I am a postdoctoral fellow in Emory University's Data and Decision Sciences department, where I direct Democracy Lab and the development of the Democracy Viewer.

I am pioneering the emerging field of "Experimental Computational Humanities" (ECH), a design-based research paradigm that integrates digital humanities, experimental humanities, and computer science to foreground experimentation and tool-building as modes of knowledge production. ECH recognizes that answering challenging questions requires lots of perspectives, which is why researching across traditional academic boundaries to learn from each other is essential, as is a willingness to experiment across both the arts and the sciences, guided by an open-minded spirit of possibility and discovery.

Design-based humanities research is especially productive in digital history. Because tools function as epistemic instruments, the design of digital infrastructure shapes which historical questions can be asked and answered, as well as how universities and communities collaborate to build systems that support analysis and knowledge creation.

Using a design-based research approach, my tools have told global history at unprecedented scale, modeling consensus and dissent across Wikipedia’s 355 languages to advance multi-scale analyses of globality, connectivity, and interrelatedness that were previously infeasible. My other project, Democracy Viewer, is a public-history web application that provides citizens and researchers with accessible text-mining tools to examine how democratic discourse has changed over time.

My transdisciplinary scholarship has appeared or is forthcoming with Cambridge University Press, Journal of Early American Studies, Journal of Digital History, Association for the Advancement of Artificial Intelligence (AAAI), and the Institute of Electrical and Electronics Engineers (IEEE). My research has been supported by the National Science Foundation (NSF), the National Endowment for the Humanities (NEH), and the National Institute of Justice (NIJ).

's Picture

Evaluating the Efficacy of LLMs to Emulate Human Personalities for Video Game Play

Evaluating the Efficacy of LLMs to Emulate Human Personalities for Video Game Play

To improve the realism of affective Non-Player Characters (NPCs) in video games, this study investigates whether Large Language Models (LLMs) can emulate human personalities. Using the Big Five framework and over 50,000 responses from the International Personality Item Pool (IPIP), LLMs were prompted with self-assessment items corresponding to various personality profiles. Their outputs were then compared to human baseline responses to evaluate accuracy and consistency. Results showed that while some local models exhibited no alignment with human profiles, certain frontier models achieved high alignment. These findings suggest that LLMs can provide a method for designing NPCs with more realistic, personality-driven behavior.