Social media bots are often treated as a homogeneous and inherently malicious presence in online environments. This thesis argues that these automated accounts actually function as as heterogeneous Cyber Social Agents (CSAs) embedded within socio-technical systems. Their behaviors, roles and impacts emerge through the interactions with other users, the content and the platform algorithms. Drawing on multi-platform datasets that span political events, crises and international discourse, this thesis characterizes a typology of CSAs and demonstrates that CSAs are deployed strategically to influence narratives, disseminate information, and amplify social signals. Contrary to common assumptions, bots and humans exhibit similar levels of moral and emotional language, and their distinguishing characteristics arise primarily through coordination patterns, network positioning, and the systematic invocation of cognitive biases. Using network analysis and coordination metrics, this thesis also shows that bot influence emerges collectively through synchronized activity rather than individual behavior, and by working together, these agents accelerate narrative spread and shape engagement and stance dynamics. Agent-based simulations further demonstrate that automation is not inherently harmful: purpose-designed "useful bots" can mitigate false information dynamics and improve information resilience. To understand the information ecology, this thesis develops computational frameworks for bot detection, archetype-based classification, and large-scale social simulation. By integrating computational methods with sociological theory, this thesis reframes bots as active social actors and provides a unified framework for studying automation, influence, and intervention in modern information ecosystems.

220 pages

Thesis Committee:
Kathleen M. Carley (Chair)
L. RIchard Carley
Nicolas Christin
Melissa Chua (Defense Science and Technology Agency)

Nicolas Christin, Head, Software and Societal Systems Department
Martial Hebert, Dean, School of Computer Science

Return to: SCS Technical Report Collection
School of Computer Science

This page maintained by reports@cs.cmu.edu