
• A preference for root simplicity was reduced, but not eliminated, by probabilistic evidence favoring more complex explanations.
• Adherence to simpler explanations led to memory distortions for past observations.

Conclusions:

• Individuals prioritize explanations with fewer unexplained root causes.
• This preference for root simplicity is robust but can be modulated by evidence and cause-effect strength.
• Root simplicity preference likely supports efficient prediction and intervention by highlighting key causal information.