Repository for the Winter 2026 Computational Social Science Workshop
Time: 11:00 AM to 12:20 PM, Thursdays Location: Room 295, 1155 E. 60th St
Damián Blasi is an ICREA Research Professor based at the Center for Brain & Cognition at the Pompeu Fabra University in Barcelona (Spain). He is also an external researcher at the Human Relations Area Files (Yale University, USA) and an associate of the Culture, Cognition, Coevolution Lab based at the Department of Human Evolutionary Biology at Harvard University.
His research centers on the diversity and evolution of languages. He wants to understand where the ~7,500 languages extant today come from (with a special emphasis in the last 12,000 years, the Holocene), how they will change in the advent of the human-machine era, and what is that languages have done to our species, our cognitions, behaviors, and cultures. He fully embraces a transdisciplinary and question-guided approach, drawing from data science, human biology, cognitive sciences, comparative linguistics, evolutionary anthropology, computational social sciences, natural language processing, and cultural evolution. A substantial proportion of his work involves inferences with small, sparse, incomplete, imbalanced, noisy and non-independent observational data.
LLM as theories of human language: the view from linguistic diversity
The prowess of LLMs to acquire complex linguistic representations has inspired a new wave of LLM-centric theories of human language. In this view, the inductive biases LLMs bring to play are not a mere artifact of the model’s architecture but a substantive claim about the very nature of possible and probable languages, such that those characteristics that are easier to LLM-model are also more represented across the world’s languages. In this presentation, I will show that - in spite of the intuitive nature of the claim - the role of functional forces in shaping linguistic diversity is, at best, modest.
James Evans is the Max Palevsky Professor of Sociology, Director of Knowledge Lab and Founding Faculty Co-Director of Chicago Center Computational Social Science at the University of Chicago, the Santa Fe Institute, and Google. Evans' research uses large-scale data, machine learning and generative models to understand how collectives of humans and machines think and what they (can) know. This involves inquiry into the emergence of ideas, shared patterns of reasoning, and processes of attention, communication, agreement, and certainty. Thinking and knowing collectives like science, the Web, and civilization as a whole involve complex networks of diverse human and machine intelligences, collaborating and competing to achieve overlapping aims. Evans' work connects the interaction of these agents with the knowledge they produce and its value for themselves and the system. His work is supported by numerous federal agencies (NSF, NIH, DOD), foundations and philanthropies, has been published in Nature, Science, PNAS, and top social and computer science outlets, and has been covered by global news outlets from the Economist, the Atlantic, and the New York Times to Le Monde, El Pais, and Die Zeit.
Scientific Misinterpretation in Policy-making
Evidence-based policymaking depends on the accurate interpretation of scientific research, yet partisan polarization has intensified concerns about the distortion of knowledge in pursuit of ideological agendas. Drawing on a large corpus of U.S. policy documents and the scientific articles they cite, we use a language-model approach to assess whether policy actors faithfully represent scientific findings. We find that think tanks are nearly twice as likely to misinterpret science as governments or intergovernmental organizations. These misinterpretations are strategically aligned with think tanks' policy positions, disproportionately target high-impact journals (impact factor ≥ 20), and are subsequently more likely to be cited by government documents. These dynamics show the structure of information laundering - the strategic diffusion of misinterpretations through long chains of indirect citation that embed distorted science into policy discourse. Replication with an embedding-based analysis of a larger sample confirms these patterns at scale. We show how the current Trump administration has "cut out the middleman", stopped citing think-tanks, and begun distorting science at think-tank levels for policy support, creating new complications of legitimacy. The findings highlight the distinctive role of partisan think tanks in shaping how misrepresented science circulates and gains legitimacy in evidence-based governance.
John Horton is the Chrysler Associate Professor of Management and an Associate Professor of Information Technologies at the MIT Sloan School of Management.
Horton's research is primarily focused on issues in information systems, market design, labor economics and organizational economics, particularly in the context of online markets. He is also interested in the effects of AI on labor markets and the potential of AI to improve social science methodology.
After completing his PhD and prior to joining NYU Stern School of Business in 2013, he served for two years as the staff economist for oDesk, an online labor market.
Horton received a BS in mathematics from the United States Military Academy at West Point and a PhD in public policy from Harvard University.
Thoughts & Hands-on Experiments with Automated Social Science
Reading List
- Why AI social simulations will be transformative
- General Social Agents
- GitHub Repo - Expected Parrot Domain-Specific Language (EDSL)
Madalina Vlasceanu is an Assistant Professor of Environmental Behavioral Sciences in the Department of Environmental Social Sciences at Stanford University’s Doerr School of Sustainability and the Director of the Climate Cognition Lab. Her research focuses on the cognitive and social processes that give rise to emergent phenomena such as collective beliefs, collective decision-making, and collective action, with direct applications to climate policy. Guided by a theoretical framework of investigation, her research employs a large array of methods including behavioral laboratory experiments, social network analysis, field studies, randomized controlled trials, megastudies, and international many-lab collaborations, with the goal of understanding the processes underlying climate awareness and action at the individual, collective, and system level.
She obtained a PhD in Psychology and Neuroscience from Princeton University in 2021 and a BA in Psychology and Economics from the University of Rochester in 2016. Prior to Stanford, she was an Assistant Professor of Psychology at New York University.
Behavioral Interventions Increasing Climate Awareness and Action at the Individual, Collective, and System Level
Given the urgency of climate change, a rapidly growing body of research across the behavioral sciences has tested interventions aimed at stimulating pro-climate beliefs and behaviors. Here, we propose a framework conceptualizing this body of work at three levels of analysis, ranging from individual cognition to collective action and systemic change. At the individual level, interventions primarily target cognitive or affective processes to increase climate beliefs and stimulate pro-environmental behaviors. Effective interventions at this level include the decreasing of spatial, temporal, and social psychological distance of climate change. At the collective level, interventions aim to stimulate climate advocacy and civic engagement, overcoming social or political barriers to climate mitigation. Promising interventions at this level include emphasizing the efficacy and emotional benefits of collective action. And at the systemic level, climate action can be facilitated, accelerated, and scaled, through structural interventions leveraging policy innovations, infrastructure development, algorithmic deployment, entertainment outlets, or educational tools. Incorporating insights across the individual collective and system levels through interdisciplinary and intersectoral collaborations stands to maximize the behavioral sciences’ contributions to the climate crisis response.
Mina Lee is an Assistant Professor in the Computer Science and Data Science Institute at the University of Chicago. Previously, she was a postdoctoral researcher in the Computational Social Science group at Microsoft Research. She received her Ph.D. in Computer Science from Stanford University in 2023. Her research is at the intersection of natural language processing (NLP) and human-computer interaction (HCI).
Her research group studies the evolving relationship between humans and AI, with a special focus on Writing, Reading, and Thinking with AI. Concretely, they design and evaluate AI systems (e.g., autocomplete system), identify opportunities and risks of AI-assisted writing, reading, and thinking (e.g., design space for writing assistants), and assess AI’s impact through user studies, controlled experiments, and surveys (e.g., reading comprehension). They also examine broader implications, such as how AI may reshape social norms around authorship (e.g., disclosure of AI use), transform education, and influence everyday communication.
Writing with AI: Capturing Its Influence, Designing Its Future
How is AI changing what we write, how we write, and who we are as writers? In this talk, I will first introduce CoAuthor, a platform that records keystroke-level human-AI interactions, and show how we can use CoAuthor to analyze AI’s effects on language, ideation, collaboration, and beyond. Second, I will present a design space of AI writing assistants derived from a systematic review of over 100 papers, highlighting potential trade-offs, alternative design choices, and gaps in current research. Finally, I will share ongoing projects and invite an open discussion on the future of writing with AI.
Reading List
- CoAuthor: Designing a Human-AI Collaborative Writing Dataset for Exploring Language Model Capabilities
- A Design Space for Intelligent and Interactive Writing Assistants
Fiery Cushman is Professor of Psychology at Harvard University, where he directs the Moral Psychology Research Laboratory. His research aims to organize the astonishing complexity of moral judgment around basic functional principles. Much of it is motivated by a simple idea: Because we use punishments and rewards to modify others’ behavior, one function of morality is to teach others how to behave, while another complementary function is to learn appropriate patterns of behavior. His laboratory investigates these issues using a wide range of methods, including surveys, laboratory behavioral studies, psychophysiology, infant and child research, functional neuroimaging, economic games and formal modeling. The ultimate goal is to use the moral domain to understand phenomena of more general importance: the balance between learned and innate contributions to cognition; the human capacity to explain, predict and evaluate others’ behavior; the relationship between automaticity and control; and the architecture of learning and decision-making in a social context. He received his BA and PhD from Harvard University, where he also completed a post-doctoral fellowship.
Cognitive Foundations of Contractualist Morality
Among scientific theories that attempt to explain the function of morality, a common theme is that morality helps people with disparate interests find mutually beneficial arrangements—the kinds of agreements that they would agree to in negotiation. The same theme arises in contractualist philosophical theories of morality. If these theories are right, what cognitive mechanisms help guide our moral judgments towards mutually beneficial agreements? After all, we cannot literally sit down and bargain over every moral norm. I will present our recent research on several cognitively efficient mechanisms that approximate the outcome of actual idealized bargaining, and that inform our moral judgments.
Reading List