{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,14]],"date-time":"2026-02-14T07:33:35Z","timestamp":1771054415985,"version":"3.50.1"},"reference-count":42,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2021,11,18]],"date-time":"2021-11-18T00:00:00Z","timestamp":1637193600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2021,11,18]],"date-time":"2021-11-18T00:00:00Z","timestamp":1637193600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100000185","name":"Defense Advanced Research Projects Agency","doi-asserted-by":"publisher","award":["HR0011939944"],"award-info":[{"award-number":["HR0011939944"]}],"id":[{"id":"10.13039\/100000185","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000185","name":"Defense Advanced Research Projects Agency","doi-asserted-by":"publisher","award":["HR0011044266"],"award-info":[{"award-number":["HR0011044266"]}],"id":[{"id":"10.13039\/100000185","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Comput Math Organ Theory"],"published-print":{"date-parts":[[2023,3]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Ground Truth program was designed to evaluate social science modeling approaches using simulation test beds with ground truth intentionally and systematically embedded to understand and model complex Human Domain systems and their dynamics Lazer et\u00a0al. (Science 369:1060\u20131062, 2020). Our multidisciplinary team of data scientists, statisticians, experts in Artificial Intelligence (AI) and visual analytics had a unique role on the program to investigate accuracy, reproducibility, generalizability, and robustness of the state-of-the-art (SOTA) causal structure learning approaches applied to fully observed and sampled simulated data across virtual worlds. In addition, we analyzed the feasibility of using machine learning models to predict future social behavior with and without causal knowledge explicitly embedded. In this paper, we first present our causal modeling approach to discover the causal structure of four virtual worlds produced by the simulation teams\u2014Urban Life, Financial Governance, Disaster and Geopolitical Conflict. Our approach adapts the state-of-the-art causal discovery (including ensemble models), machine learning, data analytics, and visualization techniques to allow a human-machine team to reverse-engineer the true causal relations from sampled and fully observed data. We next present our reproducibility analysis of two research methods team\u2019s performance using a range of causal discovery models applied to both sampled and fully observed data, and analyze their effectiveness and limitations. We further investigate the generalizability and robustness to sampling of the SOTA causal discovery approaches on additional simulated datasets with known ground truth. Our results reveal the limitations of existing causal modeling approaches when applied to large-scale, noisy, high-dimensional data with unobserved variables and unknown relationships between them. We show that the SOTA causal models explored in our experiments are not designed to take advantage from vasts amounts of data and have difficulty recovering ground truth when latent confounders are present; they do not generalize well across simulation scenarios and are not robust to sampling; they are vulnerable to data and modeling assumptions, and therefore, the results are hard to reproduce. Finally, when we outline lessons learned and provide recommendations to improve models for causal discovery and prediction of human social behavior from observational data, we highlight the importance of learning data to knowledge representations or transformations to improve causal discovery and describe the benefit of causal feature selection for predictive and prescriptive modeling.<\/jats:p>","DOI":"10.1007\/s10588-021-09351-y","type":"journal-article","created":{"date-parts":[[2021,11,18]],"date-time":"2021-11-18T09:03:05Z","timestamp":1637226185000},"page":"220-241","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["Explaining and predicting human behavior and social dynamics in simulated virtual worlds: reproducibility, generalizability, and robustness of causal discovery methods"],"prefix":"10.1007","volume":"29","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6131-3073","authenticated-orcid":false,"given":"Svitlana","family":"Volkova","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dustin","family":"Arendt","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Emily","family":"Saldanha","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Maria","family":"Glenski","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ellyn","family":"Ayton","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Joseph","family":"Cottam","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sinan","family":"Aksoy","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Brett","family":"Jefferson","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Karthnik","family":"Shrivaram","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2021,11,18]]},"reference":[{"key":"9351_CR1","doi-asserted-by":"crossref","unstructured":"Abeliuk A, Huang Z, Ferrara E, Lerman K (2020) Predictability limit of partially observed systems. Scientific Rep 10(1):1\u201310","DOI":"10.1038\/s41598-020-77091-1"},{"issue":"1","key":"9351_CR2","first-page":"171","volume":"11","author":"CF Aliferis","year":"2010","unstructured":"Aliferis CF, Statnikov A, Tsamardinos I, Mani S, Koutsoukos XD (2010) Local causal and markov blanket induction for causal discovery and feature selection for classification part i: algorithms and empirical evaluation. J Mach Learn Res 11(1):171\u2013234","journal-title":"J Mach Learn Res"},{"key":"9351_CR3","doi-asserted-by":"crossref","unstructured":"Alipourfard N, Fennell PG, Lerman K (2018) Using Simpson\u2019s paradox to discover interesting patterns in behavioral data. Preprint at arXiv:1805.03094","DOI":"10.1609\/icwsm.v12i1.15017"},{"key":"9351_CR4","doi-asserted-by":"crossref","unstructured":"Athey S (2015) Machine learning and causal inference for policy evaluation. In: Proceedings of the 21th ACM SIGKDD international conference on knowledge discovery and data mining, pp 5\u20136","DOI":"10.1145\/2783258.2785466"},{"key":"9351_CR5","unstructured":"Bengio Y (2019) From system 1 deep learning to system 2 deep learning. http:\/\/www.iro.umontreal.ca\/bengioy\/NeurIPS-11dec2019.pdfAccessed 11 Nov 2021"},{"key":"9351_CR6","first-page":"507","volume":"3","author":"DM Chickering","year":"2002","unstructured":"Chickering DM (2002) Optimal structure identification with greedy search. J Mach Learn Res 3:507\u2013554","journal-title":"J Mach Learn Res"},{"issue":"2","key":"9351_CR7","first-page":"27","volume":"11","author":"F Chollet","year":"2020","unstructured":"Chollet F (2020) A definition of intelligence for the real world. J Artif Gen Intell 11(2):27\u201330","journal-title":"J Artif Gen Intell"},{"key":"9351_CR8","unstructured":"Cottam J, Glenski M, Shaw Y, Rabello R, Golding A, Volkova S, Arendt D (2021) Graph comparison for causal discovery. Visualization in data science"},{"key":"9351_CR9","doi-asserted-by":"crossref","unstructured":"Feder A, Keith KA, Manzoor E, Pryzant R, Sridhar D, Wood-Doughty Z, Eisenstein J, Grimmer J, Reichart R, Roberts ME, et\u00a0al (2021) Causal inference in natural language processing: estimation, prediction, interpretation and beyond. Preprint at arXiv:2109.00725","DOI":"10.1162\/tacl_a_00511"},{"key":"9351_CR10","unstructured":"Gain A, Shpitser I (2018) Structure learning under missing data. In: International conference on probabilistic graphical models, PMLR, pp 121\u2013132"},{"key":"9351_CR11","doi-asserted-by":"publisher","first-page":"524","DOI":"10.3389\/fgene.2019.00524","volume":"10","author":"C Glymour","year":"2019","unstructured":"Glymour C, Zhang K, Spirtes P (2019) Review of causal discovery methods based on graphical models. Front Genet 10:524","journal-title":"Front Genet"},{"issue":"10","key":"9351_CR12","doi-asserted-by":"publisher","first-page":"1101","DOI":"10.1007\/s10654-015-9995-7","volume":"30","author":"S Greenland","year":"2015","unstructured":"Greenland S, Mansournia MA (2015) Limitations of individual causal models, causal graphs, and ignorability assumptions, as illustrated by random confounding and design unfaithfulness. Eur J Epidemiol 30(10):1101\u20131110","journal-title":"Eur J Epidemiol"},{"issue":"4","key":"9351_CR13","first-page":"1","volume":"53","author":"R Guo","year":"2020","unstructured":"Guo R, Cheng L, Li J, Hahn PR, Liu H (2020) A survey of learning causality with data: problems and methods. ACM Comput Surv (CSUR) 53(4):1\u201337","journal-title":"ACM Comput Surv (CSUR)"},{"issue":"6495","key":"9351_CR14","doi-asserted-by":"publisher","first-page":"1063","DOI":"10.1126\/science.abb6144","volume":"368","author":"J Haushofer","year":"2020","unstructured":"Haushofer J, Metcalf CJE (2020) Which interventions work best in a pandemic? Science 368(6495):1063\u20131065","journal-title":"Science"},{"issue":"6324","key":"9351_CR15","doi-asserted-by":"publisher","first-page":"486","DOI":"10.1126\/science.aal3856","volume":"355","author":"JM Hofman","year":"2017","unstructured":"Hofman JM, Sharma A, Watts DJ (2017) Prediction and explanation in social systems. Science 355(6324):486\u2013488","journal-title":"Science"},{"key":"9351_CR16","volume-title":"Thinking, fast and slow","author":"D Kahneman","year":"2011","unstructured":"Kahneman D (2011) Thinking, fast and slow. Macmillan, London"},{"key":"9351_CR17","unstructured":"King G, Nielsen R, Coberley C, Pope JE, Wells A (2011) Comparative effectiveness of matching methods for causal inference. 15(1):41\u201367"},{"key":"9351_CR18","doi-asserted-by":"crossref","unstructured":"Kummerfeld E, Rix A (2019) Simulations evaluating resampling methods for causal discovery: ensemble performance and calibration. Preprint at arXiv:1910.02047","DOI":"10.1109\/BIBM47256.2019.8983327"},{"issue":"1","key":"9351_CR19","doi-asserted-by":"publisher","first-page":"33","DOI":"10.1007\/s13194-012-0056-8","volume":"3","author":"J Ladyman","year":"2013","unstructured":"Ladyman J, Lambert J, Wiesner K (2013) What is a complex system? Eur J Philos Sci 3(1):33\u201367","journal-title":"Eur J Philos Sci"},{"key":"9351_CR20","doi-asserted-by":"publisher","DOI":"10.1017\/S0140525X16001837","author":"BM Lake","year":"2017","unstructured":"Lake BM, Ullman TD, Tenenbaum JB, Gershman SJ (2017) Building machines that learn and think like people. Behav Brain Sci. https:\/\/doi.org\/10.1017\/S0140525X16001837","journal-title":"Behav Brain Sci"},{"issue":"6507","key":"9351_CR21","doi-asserted-by":"publisher","first-page":"1060","DOI":"10.1126\/science.aaz8170","volume":"369","author":"DM Lazer","year":"2020","unstructured":"Lazer DM, Pentland A, Watts DJ, Aral S, Athey S, Contractor N, Freelon D, Gonzalez-Bailon S, King G, Margetts H et al (2020) Computational social science: obstacles and opportunities. Science 369(6507):1060\u20131062","journal-title":"Science"},{"key":"9351_CR22","doi-asserted-by":"crossref","unstructured":"Martin T, Hofman JM, Sharma A, Anderson A, Watts DJ (2016) Exploring limits to prediction in complex social systems. In: Proceedings of the 25th international conference on world wide web, pp. 683\u2013694","DOI":"10.1145\/2872427.2883001"},{"key":"9351_CR23","first-page":"1","volume":"1","author":"M Mitchell","year":"2002","unstructured":"Mitchell M, Newman M (2002) Complex systems theory and evolution. Encycl Evol 1:1\u20135","journal-title":"Encycl Evol"},{"issue":"1","key":"9351_CR24","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41562-016-0021","volume":"1","author":"MR Munaf\u00f2","year":"2017","unstructured":"Munaf\u00f2 MR, Nosek BA, Bishop DV, Button KS, Chambers CD, Du Sert NP, Simonsohn U, Wagenmakers EJ, Ware JJ, Ioannidis JP (2017) A manifesto for reproducible science. Nat Hum Behav 1(1):1\u20139","journal-title":"Nat Hum Behav"},{"key":"9351_CR25","doi-asserted-by":"publisher","first-page":"13","DOI":"10.3389\/fdata.2019.00013","volume":"2","author":"A Olteanu","year":"2019","unstructured":"Olteanu A, Castillo C, Diaz F, Kiciman E (2019) Social data: biases, methodological pitfalls, and ethical boundaries. Front Big Data 2:13","journal-title":"Front Big Data"},{"key":"9351_CR26","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511803161","volume-title":"Causality","author":"J Pearl","year":"2009","unstructured":"Pearl J (2009) Causality. Cambridge University Press, Cambridge"},{"issue":"3","key":"9351_CR27","doi-asserted-by":"publisher","first-page":"54","DOI":"10.1145\/3241036","volume":"62","author":"J Pearl","year":"2019","unstructured":"Pearl J (2019) The seven tools of causal inference, with reflections on machine learning. Commun ACM 62(3):54\u201360","journal-title":"Commun ACM"},{"key":"9351_CR28","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-019-10105-3","author":"J Runge","year":"2019","unstructured":"Runge J, Bathiany S, Bollt E, Camps-Valls G, Coumou D, Deyle E, Glymour C, Kretschmer M, Mahecha M, Mu\u00f1oz J, Nes E, Peters J, Quax R, Reichstein M, Scheffer M, Sch\u00f6lkopf B, Spirtes P, Sugihara G, Sun J, Zscheischler J (2019) Inferring causation from time series in earth system sciences. Nat Commun. https:\/\/doi.org\/10.1038\/s41467-019-10105-3","journal-title":"Nat Commun"},{"key":"9351_CR29","unstructured":"Saldanha E, Cosbey R, Ayton E, Glenski M, Cottam J, Shivaram K, Jefferson B, Hutchinson B, Arendt D, Volkova S (2020) Evaluation of algorithm selection and ensemble methods for causal discovery"},{"issue":"15","key":"9351_CR30","doi-asserted-by":"publisher","first-page":"8398","DOI":"10.1073\/pnas.1915006117","volume":"117","author":"MJ Salganik","year":"2020","unstructured":"Salganik MJ, Lundberg I, Kindel AT, Ahearn CE, Al-Ghoneim K, Almaatouq A, Altschul DM, Brand JE, Carnegie NB, Compton RJ et al (2020) Measuring the predictability of life outcomes with a scientific mass collaboration. Proc Natl Acad Sci 117(15):8398\u20138403","journal-title":"Proc Natl Acad Sci"},{"issue":"5","key":"9351_CR31","doi-asserted-by":"publisher","first-page":"612","DOI":"10.1109\/JPROC.2021.3058954","volume":"109","author":"B Sch\u00f6lkopf","year":"2021","unstructured":"Sch\u00f6lkopf B, Locatello F, Bauer S, Ke NR, Kalchbrenner N, Goyal A, Bengio Y (2021) Toward causal representation learning. Proc IEEE 109(5):612\u2013634","journal-title":"Proc IEEE"},{"issue":"3","key":"9351_CR32","doi-asserted-by":"publisher","first-page":"289","DOI":"10.1214\/10-STS330","volume":"25","author":"G Shmueli","year":"2010","unstructured":"Shmueli G et al (2010) To explain or to predict? Stat Sci 25(3):289\u2013310","journal-title":"Stat Sci"},{"key":"9351_CR33","volume-title":"Causation, prediction, and search","author":"P Spirtes","year":"2000","unstructured":"Spirtes P, Glymour CN, Scheines R, Heckerman D (2000) Causation, prediction, and search. MIT press, Cambridge"},{"issue":"6317","key":"9351_CR34","doi-asserted-by":"publisher","first-page":"1240","DOI":"10.1126\/science.aah6168","volume":"354","author":"V Stodden","year":"2016","unstructured":"Stodden V, McNutt M, Bailey DH, Deelman E, Gil Y, Hanson B, Heroux MA, Ioannidis JP, Taufer M (2016) Enhancing reproducibility for computational methods. Science 354(6317):1240\u20131241","journal-title":"Science"},{"key":"9351_CR35","unstructured":"Strobl EV (2019) Improved causal discovery from longitudinal data using a mixture of dags. In: The 2019 ACM SIGKDD workshop on causal discovery, PMLR, pp 100\u2013133"},{"key":"9351_CR36","first-page":"376","volume":"2","author":"I Tsamardinos","year":"2003","unstructured":"Tsamardinos I, Aliferis CF, Statnikov AR, Statnikov E (2003) Algorithms for large scale Markov blanket discovery. FLAIRS conference 2:376\u2013380","journal-title":"FLAIRS conference"},{"key":"9351_CR37","unstructured":"Tu R, Zhang C, Ackermann P, Mohan K, Kjellstrm H, Zhang K (2019) Causal discovery in the presence of missing data. In: The 22nd international conference on artificial intelligence and statistics, PMLR, pp 1762\u20131770"},{"key":"9351_CR38","unstructured":"Watts, Duncan J (2011) Everything is obvious:* Once you know the answer. Currency"},{"key":"9351_CR39","doi-asserted-by":"crossref","unstructured":"Watts DJ, Beck ED, Bienenstock EJ, Bowers J, Frank A, Grubesic A, Hofman J, Rohrer JM, Salganik M (2018) Explanation, prediction, and causality: three sides of the same coin?","DOI":"10.31219\/osf.io\/u6vz5"},{"key":"9351_CR40","doi-asserted-by":"publisher","DOI":"10.1017\/S0140525X20001685","author":"T Yarkoni","year":"2019","unstructured":"Yarkoni T (2019) The generalizability crisis. Behav Brain Sci. https:\/\/doi.org\/10.1017\/S0140525X20001685","journal-title":"Behav Brain Sci"},{"key":"9351_CR41","unstructured":"Yu K, Li J, Liu L (2016) A review on algorithms for constraint-based causal discovery. Preprint at arXiv:1611.03977"},{"issue":"6","key":"9351_CR42","doi-asserted-by":"publisher","first-page":"1023","DOI":"10.1007\/s10458-016-9326-8","volume":"30","author":"H Zhang","year":"2016","unstructured":"Zhang H, Vorobeychik Y, Letchford J, Lakkaraju K (2016) Data-driven agent-based modeling, with application to rooftop solar adoption. Auton Agents Multi-Agent Syst 30(6):1023\u20131049","journal-title":"Auton Agents Multi-Agent Syst"}],"container-title":["Computational and Mathematical Organization Theory"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10588-021-09351-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10588-021-09351-y\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10588-021-09351-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,3,24]],"date-time":"2023-03-24T21:17:35Z","timestamp":1679692655000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s10588-021-09351-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,11,18]]},"references-count":42,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2023,3]]}},"alternative-id":["9351"],"URL":"https:\/\/doi.org\/10.1007\/s10588-021-09351-y","relation":{},"ISSN":["1381-298X","1572-9346"],"issn-type":[{"value":"1381-298X","type":"print"},{"value":"1572-9346","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,11,18]]},"assertion":[{"value":"23 October 2021","order":1,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"18 November 2021","order":2,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}}]}}