Driven by power? Probe question and presentation format effects on causal judgment.

Area of Science:

• Cognitive Science
• Psychology
• Philosophy of Science

Background:

• Causal power, defined as the probability of a cause producing an effect without other influences, is often assumed to be calculable from covariation data.
• Counterfactual questions are proposed to aid intuitive estimation of causal power by reducing ambiguity.

Purpose of the Study:

• To analyze the effectiveness of counterfactual probe questions in eliciting accurate causal power estimations.
• To investigate the influence of question wording and information presentation on causal judgments.

Main Methods:

• Empirical study analyzing human judgments in response to counterfactual probes.
• Comparison of judgments against normative causal power estimations.
• Evaluation of an alternative Evidence Integration model.

Main Results:

• Judgments elicited by counterfactual probes do not align with normative causal power.
• Causal judgments are significantly influenced by the specific wording of probe questions.
• The presentation format of covariation information also impacts causal judgments.

Conclusions:

• Counterfactual probe questions are inadequate for normative causal power estimation.
• An Evidence Integration model provides a more parsimonious explanation for observed causal judgments.
• Future research should consider alternative models for understanding causal reasoning.