Title

Bayesian Principles and Causal Judgment

Date of Award

2007

Document Type

Dissertation

Degree Name

Doctor of Philosophy (Ph.D.)

Department

Psychology/Experimental

First Advisor

Richard Anderson

Abstract

There has been extensive research on the ability to perceive causal and correlational relationships between variables. Recently, Bayesian principles have been brought to bear on this topic. The purpose of the present study was to assess whether people are sensitive to the probabilities of samples with varying mathematical characteristics drawn from correlated versus uncorrelated populations. Also of interest was the formation of judgments of causality, examined using concurrent process tracing techniques, including verbal protocol. Participants were presented with seven samples, three with determinate and four with indeterminate correlations, and were asked to rank them with respect to the likelihood that the sample indicated a positive causal relationship. Participants did show, under some circumstances, sensitivity to the probability that a sample had been drawn from a correlated versus an uncorrelated population, supporting the hypothesis that people take an inferential approach to causal judgment.