Bayesian generic priors for causal learning.

Abstract

The article presents a Bayesian model of causal learning that incorporates generic priors-systematic assumptions about abstract properties of a system of cause-effect relations. The proposed generic priors for causal learning favor sparse and strong (SS) causes-causes that are few in number and high in their individual powers to produce or prevent effects. The SS power model couples these generic priors with a causal generating function based on the assumption that unobservable causal influences on an effect operate independently (P. W. Cheng, 1997). The authors tested this and other Bayesian models, as well as leading nonnormative models, by fitting multiple data sets in which several parameters were varied parametrically across multiple types of judgments. The SS power model accounted for data concerning judgments of both causal strength and causal structure (whether a causal link exists). The model explains why human judgments of causal structure (relative to a Bayesian model lacking these generic priors) are influenced more by causal power and the base rate of the effect and less by sample size. Broader implications of the Bayesian framework for human learning are discussed.

Analogical and category-based inference: A theoretical integration with Bayesian causal models.

“…Although alternative priors could be considered (Lu et al, 2008), uniform priors allow us to keep the model parameter-free. 3 The basic theory can readily be extended to situations in which the reasoner begins with specific priors about the source or target.…”

“…This probability can be directly computed with the Bayesian extension of the power PC theory (Cheng, 1997;Griffiths & Tenenbaum, 2005;Lu et al, 2008). Additional mathematical details are presented in Appendix A.…”

“…These are the cases for which the Bayesian power PC theory of causal learning and inference (Cheng, 1997;Lu et al, 2008) has been most firmly established. Space precludes a full review of proposed models of causal judgments.…”

“…We describe a computational theory of inference that is tightly coupled with the framework provided by causal models (Cheng, 1997;Griffiths & Tenenbaum, 2005;Lu et al, 2008;Waldmann & Holyoak, 1992;Waldmann & Martignon, 1998). In order to definitively differentiate the meaning of causal relations from their syntactic form, we consider not only generative causes (those that make their effect happen) but also preventive causes (those that stop their effect from happening).…”

“…Recent computational studies have developed detailed models that estimate probability distributions of causal strength by combining priors and observations (Griffiths & Tenenbaum, 2005;Lu et al, 2008).…”