Detecting how topological, social, and ontological hierarchies are reflected in human behavior Xiaohuan Xia

Format:

Book

Thesis/Dissertation

Author/Creator:

Xia, Xiaohuan, author.

Contributor:

University of Pennsylvania. Bioengineering., degree granting institution.

Language:

English

Subjects (All):

Social psychology.

Sociology.

Behavioral psychology.

0451.

0626.

0384.

Local Subjects:

Social psychology.

Sociology.

Behavioral psychology.

0451.

0626.

0384.

Genre:

Academic theses

Physical Description:

1 online resource (175 pages)

Contained In:

Dissertations Abstracts International 87-12B

Place of Publication:

Ann Arbor : ProQuest Dissertations and Theses, 2026

Language Note:

English

Summary:

Humans are a very complex species. Individually, a person can display a wide range of behaviors. Collectively, sociocultural and other hierarchies in the environment shape these humans to produce behavior that can be highly heterogeneous at the individual level but regular and patterned at the collective level. Thus, detecting how different hierarchies are reflected in human behavior at individual and collective level can give us insights into how the hierarchical environment can shape what aspects of human behavior in what ways.To study how complex human behavior interacts with hierarchies at an individual and a collective scales, one may leverage multidisciplinary research and network science. Multidisciplinary research can develop new knowledge and ideas through combining the skills and perspectives from multiple disciplines. This enables tackling some problems of interactions between hierarchies and human behavior with novel problem-solving strategies that leverage the strengths of several fields. Network science, itself a multidisciplinary field, studies interacting entities in a system by modeling the elements with nodes and the interactions between elements with edges. In the context of studying human behavior at a collective scale, networks can be used to model a group of individuals and to then extract insightful measures based on the connectivity among these individuals. Together with multidisciplinary tools and findings, one may extract from these interacting individuals on a network valuable insights that otherwise easily stay hidden from a single disciplinary viewpoint.In this dissertation, we investigated how three types of hierarchies were reflected in human behavior using multidisciplinary tools and insights across scales. Chapter 2 is an experimental study using a statistical mechanics-inspired computational memory model with simulations to predict and explain how the learning of hierarchical structure in transition networks can be reflected in reaction times in a motor response task. Going from individual level to collective level analysis, Chapter 3 utilized large language models that we validated to measure the sentiment in researchers' writing when they cited each other. We then used a social psychology theory to explain how the observed sentiment on a collective level tracked sociocultural and scientific hierarchies of the writers. Building on our prior work studying how sociocultural hierarchies were reflected in citation sentiment, Chapter 4 computationally captured an ontological hierarchy-changing behavior, anthropomorphism, and explained how this hierarchy-altering behavior can reflect sociocultural hierarchies of the writers. The study was grounded in multiple individual-level psychological theories to explain anthropomorphism tendencies collectively using factors along a sociocultural hierarchy: team size, collaboration frequency, and collectivism.Together, this body of work leveraged multiple disciplines and theories to detect and understand how individual-level and collective-level human behavior can reflect topological, social, and ontological hierarchies

Notes:

Source: Dissertations Abstracts International, Volume: 87-12, Section: B.

Advisors: Bassett, Dani S. Committee members: Lydon-Staley, David; Stocker, Alan A.; Daniilidis, Kostas

Ph.D. University of Pennsylvania 2026

Vendor supplied data

Local Notes:

School code: 0175

ISBN:

9798247972907

Access Restriction:

Restricted for use by site license