Analogical Inferences in Causal Systems

Abstract

Analogical and causal reasoning theories both seek to explain patterns of inductive inference. Researchers have claimed that reasoning scenarios incorporating aspects of both analogical comparison and causal thinking necessitate a new model of inductive inference (Holyoak, Lee, & Lu, 2010; Lee & Holyoak, 2008). This paper takes an opposing position, arguing that features of analogical models make correct claims about inference patterns found among causal analogies, including analogies with both generative and preventative relations. Experiment 1 demonstrates that analogical inferences for these kinds of causal systems can be explained by alignment of relational structure, including higher-order relations. Experiment 2 further demonstrates that inferences strengthened by matching higher-order relations are not guided by the transfer of probabilistic information about a cause from base to target. We conclude that causal analogies behave like analogies in general—analogical mapping provides candidate inferences which can then be reasoned about in the target.

4 Figures and Tables

Cite this paper

@inproceedings{Myers2017AnalogicalII, title={Analogical Inferences in Causal Systems}, author={Matthew Myers}, year={2017} }