Fact-checking information from large language models can decrease headline discernment

Matthew R. DeVerna, Harry Yaojun Yan, Kai-Cheng Yang, and Filippo Menczer
PNAS 121 (50) e2322823121

This study explores how large language models (LLMs) used for fact-checking affect the perception and dissemination of political news headlines. Despite the growing adoption of AI and tests of its ability to counter online misinformation, little is known about how people respond to LLM-driven fact-checking. This experiment reveals that even LLMs that accurately identify false headlines do not necessarily enhance users’ abilities to discern headline accuracy or promote accurate news sharing. LLM fact checks can actually reduce belief in true news wrongly labeled as false and increase belief in dubious headlines when the AI is unsure about an article’s veracity. These findings underscore the need for research on AI fact-checking’s unintended consequences, informing policies to enhance information integrity in the digital age.

Read the full article in PNAS: doi:10.1073/pnas.2322823121

NECSI Winter Session 2025

Explore the world of complexity science with NECSI’s Winter Session 2025. This specially designed course offers an in-depth understanding of complex systems, combining theoretical foundations with real-world applications. Whether you aim to deepen your expertise or begin a new intellectual journey, this program provides a comprehensive and academically rigorous exploration of the field.

Course Dates: 10-21 February 2025

Details at: necsi.edu

Co-evolution of behaviour and beliefs in social dilemmas: estimating material, social, cognitive and cultural determinants

Sergey Gavrilets, Denis Tverskoi, Nianyi Wang, Xiaomin Wang, Juan Ozaita, Boyu Zhang, Angel Sánchez, and Giulia Andrighetto

Evolutionary Human Sciences , Volume 6 , 2024 , e50

Understanding and predicting human cooperative behaviour and belief dynamics remains a major challenge both from the scientific and practical perspectives. Because of the complexity and multiplicity of material, social and cognitive factors involved, both empirical and theoretical work tends to focus only on some snippets of the puzzle. Recently, a mathematical theory has been proposed that integrates material, social and cognitive aspects of behaviour and beliefs dynamics to explain how people make decisions in social dilemmas within heterogeneous groups. Here we apply this theory in two countries, China and Spain, through four long-term behavioural experiments utilising the Common Pool Resources game and the Collective Risk game. Our results show that material considerations carry the smallest weight in decision-making, while personal norms tend to be the most important factor. Empirical and normative expectations have intermediate weight in decision-making. Cognitive dissonance, social projection, logic constraints and cultural background play important roles in both decision-making and beliefs dynamics. At the individual level, we observe differences in the weights that people assign to factors involved in the decision-making and belief updating process. We identify different types of prosociality and rule-following associated with cultural differences, various channels for the effects of messaging, and culturally dependent interactions between sensitivity to messaging and conformity. Our results can put policy and information design on firmer ground, highlighting the need for interventions tailored to the situation at hand and to individual characteristics. Overall, this work demonstrates the theoretical and practical power of the theory in providing a more comprehensive understanding of human behaviour and beliefs.

Read the full article at: www.cambridge.org

The Atlas of Social Complexity, by Brian Castellani and Lasse Gerrits

Embark on a riveting journey through the study of social complexity with The Atlas of Social Complexity. Over three decades of scientific exploration unfold, unravelling the enigmatic threads that compose the fabric of society. From the dance of bacteria, to human-machine interactions, to the ever-shifting dynamics of power in social networks, this Atlas maps the evolution of our understanding of social complexity.

Brian Castellani’s and Lasse Gerrits’ Atlas is not merely retrospective. It is a compass pointing to uncharted territories: new directions for research and intellectual debate. With wit and insight, they invite the reader to ponder unanswered questions, taking them on a quest for alternative ways to understand the intricate complexities of societies.

The Atlas of Social Complexity is a thrilling expedition into the heart of what makes us human: from cognition, emotion, consciousness, the dynamics of human psychology, to social networks, collective behaviour, politics and governance, technology and planning, and the practice of social interventions. The Atlas also visits cross-cutting themes such as intersectionality, configurational complexity, and research methods.

Organised around six transdisciplinary themes and twenty-four topics the Atlas is an invaluable resource for all social science and complexity science scholars and students interested in new ideas and new ways of working in social complexity. It paves the way for the next generation of research in the study of social complexity.

Read the full article at: www.e-elgar.com

Spaces of mathematical chemistry

Guillermo Restrepo

Theory in Biosciences Volume 143, pages 237–251, (2024)

In an effort to expand the domain of mathematical chemistry and inspire research beyond the realms of graph theory and quantum chemistry, we explore five mathematical chemistry spaces and their interconnectedness. These spaces comprise the chemical space, which encompasses substances and reactions; the space of reaction conditions, spanning the physical and chemical aspects involved in chemical reactions; the space of reaction grammars, which encapsulates the rules for creating and breaking chemical bonds; the space of substance properties, covering all documented measurements regarding substances; and the space of substance representations, composed of the various ontologies for characterising substances.

Read the full article at: link.springer.com