Generative Social Science: Studies in Agent-Based Computational Modeling

Eric Jones

DOI: https://doi.org/10.5038/2162-4593.11.1.8

2007-01-01

Journal of Ecological Anthropology

Abstract:By Eric C. Jones, Published on 01/01/07

What problem does this paper attempt to address?

Agent-based computational models and generative social science

J. Epstein

DOI: https://doi.org/10.1002/(SICI)1099-0526(199905/06)4:5%3C41::AID-CPLX9%3E3.0.CO;2-F

1999-05-01

Abstract:This article argues that the agent-based computational model permits a distinctive approach to social science for which the term “generative” is suitable. In defending this terminology, features distinguishing the approach from both “inductive” and “deductive” science are given. Then, the following specific contributions to social science are discussed: The agent-based computational model is a new tool for empirical research. It offers a natural environment for the study of connectionist phenomena in social science. Agent-based modeling provides a powerful way to address certain enduring—and especially interdisciplinary—questions. It allows one to subject certain core theories—such as neoclassical microeconomics—to important types of stress (e.g., the effect of evolving preferences). It permits one to study how rules of individual behavior give rise—or “map up”—to macroscopic regularities and organizations. In turn, one can employ laboratory behavioral research findings to select among competing agent-based (“bottom up”) models. The agent-based approach may well have the important effect of decoupling individual rationality from macroscopic equilibrium and of separating decision science from social science more generally. Agent-based modeling offers powerful new forms of hybrid theoretical-computational work; these are particularly relevant to the study of non-equilibrium systems. The agentbased approach invites the interpretation of society as a distributed computational device, and in turn the interpretation of social dynamics as a type of computation. This interpretation raises important foundational issues in social science—some related to intractability, and some to undecidability proper. Finally, since “emergence” figures prominently in this literature, I take up the connection between agent-based modeling and classical emergentism, criticizing the latter and arguing that the two are incompatible. ! 1999 John Wiley & Sons, Inc.

Sociology,Computer Science
Computational Social Science: Agent-based social simulation

N. Gilbert

Abstract:Abstract versus Descriptive . Models can vary in the degree to which they attempt to incorporate the detail of particular targets. An example of a model which aims to be a detailed representation of a specific location and the developments there is the work by Dean et al. (1999) on the Long House Valley, in northern Arizona near Monument Valley. The model covers a time from about A.D. 400 to 1400 and consists of agent households that inhabit a digitized version of the Long House Valley landscape. Agents have rules for determining their agricultural practices and residential locations, as well as for reproduction and mortality. Each run of the model generates a unique history of population, agricultural output, and settlement patterns which can be compared with archaeological evidence from the Valley. In contrast, a series of papers (Conte and Castelfranchi 1995, Castelfranchi, Conte and Paolucci 1998, Saam and Harrer 1999, Staller and Petta 2001, Flenthe, Polani and Uthman 2001, Hales 2002, Younger 2004) has explored the relationship between norms and social inequality using a very simple ‘game’ in which agents controlled by ‘norms’ (i.e. behavioural rules) search a regular grid for ‘food’, which they consume to maintain their energy level or ‘strength’. These authors carried out experiments to study the mean and variance of the distribution of strength under various normative arrangements. Two examples are ‘blind aggression’, when agents attack other agents to grab their food, regardless of whether the attacked agent is stronger than they are, and ‘finders-keeper’, when agents respect ‘property rights’ and do not attack other

Sociology,Computer Science,Geography
On agent-based modeling and computational social science

Rosaria Conte,Mario Paolucci

DOI: https://doi.org/10.3389/fpsyg.2014.00668

IF: 3.8

2014-07-14

Frontiers in Psychology

Abstract:In the first part of the paper, the field of agent-based modeling (ABM) is discussed focusing on the role of generative theories, aiming at explaining phenomena by growing them. After a brief analysis of the major strengths of the field some crucial weaknesses are analyzed. In particular, the generative power of ABM is found to have been underexploited, as the pressure for simple recipes has prevailed and shadowed the application of rich cognitive models. In the second part of the paper, the renewal of interest for Computational Social Science (CSS) is focused upon, and several of its variants, such as deductive, generative, and complex CSS, are identified and described. In the concluding remarks, an interdisciplinary variant, which takes after ABM, reconciling it with the quantitative one, is proposed as a fundamental requirement for a new program of the CSS.

psychology, multidisciplinary
Psychologically-Valid Generative Agents: A Novel Approach to Agent-Based Modeling in Social Sciences

Konstantinos Mitsopoulos,Ritwik Bose,Brodie Mather,Archna Bhatia,Kevin Gluck,Bonnie Dorr,Christian Lebiere,Peter Pirolli

DOI: https://doi.org/10.1609/aaaiss.v2i1.27698

2024-01-22

Abstract:Incorporating dynamic realistic human behaviors in population-scale computational models has been challenging. While some efforts have leveraged behavioral theories from social science, validated theories specifically applicable to Agent-based modeling remain limited. Existing approaches lack a comprehensive framework to model the situated, adaptive nature of human cognition and choice. To address these challenges, this paper proposes a novel framework, Psychologically-Valid Generative Agents. These agents consist of a Cognitive Architecture that provides data-driven and cognitively-constrained decision-making functionality, and a Large Language Model that generates human-like linguistic data. In addition, our framework benefits from Stance Detection, a Natural Language Processing technique, that allows highly personalize initialization of the agents, based on real-world data, within Agent-based modeling simulations. This combination provides a flexible yet structured approach to endogenously represent how people perceive, deliberate, and respond to social or other types of complex decision-making dynamics. Previous work has demonstrated promising results by using a subset of the components of our proposed architecture. Our approach has the potential to exhibit highly-realistic human behavior and can be used across a variety of domains (e.g., public health, group dynamics, social and psychological sciences, and financial markets).
Generative Agent-Based Modeling: Unveiling Social System Dynamics through Coupling Mechanistic Models with Generative Artificial Intelligence

Navid Ghaffarzadegan,Aritra Majumdar,Ross Williams,Niyousha Hosseinichimeh

DOI: https://doi.org/10.1002/sdr.1761

2023-09-21

Abstract:We discuss the emerging new opportunity for building feedback-rich computational models of social systems using generative artificial intelligence. Referred to as Generative Agent-Based Models (GABMs), such individual-level models utilize large language models such as ChatGPT to represent human decision-making in social settings. We provide a GABM case in which human behavior can be incorporated in simulation models by coupling a mechanistic model of human interactions with a pre-trained large language model. This is achieved by introducing a simple GABM of social norm diffusion in an organization. For educational purposes, the model is intentionally kept simple. We examine a wide range of scenarios and the sensitivity of the results to several changes in the prompt. We hope the article and the model serve as a guide for building useful diffusion models that include realistic human reasoning and decision-making.

Artificial Intelligence,Machine Learning,Multiagent Systems,Adaptation and Self-Organizing Systems,Physics and Society
Data-driven agent-based modeling in computational social science

Jan Lorenz

DOI: https://doi.org/10.4324/9781003024583-11

2021-11-10

Abstract:With agent-based models, computational social scientists want to explain the emergence of social phenomena, such as ethnic and social segregation, opinion polarization, the rise of mass protests, economic and political cycles, or the ups and downs of epidemic spreading of infectious diseases. While modeling, researchers may experiment with theories and assumptions about the behavior of individuals and link them to the societal level through computer simulation of repeated social interaction of many artificial individuals. Similarly, researchers play with institutional settings modifying the context in which agents interact. Such agent-based models can be seen as a tool to explore multi-player games where actors’ rationality is bounded. In both senses, agent-based modeling is intrinsically theory driven. The strength of agent-based modeling is the quantitative causal understanding of emergence through complex nonlinear dynamics on the macro-level. Therefore, methods of validation with empirical data are less easily specified and more diverse than in variable-based regression models. In particular, the nature of the data–theory link is closely related to the main research purpose and can deviate from validation. Nevertheless, agent-based models are to explain real-world phenomena, and thus a useful model should have a reflection in empirical data and the other way round. This chapter is about how agent-based modeling can be data-driven. This includes (1) the consideration of data structure in the model building process, (2) a parallel data exploration searching and quantifying “stylized facts” to later use for validation, (3) the iteration of model revisions to increase its replicative validity, and (4) the calibration of model parameters with data for forecasting or counterfactual simulations. These aspects of data-driven agent-based modeling will be exemplified by two examples about segregation and polarization. This chapter is about how agent-based modeling can be data-driven. This includes: the consideration of data structure in the model building process, a parallel data exploration searching and quantifying “stylized facts” to later use for validation, the iteration of model revisions to increase its replicative validity, and the calibration of model parameters with data for forecasting or counterfactual simulations. Agent-based modeling and simulation are core methods in computational social science, besides making sense of large amounts of human-generated data and studying and shaping how digitization changes human societies. The segregation model’s fundamental insight is that ethnic homophily can trigger heavily segregated towns even when similarity preferences are mild. For societies in the real world, macroscopic transitions at critical thresholds in agent-based models imply that small changes in its population or environmental conditions can trigger substantial qualitative changes. The data-driven way to derive stylized facts is exploratory data analysis, which is the core part of the data science process.
An Introduction to Agent Based Modelling and Simulation of Social Processes

Armano Srbljinovic,Ognjen Skunca

DOI: https://doi.org/10.48550/arXiv.cond-mat/0409312

2004-09-13

Abstract:The paper provides an introduction to agent-based modelling and simulation of social processes. Reader is introduced to the worldview underlying agent-based models, some basic terminology, basic properties of agent-based models, as well as to what one can and what cannot expect from such models, particularly when they are applied to social-scientific investigation. Special attention is given to the issues of validation.

Other Condensed Matter
Epidemic Modeling with Generative Agents

Ross Williams,Niyousha Hosseinichimeh,Aritra Majumdar,Navid Ghaffarzadegan

2023-07-11

Abstract:This study offers a new paradigm of individual-level modeling to address the grand challenge of incorporating human behavior in epidemic models. Using generative artificial intelligence in an agent-based epidemic model, each agent is empowered to make its own reasonings and decisions via connecting to a large language model such as ChatGPT. Through various simulation experiments, we present compelling evidence that generative agents mimic real-world behaviors such as quarantining when sick and self-isolation when cases rise. Collectively, the agents demonstrate patterns akin to multiple waves observed in recent pandemics followed by an endemic period. Moreover, the agents successfully flatten the epidemic curve. This study creates potential to improve dynamic system modeling by offering a way to represent human brain, reasoning, and decision making.

Artificial Intelligence,Multiagent Systems,General Economics,Adaptation and Self-Organizing Systems,Physics and Society
Agent-based computational economics: growing economies from the bottom up

Leigh Tesfatsion

DOI: https://doi.org/10.1162/106454602753694765

Abstract:Agent-based computational economics (ACE) is the computational study of economies modeled as evolving systems of autonomous interacting agents. Thus, ACE is a specialization of economics of the basic complex adaptive systems paradigm. This study outlines the main objectives and defining characteristics of the ACE methodology and discusses similarities and distinctions between ACE and artificial life research. Eight ACE research areas are identified, and a number of publications in each area are highlighted for concrete illustration. Open questions and directions for future ACE research are also considered. The study concludes with a discussion of the potential benefits associated with ACE modeling, as well as some potential difficulties.
Can Agents Spontaneously Form a Society? Introducing a Novel Architecture for Generative Multi-Agents to Elicit Social Emergence

H. Zhang,J. Yin,M. Jiang,C. Su

2024-09-10

Abstract:Generative agents have demonstrated impressive capabilities in specific tasks, but most of these frameworks focus on independent tasks and lack attention to social interactions. We introduce a generative agent architecture called ITCMA-S, which includes a basic framework for individual agents and a framework called LTRHA that supports social interactions among multi-agents. This architecture enables agents to identify and filter out behaviors that are detrimental to social interactions, guiding them to choose more favorable actions. We designed a sandbox environment to simulate the natural evolution of social relationships among multiple identity-less agents for experimental evaluation. The results showed that ITCMA-S performed well on multiple evaluation indicators, demonstrating its ability to actively explore the environment, recognize new agents, and acquire new information through continuous actions and dialogue. Observations show that as agents establish connections with each other, they spontaneously form cliques with internal hierarchies around a selected leader and organize collective activities.

Multiagent Systems,Artificial Intelligence,Human-Computer Interaction,Machine Learning
Advances in the agent-based modeling of economic and social behavior

Mitja Steinbacher,Matthias Raddant,Fariba Karimi,Eva Camacho Cuena,Simone Alfarano,Giulia Iori,Thomas Lux

DOI: https://doi.org/10.1007/s43546-021-00103-3

2021-07-01

Abstract:Abstract In this review we discuss advances in the agent-based modeling of economic and social systems. We show the state of the art of the heuristic design of agents and how behavioral economics and laboratory experiments have improved the modeling of agent behavior. We further discuss how economic networks and social systems can be modeled and we discuss novel methodology and data sources. Lastly, we present an overview of estimation techniques to calibrate and validate agent-based models and show avenues for future research.
The Design Of A Network-Structured Artificial Social System Based On Agent-Based Modeling

Wei Zhou,Zhichuan Huang,Yichuan Jiang

2009-01-01

Abstract:In previous work, in order to simulate the phenomenon of emergence in the social system, researchers have built kinds of artificial social systems. But most of these systems are just by the mathematics models, so it is hard to complete the work when the real system is complex and dynamic. This paper presents a new network-structured artificial social system based on agent-based modeling. We proposed a process of agent-based modeling and described a system in the paper which is designed based on this modeling. In the presented system we took social individuals as autonomous agents. and set random social ranking relations among them, so that, these agents in the network-structure system could take relevant actions according to their social positions, after enough time, it would lead to the phenomenon of community modularity convergence, and that can be proved by our experimental result.
Agent‐based models in sociology

F. Squazzoni,F. Bianchi

DOI: https://doi.org/10.1002/wics.1356

2015-07-01

Abstract:This article looks at 20 years of applications of agent‐based models (ABMs) in sociology and, in particular, their explanatory achievements and methodological insights. These applications have helped sociologists to examine agent interaction in social outcomes and have helped shift analyses away from structural and aggregate factors, to the role of agency. They have improved the realism of the micro‐behavioral foundations of sociological models, by complementing analytic modeling and game theory–inspired analyses. Secondly, they have helped us to dissect the role of social structures in constraining individual behavior more precisely than in variable‐based sociology. Finally, simulation outcomes have given us a more dynamic view of the interplay between individual behavior and social structures, thus promoting a more evolutionary and process‐based approach to social facts. Attention here has been paid to social norms, social influence, and culture dynamics, across different disciplines such as behavioral sciences, complexity science, sociology, and economics. We argue that these applications can help sociology to achieve more rigorous research standards, by promoting a modeling environment and providing tighter cross‐disciplinary integration. Recently, certain methodological improvements toward model standardization, replication, and validation have been achieved. As a result, the impact of these models in sociology is expected to grow even more in the future. WIREs Comput Stat 2015, 7:284–306. doi: 10.1002/wics.1356

Sociology
Generate Descriptive Social Networks for Large Populations from Available Observations: A Novel Methodology and a Generator

Samuel Thiriot

DOI: https://doi.org/10.48550/arXiv.2003.02213

2020-03-05

Abstract:When modeling a social dynamics with an agent-oriented approach, researchers have to describe the structure of interactions within the population. Given the intractability of extensive network collecting, they rely on random network generators that are supposed to explore the space of plausible networks. We first identify the needs of modelers, including placing heterogeneous agents on the network given their attributes and differentiating the various types of social links that lead to different interactions. We point out the existence of data in the form of scattered statistics and qualitative observations, that should be used to parameter the generator. We propose a new approach peculiar to agent-based modeling, in which we will generate social links from individuals' observed attributes, and return them as a multiplex network. Interdependencies between socioeconomic attributes, and generative rules, are encoded as Bayesian networks. A methodology guides modelers through the formalization of these parameters. This approach is illustrated by describing the structure of interactions that supports diffusion of contraceptive solutions in rural Kenya.

Social and Information Networks,Applications
Social Dynamics of Science

Xiaoling Sun,Jasleen Kaur,Staša Milojević,Alessandro Flammini,Filippo Menczer

DOI: https://doi.org/10.1038/srep01069

2012-09-22

Abstract:The birth and decline of disciplines are critical to science and society. However, no quantitative model to date allows us to validate competing theories of whether the emergence of scientific disciplines drives or follows the formation of social communities of scholars. Here we propose an agent-based model based on a \emph{social dynamics of science,} in which the evolution of disciplines is guided mainly by the social interactions among scientists. We find that such a social theory can account for a number of stylized facts about the relationships between disciplines, authors, and publications. These results provide strong quantitative support for the key role of social interactions in shaping the dynamics of science. A "science of science" must gauge the role of exogenous events, such as scientific discoveries and technological advances, against this purely social baseline.

Physics and Society,Digital Libraries,Social and Information Networks
Agent-Based Models in Social Physics

Le Anh Quang,Nam Jung,Eun Sung Cho,Jae Hwan Choi,Jae Woo Lee

DOI: https://doi.org/10.3938/jkps.72.1272

2018-06-12

Abstract:We review the agent-based models (ABM) on social physics including econophysics. The ABM consists of agent, system space, and external environment. The agent is autonomous and decides his/her behavior by interacting with the neighbors or the external environment with the rules of behavior. Agents are irrational because they have only limited information when they make decisions. They adapt using learning from past memories. Agents have various attributes and are heterogeneous. ABM is a non-equilibrium complex system that exhibits various emergence phenomena. The social complexity ABM describes human behavioral characteristics. In ABMs of econophysics, we introduce the Sugarscape model and the artificial market models. We review minority games and majority games in ABMs of game theory. Social flow ABM introduces crowding, evacuation, traffic congestion, and pedestrian dynamics. We also review ABM for opinion dynamics and voter model. We discuss features and advantages and disadvantages of Netlogo, Repast, Swarm, and Mason, which are representative platforms for implementing ABM.

Physics and Society
Generative Agent Simulations of 1,000 People

Joon Sung Park,Carolyn Q. Zou,Aaron Shaw,Benjamin Mako Hill,Carrie Cai,Meredith Ringel Morris,Robb Willer,Percy Liang,Michael S. Bernstein

2024-11-15

Abstract:The promise of human behavioral simulation--general-purpose computational agents that replicate human behavior across domains--could enable broad applications in policymaking and social science. We present a novel agent architecture that simulates the attitudes and behaviors of 1,052 real individuals--applying large language models to qualitative interviews about their lives, then measuring how well these agents replicate the attitudes and behaviors of the individuals that they represent. The generative agents replicate participants' responses on the General Social Survey 85% as accurately as participants replicate their own answers two weeks later, and perform comparably in predicting personality traits and outcomes in experimental replications. Our architecture reduces accuracy biases across racial and ideological groups compared to agents given demographic descriptions. This work provides a foundation for new tools that can help investigate individual and collective behavior.

Artificial Intelligence,Human-Computer Interaction,Machine Learning
Generative Agents: Interactive Simulacra of Human Behavior

Joon Sung Park,Joseph C. O'Brien,Carrie J. Cai,Meredith Ringel Morris,Percy Liang,Michael S. Bernstein

2023-08-06

Abstract:Believable proxies of human behavior can empower interactive applications ranging from immersive environments to rehearsal spaces for interpersonal communication to prototyping tools. In this paper, we introduce generative agents--computational software agents that simulate believable human behavior. Generative agents wake up, cook breakfast, and head to work; artists paint, while authors write; they form opinions, notice each other, and initiate conversations; they remember and reflect on days past as they plan the next day. To enable generative agents, we describe an architecture that extends a large language model to store a complete record of the agent's experiences using natural language, synthesize those memories over time into higher-level reflections, and retrieve them dynamically to plan behavior. We instantiate generative agents to populate an interactive sandbox environment inspired by The Sims, where end users can interact with a small town of twenty five agents using natural language. In an evaluation, these generative agents produce believable individual and emergent social behaviors: for example, starting with only a single user-specified notion that one agent wants to throw a Valentine's Day party, the agents autonomously spread invitations to the party over the next two days, make new acquaintances, ask each other out on dates to the party, and coordinate to show up for the party together at the right time. We demonstrate through ablation that the components of our agent architecture--observation, planning, and reflection--each contribute critically to the believability of agent behavior. By fusing large language models with computational, interactive agents, this work introduces architectural and interaction patterns for enabling believable simulations of human behavior.

Human-Computer Interaction,Artificial Intelligence,Machine Learning
A Simple Generative Model of Collective Online Behaviour

James P. Gleeson,Davide Cellai,Jukka-Pekka Onnela,Mason A. Porter,Felix Reed-Tsochas

DOI: https://doi.org/10.1073/pnas.1313895111

2014-05-29

Abstract:Human activities increasingly take place in online environments, providing novel opportunities for relating individual behaviours to population-level outcomes. In this paper, we introduce a simple generative model for the collective behaviour of millions of social networking site users who are deciding between different software applications. Our model incorporates two distinct components: one is associated with recent decisions of users, and the other reflects the cumulative popularity of each application. Importantly, although various combinations of the two mechanisms yield long-time behaviour that is consistent with data, the only models that reproduce the observed temporal dynamics are those that strongly emphasize the recent popularity of applications over their cumulative popularity. This demonstrates---even when using purely observational data without experimental design---that temporal data-driven modelling can effectively distinguish between competing microscopic mechanisms, allowing us to uncover new aspects of collective online behaviour.

Physics and Society,Social and Information Networks
Virtual social science

Stefan Thurner

DOI: https://doi.org/10.48550/arXiv.1811.08156

2018-11-20

Abstract:Can we describe social systems quantitatively and predictively, when we know all the actions, interactions, and states of individuals? We interpret human societies as co-evolutionary systems of individuals and their interactions. Based on unique data of a society of computer game players, where all actions and interactions between all players are known, we show that this might indeed be possible. Within this framework we address a number of sociological classics, including formation of social networks, strength of relations, group formation, hierarchical organization, aggression management, gender differences, mobility, and wealth-inequality. We discover behavioral and organizational patterns of the homo sapiens and its society that were not visible with traditional methodology from the social sciences.

Physics and Society

Generative Social Science: Studies in Agent-Based Computational Modeling

Agent-based computational models and generative social science

Computational Social Science: Agent-based social simulation

On agent-based modeling and computational social science

Psychologically-Valid Generative Agents: A Novel Approach to Agent-Based Modeling in Social Sciences

Generative Agent-Based Modeling: Unveiling Social System Dynamics through Coupling Mechanistic Models with Generative Artificial Intelligence

Data-driven agent-based modeling in computational social science

An Introduction to Agent Based Modelling and Simulation of Social Processes

Epidemic Modeling with Generative Agents

Agent-based computational economics: growing economies from the bottom up

Can Agents Spontaneously Form a Society? Introducing a Novel Architecture for Generative Multi-Agents to Elicit Social Emergence

Advances in the agent-based modeling of economic and social behavior

The Design Of A Network-Structured Artificial Social System Based On Agent-Based Modeling

Agent‐based models in sociology

Generate Descriptive Social Networks for Large Populations from Available Observations: A Novel Methodology and a Generator

Social Dynamics of Science

Agent-Based Models in Social Physics

Generative Agent Simulations of 1,000 People

Generative Agents: Interactive Simulacra of Human Behavior

A Simple Generative Model of Collective Online Behaviour

Virtual social science