Explaining away: significance of priors, diagnostic reasoning and structural complexity
- Alice Liefgreen, Department of Experimental Psychology, University College London, London, United Kingdom
- Marko Tesic, Department of Psychological Sciences, Birbeck, University of London, London, United Kingdom
- David Lagnado, Department of Experimental Psychology, University College London, London, United Kingdom
AbstractRecent research suggests that people do not perform well on some of the most crucial components of causal reasoning: probabilistic independence, diagnostic reasoning, and explaining away. Despite this, it remains unclear what contexts would affect people's reasoning in these domains. In the present study we investigated the influence of manipulating priors of causes and structural complexity of Causal Bayesian Networks (CBNs) on the above components. Overall we found that participants largely accepted the priors and understood probabilistic independence, but engaged in inaccurate diagnostic reasoning and insufficient explaining away behavior. Moreover, the effect of manipulating priors on participants' performance in diagnostic reasoning and explaining away was significantly larger in a structurally less complex CBN than in a structurally more complex CBN.
Return to previous page