Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

The Role of Granularity in Causal Learning

Soo, Kevin, W (2019) The Role of Granularity in Causal Learning. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

Download (1MB) | Preview


Prior experiments on causal learning have typically investigated how people learn about the relationships between binary variables (e.g., patients either take or do not take a drug, and either exhibit or do not exhibit a particular symptom). Such experiments are often oversimplifications of real-world learning contexts, in which people have to learn about relationships between causes and effects of varying granularities (i.e. how many levels a variable has). In this dissertation, I explored how the granularities of a cause and effect influenced peoples’ estimates of the strength of causal relationships. Four experiments were conducted in which participants learned about a cause-effect relationship by observing a cause and effect over multiple trials and making a judgment about the causal strength. On each trial, participants first viewed the state of the cause and predicted the state of the effect. Participants made stronger causal strength judgments when the effect was more coarse-grained, despite the objective causal strength being fixed (Experiment 1). The influence of the effect’s granularity was due to participants perceiving the prediction task as subjectively easier when it involved a coarse-grained effect, and not due to feedback they received for their predictions (Experiment 2). These findings supported the newly proposed feelings-of-success heuristic; I proposed that participants made judgments of objective causal strength by substituting their subjective feelings of how successfully they made predictions of the effect. In support of this hypothesis, participants’ judgments of how successful they were in the prediction task mediated the relationship between the granularity of the effect and their judgments of objective causal strength (Experiment 3). Finally, the influence of the effect’s granularity was attenuated when participants did not make explicit predictions, suggesting that the effect’s granularity influenced causal strength judgments via the subjective feelings associated with the act of prediction (Experiment 4). Collectively, these studies show that while people are generally accurate when estimating causal strength, real-world factors like the granularity of variables can lead to biases in judgments.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Soo, Kevin, Wkevin.soo@pitt.edukws100000-0002-3927-6384
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairRottman, Benjamin, Mrottman@pitt.edu0000-0002-4718-3970
Committee MemberLibertus, Melissa,
Committee MemberNokes-Malach, Timothy,
Committee MemberSchunn, Christian,
Committee MemberWoodward, James,
Date: 25 June 2019
Date Type: Publication
Defense Date: 26 March 2019
Approval Date: 25 June 2019
Submission Date: 8 April 2019
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 143
Institution: University of Pittsburgh
Schools and Programs: Dietrich School of Arts and Sciences > Psychology
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: causal learning; causal reasoning; granularity; prediction
Date Deposited: 25 Jun 2019 21:54
Last Modified: 25 Jun 2019 21:54


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item