Date of Award

2020

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Graduate Group

Communication

First Advisor

John B. Jemmott

Abstract

Segregation between different social groups has been suspected to exacerbate cognitive bias and misperception. However, a lack of empirical evidence has prevented scholars from identifying the causal effects of social segregation. This dissertation aimed to examine the effects of social segregation on information processing based on multiple studies using web-based randomized controlled experiments. Study 1 examined the effects of social segregation on cognitive bias in processing political information. The results showed that people in segregated networks are less likely to perceive that information they exchanged favored or challenged positions of their own social groups. These results indicate the existence of the “network backfire effect”: network interventions that attempt to mitigate social segregation by creating cross-cutting connections can aggravate cognitive bias of individuals. Study 2 investigated how segregation based on racial identities influences people’s ability to distinguish true and false information about health. The results revealed that social segregation reduces the difference between perceived credibility of true information and that of false information. The findings imply the existence of the “network ambiguation effect”: social segregation can undermine individuals’ ability to distinguish falsehood and truth and blur the boundary between the two, amplifying negative consequences of misinformation. Study 3 was designed to test if segregation between different political groups increases perceived credibility of health misinformation. The results indicate that political segregation inflates the credibility of health misinformation, supporting the “spillover effect” of political polarization on health misperception. This dissertation demonstrated that social segregation influences the information processing of individuals in social networks. More research is needed to understand how various aspects of social networks affect information processing and behavior.

Embargoed

Available to all on Tuesday, August 06, 2024

Share

COinS