Pith. sign in

REVIEW 1 major objections 4 minor 1 cited by

This paper argues that AI and the social and behavioral sciences have shaped each other for 75 years through a two-part rhythm: quick adoption of each new technology, then slower science-driven refinement.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

AI and social science have co-evolved for 75 years through rapid technological adoption and slower scientific consolidation, with direct human-focused AI studies still scarce.

T0 review reviewed 2026-08-04 challenge →

load-bearing objection Readable, well-sourced historical synthesis; the two-process thesis is plausible but under-tested and needs sharper caveats. the 1 major comments →

arxiv 2510.05743 v3 pith:CSKBIHOJ submitted 2025-10-07 cs.AI cs.CY

Artificially intelligent agents in the social and behavioral sciences: A history and outlook

Petter Holme , Milena Tsvetkova This is my paper
classification cs.AI cs.CY
keywords artificial intelligencesocial sciencebehavioral scienceagent-based modelslarge language modelshistory of computingcyberneticscomputational social science
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper tries to establish that the relationship between artificial intelligence and the social and behavioral sciences is not a one-way import of tools but a long, bidirectional co-evolution. It identifies two recurring processes: technically minded social scientists quickly pick up each new AI breakthrough and design new kinds of studies with it, and then these research lines settle into a slower evolution organized around scientific content. The authors argue that this pattern has repeated from early computer simulations to today's large language models, and that the same technologies are now changing the society being studied. A sympathetic reader would care because this framing corrects the common picture of AI arriving from afar and clarifies why the current wave of generative-AI social science is both promising and epistemically tricky.

Core claim

The central claim is that the history of AI in the social and behavioral sciences is best understood as a two-process dynamic. First, open-minded social and behavioral scientists rapidly adopt any major technical breakthrough in AI and use it to create new types of investigations—this explains why the first social-science Ph.D. based on computer science appeared only about five years after the first programmable computer, and why chatbots were quickly inserted into experiments after 2022. Second, each such burst branches into a slower, content-centered line of research that evolves around the science itself rather than the technology. The paper presents this as a corrective to the view of AI

What carries the argument

The paper's organizing device is the two-process model of scientific change: a fast adoption loop (new AI capability meets open-minded social scientists) followed by a slow content loop (research questions and methods settle around the science). Around this it places a taxonomy of AI roles—AI as analysis tool, as simulation/digital twin, as experimental participant, and as an object or social force—which lets the authors order a 75-year chronology as parallel streams rather than a single timeline.

Load-bearing premise

The narrative rests on the reliability and representativeness of roughly 250 cited secondary sources, including several specific first claims (first social-science Ph.D. based on computer science, first AI program); the paper does not independently verify these historical episodes.

What would settle it

A single well-documented case where a major AI breakthrough was ignored by social scientists for a decade or more, or where a hot line of AI-social research died out without leaving a content-driven branch, would weaken the two-process pattern. Concretely, a systematic bibliometric study showing that leading social-science journals imported AI methods but never exported ideas back to AI—one-way flow—would falsify the bidirectional claim.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • If the pattern holds, today's wave of generative-AI experiments will not be a passing fad: the early chatbot studies will seed a slower, content-driven research program that outlives the hype.
  • The history implies that AI's influence on these sciences has never been one-directional; social-science ideas have shaped AI design, so accounts that treat AI as an external import are incomplete.
  • The current replace-humans-with-chatbots strategy is a low-creativity early phase; the paper predicts more substantive work will emerge as studies move beyond documenting AI capabilities toward understanding human behavior.
  • The same technologies that provide research tools are altering social behavior itself, creating a feedback loop that social scientists must study as part of the subject matter.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • A testable implication of the two-process model: bibliometric traces of citation flows between AI venues and social-science venues should show rapid cross-disciplinary citation bursts after each major capability jump, followed by a settling into within-field citation clusters; this can be checked.
  • The paper's missing-topic observation suggests the highest-value near-term research may be using LLM-agent experiments not to replace human subjects but to generate hypotheses about human behavior that are then tested on humans.
  • If the two-process rhythm is real, the current enthusiasm may follow the arc of earlier waves, and infrastructure for replicability (versioned models, shared prompts, benchmark batteries) will determine whether the content-centered branch thrives or decays.
  • The paper's own admission that it sketches rather than reviews means its chronology should be read as an interpretation; independent archives of early social simulations could confirm or revise the first claims.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

1 major / 4 minor

Summary. This paper surveys the history of artificial intelligence in the social and behavioral sciences, from early computer simulations and cybernetics to contemporary large language models. The authors argue that the relationship has been bidirectional and co-evolutionary, and they propose two main processes: rapid adoption of technical breakthroughs by social and behavioral scientists, followed by slower, content-centered evolution of those research lines. The paper is explicitly non-comprehensive and sketches multiple streams of research, including systems dynamics, symbolic AI and expert systems, connectionism, complexity science, agent-based models, network science, big data, crowdsourcing, and generative AI. The central contribution is an interpretive historical narrative rather than a new empirical result.

Significance. If accepted, the paper provides a valuable corrective to the one-way 'AI as an imported tool' narrative and offers a coherent organizational framework for a disparate literature. It is broad, well-referenced, and clearly structured, with useful schematic figures and an honest statement of scope. The emphasis on bidirectional influence between AI and the social/behavioral sciences is a useful perspective for both historians and practitioners. The main weakness is that the two-process thesis is stated as a general pattern but is supported only by curated examples; the paper does not define its key terms or address counterexamples, so the thesis remains an interpretive gloss rather than a demonstrated historical generalization.

major comments (1)
  1. [Section 1] The central claim is the two-process thesis: 'any technical breakthrough is typically rapidly adopted by open-minded social and behavioral scientists' and then 'branches off to follow a slower-paced evolution centered around the scientific content.' This is a strong universal generalization, but the paper does not define 'rapidly,' 'line of research,' or 'branching,' nor does it provide a systematic basis for the 'typically.' The evidence is a deliberately non-comprehensive selection of mostly successful examples. The paper should either soften the thesis to an interpretive proposal—e.g., 'we highlight two recurring patterns'—or add a discussion of selection criteria and plausible counterexamples (neglected breakthroughs, dead research lines, one-way transfers without feedback). As written, the narrative does not establish the general claim; it is the load-bearing element of the paper's
minor comments (4)
  1. [Section 8.2] The sentence 'Tessler et al., found AI capable of' is incomplete; the finding is missing. Please complete the sentence and ensure the reference is properly cited.
  2. [Section 2] The claim that Hägerstrand's 1953 thesis is 'the first social-science Ph.D. thesis based on computer science' is a strong precedence claim with no corroborating evidence. Since precedents of this kind are difficult to establish, consider replacing 'the first' with 'one of the first' or providing a source that makes the precedence claim.
  3. [Figure 1 caption] Typo: 'Illustation' should be 'Illustration.'
  4. [Section 7.3] Typo: 'nacent' should be 'nascent.' Also, 'social and behavior sciences' appears to be missing 'behavioral' or should be 'social and behavioral sciences.'

Circularity Check

0 steps flagged

No significant circularity: a historical review with no fitted parameters, derivations, or self-cited uniqueness arguments; self-citations appear only as examples of research streams.

full rationale

This is a narrative review, not a derivation. The paper's two-process thesis—that technical breakthroughs are typically rapidly adopted by social/behavioral scientists and then branch into slower, content-centered evolution—is an interpretive historical claim supported by selected episodes. The authors explicitly disclaim comprehensiveness ('we will not give a comprehensive review of the recent literature as other papers do that[59], but we will sketch the current state and trends'), which raises representativeness or cherry-picking concerns, but cherry-picking is not circularity. There are no equations, no fitted parameters, no statistical 'predictions' forced by construction, and no uniqueness theorems invoked to make a model choice appear forced. The self-citations (e.g., Holme & Ghoshal 2006/2009 for networking agents, Tsvetkova et al. 2017/2024 for human-machine networks, Holme 2022 for complexity science, Han et al. 2025 for LLM cooperation) are used as historical examples of ongoing research lines, not as load-bearing evidence for the paper's general thesis. Even where 'first' claims rely on secondary sources and are not independently verified, that is an evidential/correctness issue, not a circular reduction of the argument to its own inputs. Accordingly, no circular step is identified and the score is 0.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

This is a historical review; the central claim depends on the reliability of cited sources and the chosen interpretive framing. There are no fitted parameters or invented entities. The paper's own scope disclaimer (Section 1) limits the evidentiary burden but also prevents it from being a comprehensive or verifiable account.

axioms (2)
  • domain assumption The cited references accurately and representatively describe the historical development of AI in the social and behavioral sciences.
    The paper builds its chronological narrative on secondary sources (e.g., McCorduck 1991 for AI history, Medina 2011 for Cybersyn). If these are inaccurate or the selection is biased, the paper's timeline and thesis are unsupported. No independent verification is provided.
  • ad hoc to paper The five-motivation classification and the two-process thesis are adequate interpretive frameworks for the history.
    The paper imposes a structure on historical material that is not itself derived from the sources; it is a framing choice specific to this paper. If this framing is rejected, the review becomes a collection of anecdotes rather than a coherent thesis.

reviewed 2026-08-04 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Artificially intelligent agents in the social and behavioral sciences: A history and outlook." pith.science (2026). https://pith.science/paper/CSKBIHOJ

@misc{pith2026251005743,
  author       = {Pith},
  title        = {Pith review of: Artificially intelligent agents in the social and behavioral sciences: A history and outlook},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CSKBIHOJ}},
  note         = {Machine review of arXiv:2510.05743}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

We review the historical development and current trends of artificially intelligent agents (agentic AI) in the social and behavioral sciences: from the first programmable computers, and social simulations soon thereafter, to today's experiments with large language models. This overview emphasizes the role of AI in the scientific process and the changes brought about, both through technological advancements and the broader evolution of science from around 1950 to the present. Some of the specific points we cover include: the challenges of presenting the first social simulation studies to a world unaware of computers, the rise of social systems science, intelligent game theoretic agents, the age of big data and the epistemic upheaval in its wake, and the current enthusiasm around applications of generative AI, and many other topics. A pervasive theme is how deeply entwined we are with the technologies we use to understand ourselves.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. When Bots Join the Team: Bot Adoption and the Institutional Fabric of Open-Source Software Projects

    cs.AI 2026-07 conditional novelty 6.0

    Across 2,991 GitHub projects, adopting a first bot is followed at the adoption month by more repeated collaboration, more bot-name references, fewer conflict cascades, and more distinctive output.

Reference graph

Works this paper leans on

249 extracted references · 15 linked inside Pith · cited by 1 Pith paper

  1. [1]

    H. Abbey . An examination of the Reed-Frost theory of epidemics.Hum. Biol., 24:201–233, 1952

    1952

  2. [2]

    G. Aher, R. I. Arriaga, and A. T . Kalai. Using large lan- guage models to simulate multiple humans and repli- cate human subject studies. InProceedings of the 40th International Conference on Machine Learning, ICML’23, page 17. JMLR.org, 2023

    2023

  3. [3]

    Anderson

    C. Anderson. The end of theory: The data deluge makes the scientific method obsolete.Wired, 16(7):16– 17, 2008

    2008

  4. [4]

    L. P . Argyle, E. C. Busby , J. R. Gubler, A. Lyman, J. Ol- cott, J. Pond, and D. Wingate. Testing theories of po- litical persuasion using AI.Proc. Natl. Acad. Sci. USA, 122(18):e2412815122, 2025

    2025

  5. [5]

    K. J. Arrow. The economic implications of learning by doing.Rev. Econ. Stud., 29(3):155–173, 1962

    1962

  6. [6]

    W . B. Arthur. Competing technologies, increasing returns, and lock-in by historical events.Econ. J., 99(394):116–131, 1989

    1989

  7. [7]

    W . B. Arthur.Complexity and the Economy. Oxford Uni- versity Press, Oxford, 2014

    2014

  8. [8]

    W . R. Ashby .Design for a Brain. John Wiley & Sons, New York, 1954

    1954

  9. [9]

    Axelrod.The Evolution of Cooperation

    R. Axelrod.The Evolution of Cooperation. Basic Books, New York, NY, 1984

    1984

  10. [10]

    Axtell and J

    R. Axtell and J. D. Farmer. Agent-based modeling in economics and finance: Past, present, and future. Tech- nical Report 2022-10, Oxford Institute for New Eco- nomic Thinking, 2022

    2022

  11. [11]

    R. L. Axtell, J. M. Epstein, J. S. Dean, G. J. Gumerman, A. C. Swedlund, J. Harburger, S. Chakravarty , R. Ham- mond, J. Parker, and M. Parker. Population growth and collapse in a multiagent model of the Kayenta Anasazi in Long House Valley .Proc. Natl. Acad. Sci. USA, 99:7275–7279, 2002

    2002

  12. [12]

    C. A. Bail. Can generative AI improve social science? Proc. Natl. Acad. Sci. USA, 121(21):e2314021121, 2024

    2024

  13. [13]

    Bala and S

    V . Bala and S. Goyal. A noncooperative model of network formation.Econometrica, 68(5):1181–1229, 2000

    2000

  14. [14]

    Barabási.Network Science

    A.-L. Barabási.Network Science. Cambridge University Press, Cambridge, 2016

    2016

  15. [15]

    Basar, D

    S. Basar, D. Coupland, and H. U. Obrist.The Age of Earthquakes: A Guide to the Extreme Present. Penguin Books, London, 2015

    2015

  16. [16]

    Baudrillard.Simulacra and Simulation

    J. Baudrillard.Simulacra and Simulation. University of Michigan Press, Ann Arbor MI, 1994

    1994

  17. [17]

    Baumann, P

    J. Baumann, P . Röttger, A. Urman, A. Wendsjö, F . M. P . del Arco, J. B. Gruber, and D. Hovy . Large language model hacking: Quantifying the hidden risks of using LLMs for text annotation. Preprint arXiv:2509.08825, 2025

    arXiv 2025

  18. [18]

    Beer.A Platform for Change

    S. Beer.A Platform for Change. John Wiley & Sons, New York, 1975

    1975

  19. [19]

    Beltratti, S

    A. Beltratti, S. Margarita, and P . Terna.Neural Net- works for Economic and Financial Modelling. Interna- tional Thomson Computer Press, London, 1999

    1999

  20. [20]

    V . J. Bermejo, A. Gago, R. H. Gálvez, and N. Harari. LLMs outperform outsourced human coders on com- plex textual analysis. Preprint SSRN/5020034, 2024

    arXiv 2024

  21. [21]

    L. M. A. Bettencourt.Introduction to Urban Science. MIT Press, Cambridge MA, 2022

    2022

  22. [22]

    C. Biever. ChatGPT broke the Turing test: the race is on for new ways to assess AI.Nature, 619:686–689, 2023

    2023

  23. [23]

    M. Binz, E. Akata, M. Bethge, F . Brändle, F . Callaway , J. Coda-Forno, P . Dayan, C. Demircan, M. K. Eckstein, N. Éltet ˝o, T . L. Griffiths, S. Haridi, A. K. Jagadish, L. Ji-An, A. Kipnis, S. Kumar, T . Ludwig, M. Mathony , M. Mattar, A. Modirshanechi, S. S. Nath, J. C. Peter- son, M. Rmus, E. M. Russek, T . Saanum, J. A. Schu- bert, L. M. S. Buschoff,...

    2025

  24. [24]

    M. A. Boden.Mind as Machine: A History of Cognitive Science. Oxford University Press, Oxford, 2006

    2006

  25. [25]

    Boehm.Hierarchy in the Forest: The Evolution of Egalitarian Behavior

    C. Boehm.Hierarchy in the Forest: The Evolution of Egalitarian Behavior. Harvard University Press, Cam- bridge MA, 1999

    1999

  26. [26]

    Bohannon

    J. Bohannon. Social science for pennies.Science, 334(6054):307–307, 2011

    2011

  27. [27]

    Bohannon

    J. Bohannon. Mechanical turk upends social sciences. Science, 352(6291):1263–1264, 2016

    2016

  28. [28]

    D. A. Boiko, R. MacKnight, B. Kline, and G. Gomes. Au- tonomous chemical research with large language mod- els.Nature, 624:570–578, 2023

    2023

  29. [29]

    Boissevain.Friends of Friends: Networks Manipula- tors and Coalitions

    J. Boissevain.Friends of Friends: Networks Manipula- tors and Coalitions. Routledge, Oxford, 1974

    1974

  30. [30]

    Brynjolfsson, D

    E. Brynjolfsson, D. Li, and L. Raymond. Generative AI at work.Q. J. Econ., 140(2):889–942, 2025

    2025

  31. [31]

    J. W . Burton, E. Lopez-Lopez, S. Hechtlinger, Z. Rah- wan, S. Aeschbach, M. A. Bakker, J. A. Becker, A. Berditchevskaia, J. Berger, L. Brinkmann, L. Flek, S. M. Herzog, S. Huang, S. Kapoor, A. Narayanan, A.-M. Nussberger, T . Yasseri, P . Nickl, A. Almaatouq, R. Her- twig, et al. How large language models can reshape collective intelligence.Nat. Hum. Beh...

    2024

  32. [32]

    Carley , J

    K. Carley , J. Kjaer-Hansen, A. Newell, and M. Prietula. Plural-Soar: A prolegomenon to artificial agents and organizational behavior. In M. Masuch and M. Varglien, editors,Artificial Intelligence in Organization and Man- agement Theory, pages 87–118. North-Holland, Ams- terdam, 1992

    1992

  33. [33]

    Carley and A

    K. Carley and A. Newell. The nature of the social agent. J. Math. Sociol., 19(4):221–262, 1994

    1994

  34. [34]

    Charpentier, R

    A. Charpentier, R. Élie, and C. Remlinger. Reinforce- ment learning in economics and finance.Comput. Econ., 62:425–462, 2023

    2023

  35. [35]

    Z. Chu, S. Gianvecchio, H. Wang, and S. Jajodia. Who is tweeting on Twitter: human, bot, or cyborg? In Proceedings of the 26th Annual Computer Security Ap- plications Conference, page 21–30, 2010. Artificially intelligent agents in the social and behavioral sciences HOLME & TSVETKOV A | 14

    2010

  36. [36]

    M. Chui, E. Hazan, R. Roberts, A. Singla, K. Smaje, A. Sukharevsky , L. Yee, and R. Zemmel. The eco- nomic potential of generative AI: The next productivity frontier. Technical report, McKinsey & Company , June 2023

    2023

  37. [37]

    K. M. Colby . Computer simulation of a neurotic process. In S. S. Tompkins and S. Messick, editors,Computer Simulation of Personality. John Wiley & Sons, New York, 1963

    1963

  38. [38]

    J. S. Coleman. Social theory , social research, and a the- ory of action.Am. J. Sociol., 91:1309–1335, 1986

    1986

  39. [39]

    T . H. Costello, G. Pennycook, and D. G. Rand. Durably reducing conspiracy beliefs through dialogues with AI. Science, 385(6714):eadq1814, 2024

    2024

  40. [40]

    J. A. Danowski and P . Edison-Swift. Crisis effects on intraorganizational computer-based communica- tion.Commun. Res., 12(2):251–270, 1985

    1985

  41. [41]

    Dautenhahn, editor.Human Cognition and Social Agent Technology

    K. Dautenhahn, editor.Human Cognition and Social Agent Technology. John Benjamins, Amsterdam, 2000

    2000

  42. [42]

    De Sola Pool, R

    I. De Sola Pool, R. P . Abelson, and S. Popkin.Can- didates, Issues, and Strategies: A Computer Simulation of the 1960 And 1964 Presidential Elections. The MIT Press, Cambridge MA, 1965

    1960

  43. [43]

    de Sola Pool and A

    I. de Sola Pool and A. R. Kessler. The kaiser, the tsar, and the computer: Information processing in a crisis. Am. Behav. Sci., 8:31–38, 1965

    1965

  44. [44]

    Dell’Acqua, E

    F . Dell’Acqua, E. McFowland III, E. R. Mollick, H. Lifshitz-Assaf, K. Kellogg, S. Rajendran, L. Krayer, F . Candelon, and K. R. Lakhani. Navigating the jagged technological frontier: Field experimental evidence of the effects of ai on knowledge worker productivity and quality . Technical Report 24-013, Harvard Business School, 2024

    2024

  45. [45]

    Devlin, M.-W

    J. Devlin, M.-W . Chang, K. Lee, and K. Toutanova. BERT: Pre-training of deep bidirectional trans- formers for language understanding. Preprint arXiv:1810.04805, 2018

    Pith/arXiv arXiv 2018

  46. [46]

    P . S. Dodds, R. Muhamad, and D. J. Watts. An experi- mental study of search in global social networks.Sci- ence, 301(5634):827–829, 2003

    2003

  47. [47]

    J. Durkin. Expert systems: A view of the field.IEEE Intell. Syst., 11:56–63, 1996

    1996

  48. [48]

    J. M. Epstein and R. Axtell.Growing Artificial Societies: Social Science from the Bottom Up. Brookings Institution Press, Washington DC, 1996

    1996

  49. [49]

    J. D. Farmer. Quantitative agent-based models: a promising alternative for macroeconomics.Oxf. Rev. Econ. Policy, page graf027, 2025

    2025

  50. [50]

    J. D. Farmer and A. d’A. Belin. Artificial life: The com- ing evolution. In C. G. Langton, C. Taylor, J. D. Farmer, and S. Rasmussen, editors,Artificial Life II, pages 815–

  51. [51]

    Fehr and S

    E. Fehr and S. Gächter. Cooperation and punishment in public goods experiments.Am. Econ. Rev., 90(4):980– 994, 2000

    2000

  52. [52]

    Feigenbaum

    E. Feigenbaum. Computer simulation of human behav- ior. InMidwest Human Factors Symposium, pages 1–8. AC Spark Plug, Chicago, 1963

    1963

  53. [53]

    J. H. Feldman and D. H. Ballard. Connectionist models and their properties.Cogn. Sci., 6:205–254, 1982

    1982

  54. [54]

    Fontana, F

    N. Fontana, F . Pierri, and L. M. Aiello. Nicer than hu- mans: How do large language models behave in the prisoner’s dilemma?Proc. Int. AAAI Conf. Weblogs. Soc. Media., 19(1):522–535, Jun. 2025

    2025

  55. [55]

    J. W . Forrester. Industrial dynamics.Harv. Bus. Rev., 16:37–66, 1958

    1958

  56. [56]

    L. C. Freeman.The Development of Social Network Anal- ysis: A Study in the Sociology of Science. Empirical Press, Vancouver, 2004

    2004

  57. [57]

    K. Frenken. A complexity approach to innovation net- works. the case of the aircraft industry (1909–1997). Res. Policy, 29(2):257–272, 2000

    1909

  58. [58]

    Froese and T

    T . Froese and T . Ziemke. Enactive artificial intelli- gence: Investigating the systemic organization of life and mind.Artif. Intell., 174(3-4):466–500, 2009

    2009

  59. [59]

    C. Gao, X. Lan, N. Li, Y. Yuan, J. Ding, Z. Zhou, F . Xu, and Y. Li. Large language models empowered agent- based modeling and simulation: a survey and perspec- tives.Humanit. Soc. Sci. Commun., 11:1259, 2024

    2024

  60. [60]

    Gautam, A

    H. Gautam, A. Gaur, and D. K. Yadav. A survey on the impact of pre-trained language models in sentiment classification task.Int. J. Data Sci. Anal., 2025

    2025

  61. [61]

    D. H. Gelernter.Mirror Worlds. Oxford University Press, New York, 1991

    1991

  62. [62]

    Gershenson, V

    C. Gershenson, V . Trianni, J. Werfel, and H. Sayama. Self-organization and artificial life.Artif. Life, 26(3):391–408, 2020

    2020

  63. [63]

    J. J. Gibson.The Senses Considered as Perceptual Sys- tems. Houghton Mifflin, Boston MA, 1966

    1966

  64. [64]

    Giddens.The Constitution of Society

    A. Giddens.The Constitution of Society. Polity Press, Cambridge, 1984

    1984

  65. [65]

    J. P . Gilbert and E. A. Hammel. Computer simulation and analysis of problems in kinship and social struc- ture.Am. Anthropol., 68:71–93, 1966

    1966

  66. [66]

    Gilbert and R

    N. Gilbert and R. Conte.Artificial Societies: The Com- puter Simulation of Social Life. Routledge, London, 1995

    1995

  67. [67]

    Gilbert and K

    N. Gilbert and K. G. Troitzsch.Simulation for the social scientist. Open University Press, New York, 2005

    2005

  68. [68]

    M. S. Granovetter. The strength of weak ties.Am. J. Sociol., 78(6):1360–1380, 1973

    1973

  69. [69]

    Graves, A.-R

    A. Graves, A.-R. Mohamed, and G. Hinton. Speech recognition with deep recurrent neural networks. In IEEE International Conference on Acoustics, Speech and Signal Processing, pages 6645–6649, 2013

    2013

  70. [70]

    Grimmer, M

    J. Grimmer, M. E. Roberts, and B. M. Stewart. Ma- chine learning for social science: An agnostic approach. Annu. Rev. Polit. Sci., 24:395–419, 2021

    2021

  71. [71]

    Grossmann, M

    I. Grossmann, M. Feinberg, D. C. Parker, N. A. Chris- takis, P . E. Tetlock, and W . A. Cunningham. AI and the transformation of social science research.Science, 380:1108–1109, 2023

    2023

  72. [72]

    J. T . Gullahorn and J. E. Gullahorn. Some computer applications in social science.Am. Sociol. Rev., 30:353– 365, 1965

    1965

  73. [73]

    Hackenburg and H

    K. Hackenburg and H. Margetts. Evaluating the persuasive influence of political microtargeting with large language models.Proc. Natl. Acad. Sci. USA, 121(24):e2403116121, 2024

    2024

  74. [74]

    Hagendorff, S

    T . Hagendorff, S. Fabi, and M. Kosinski. Human-like in- tuitive behavior and reasoning biases emerged in large language models but disappeared in ChatGPT .Nat. Artificially intelligent agents in the social and behavioral sciences HOLME & TSVETKOV A | 15 Comput. Sci., 3(10):833–838, 2023

    2023

  75. [75]

    Hägerstrand.Innovationsförloppet ur Korologisk Syn- punkt

    T . Hägerstrand.Innovationsförloppet ur Korologisk Syn- punkt. Gleerup, Lund, 1953

    1953

  76. [76]

    Haigh, M

    T . Haigh, M. Priestley , and C. Rope. Los Alamos bets on ENIAC: Nuclear Monte Carlo simulations, 1947-1948. IEEE Ann. Hist. Comput., 36(3):42–63, 2014

    1947

  77. [77]

    Halevy , P

    A. Halevy , P . Norvig, and F . Pereira. The unreasonable effectiveness of data.IEEE Intell. Syst., 24(2):8–12, 2009

    2009

  78. [78]

    J. Han, B. Battu, I. Romi ´c, T . Rahwan, and P . Holme. Static network structure cannot stabilize cooperation among large language model agents.PLOS ONE, 20(5):0320094, 2025

    2025

  79. [79]

    H. H. Harman. Simulation: A review. InProc. Western Joint IRE-AIEE-ACM Computer Conference, pages 1–9, 1961

    1961

  80. [80]

    F . A. Hayek.The Sensory Order. University of Chicago Press, Chicago, 1952

    1952

Showing first 80 references.

This paper was first reviewed by deepseek-v4-flash on August 4, 2026.