Evolution of a theory of mind

Tom Lenaerts; Marco Saponara; Jorge M. Pacheco; Francisco C. Santos

doi:10.1016/j.isci.2024.108862

Evolution of a theory of mind

Tom Lenaerts, Marco Saponara, Jorge M. Pacheco, Francisco C. Santos

Research output: Contribution to journal › Article › peer-review

5 Downloads (Pure)

Abstract

Even though the Theory of Mind in upper primates has been under investigation for decades, how it may evolve remains an open problem. We propose here an evolutionary game theoretical model where a finite population of individuals may use reasoning strategies to infer a response to the anticipated behavior of others within the context of a sequential dilemma, i.e., the Centipede Game. We show that strategies with bounded reasoning evolve and flourish under natural selection, provided they are allowed to make reasoning mistakes and a temptation for higher future gains is in place. We further show that non-deterministic reasoning co-evolves with an optimism bias that may lead to the selection of new equilibria, closely associated with average behavior observed in experimental data. This work reveals both a novel perspective on the evolution of bounded rationality and a co-evolutionary link between the evolution of Theory of Mind and the emergence of misbeliefs.

Original language	English
Article number	108862
Number of pages	13
Journal	iScience
Volume	27
Issue number	2
DOIs	https://doi.org/10.1016/j.isci.2024.108862
Publication status	Published - 16 Feb 2024

Bibliographical note

Funding Information:
T.L. thanks the Université Libre de Bruxelles and the F.R.S-FNRS for awarding a sabbatical leave between October 2021 and September 2022, providing the time to think about Theory of Mind and related topics, and to create the results reported in this article. T.L. is grateful to the Center for Human Compatible AI (UC Berkeley), the Group on AI for People and Society (Técnico Lisboa) and the Responsible and Social Aware AI groups (Umeå university) for hosting him in that period. The resources and services used in this work were provided by the VSC (Flemish Supercomputer Center), funded by the Research Foundation Flanders ( FWO ) and the Flemish Government .

Funding Information:
T.L. and M.S. gratefully acknowledges the research support of the F.R.S. - FNRS (grants 31257234 and 40007793 ). T.L. further acknowledges the support of the FWO (grant G.0391.13N ), the Service Public de Wallonie Recherche (grant 2010235–ARIAC ) by DigitalWallonia4.ai and the Flemish Government through the AI Research Program . T.L. and F.C.S. also acknowledges the support of TAILOR , a project funded by the EU Horizon 2020 research and innovation program (grant 952215 ). J.M.P and F.C.S gratefully acknowledge funding from Fundação para a Ciência e a Tecnologia Portugal through grants PTDC/MAT-APL/6804/2020 , PTDC/CCI-INF/7366/2020 , and UIDB/50021/2020 .

Funding Information:
T.L. thanks the Université Libre de Bruxelles and the F.R.S-FNRS for awarding a sabbatical leave between October 2021 and September 2022, providing the time to think about Theory of Mind and related topics, and to create the results reported in this article. T.L. is grateful to the Center for Human Compatible AI (UC Berkeley), the Group on AI for People and Society (Técnico Lisboa) and the Responsible and Social Aware AI groups (Umeå university) for hosting him in that period. The resources and services used in this work were provided by the VSC (Flemish Supercomputer Center), funded by the Research Foundation Flanders (FWO) and the Flemish Government. T.L. and M.S. gratefully acknowledges the research support of the F.R.S. - FNRS (grants 31257234 and 40007793). T.L. further acknowledges the support of the FWO (grant G.0391.13N), the Service Public de Wallonie Recherche (grant 2010235–ARIAC) by DigitalWallonia4.ai and the Flemish Government through the AI Research Program. T.L. and F.C.S. also acknowledges the support of TAILOR, a project funded by the EU Horizon 2020 research and innovation program (grant 952215). J.M.P and F.C.S gratefully acknowledge funding from Fundação para a Ciência e a Tecnologia Portugal through grants PTDC/MAT-APL/6804/2020 and PTDC/CCI-COM/366/2020. Conceptualization, T.L. M.S. J.M.P, and F.C.S.; methodology, T.L. M.S. J.M.P, and F.C.S.; software T.L. and M.S. ; formal analysis, T.L. M.S. J.M.P, and F.C.S. ; writing – original draft, T.L. M.S. J.M.P, and F.C.S.; writing – review and editing, T.L. M.S. J.M.P, and F.C.S.; visualization, T.L. M.S. J.M.P, and F.C.S; funding acquisition, T.L. J.M.P, and F.C.S.; project administration, T.L. J.M.P, and F.C.S. The authors declare no competing interests.

Publisher Copyright:
© 2024 The Author(s)

Keywords

Behavioral neuroscience
cognitive neuroscience
game playing

Access to Document

10.1016/j.isci.2024.108862Licence: CC BY

108142863Final published version, 3.8 MBLicence: CC BY

Cite this

@article{a824093485d5405fb29e2bc50192322d,

title = "Evolution of a theory of mind",

abstract = "Even though the Theory of Mind in upper primates has been under investigation for decades, how it may evolve remains an open problem. We propose here an evolutionary game theoretical model where a finite population of individuals may use reasoning strategies to infer a response to the anticipated behavior of others within the context of a sequential dilemma, i.e., the Centipede Game. We show that strategies with bounded reasoning evolve and flourish under natural selection, provided they are allowed to make reasoning mistakes and a temptation for higher future gains is in place. We further show that non-deterministic reasoning co-evolves with an optimism bias that may lead to the selection of new equilibria, closely associated with average behavior observed in experimental data. This work reveals both a novel perspective on the evolution of bounded rationality and a co-evolutionary link between the evolution of Theory of Mind and the emergence of misbeliefs.",

keywords = "Behavioral neuroscience, cognitive neuroscience, game playing",

author = "Tom Lenaerts and Marco Saponara and Pacheco, {Jorge M.} and Santos, {Francisco C.}",

note = "Funding Information: T.L. thanks the Universit{\'e} Libre de Bruxelles and the F.R.S-FNRS for awarding a sabbatical leave between October 2021 and September 2022, providing the time to think about Theory of Mind and related topics, and to create the results reported in this article. T.L. is grateful to the Center for Human Compatible AI (UC Berkeley), the Group on AI for People and Society (T{\'e}cnico Lisboa) and the Responsible and Social Aware AI groups (Ume{\aa} university) for hosting him in that period. The resources and services used in this work were provided by the VSC (Flemish Supercomputer Center), funded by the Research Foundation Flanders ( FWO ) and the Flemish Government . Funding Information: T.L. and M.S. gratefully acknowledges the research support of the F.R.S. - FNRS (grants 31257234 and 40007793 ). T.L. further acknowledges the support of the FWO (grant G.0391.13N ), the Service Public de Wallonie Recherche (grant 2010235–ARIAC ) by DigitalWallonia4.ai and the Flemish Government through the AI Research Program . T.L. and F.C.S. also acknowledges the support of TAILOR , a project funded by the EU Horizon 2020 research and innovation program (grant 952215 ). J.M.P and F.C.S gratefully acknowledge funding from Funda{\c c}{\~a}o para a Ci{\^e}ncia e a Tecnologia Portugal through grants PTDC/MAT-APL/6804/2020 , PTDC/CCI-INF/7366/2020 , and UIDB/50021/2020 . Funding Information: T.L. thanks the Universit{\'e} Libre de Bruxelles and the F.R.S-FNRS for awarding a sabbatical leave between October 2021 and September 2022, providing the time to think about Theory of Mind and related topics, and to create the results reported in this article. T.L. is grateful to the Center for Human Compatible AI (UC Berkeley), the Group on AI for People and Society (T{\'e}cnico Lisboa) and the Responsible and Social Aware AI groups (Ume{\aa} university) for hosting him in that period. The resources and services used in this work were provided by the VSC (Flemish Supercomputer Center), funded by the Research Foundation Flanders (FWO) and the Flemish Government. T.L. and M.S. gratefully acknowledges the research support of the F.R.S. - FNRS (grants 31257234 and 40007793). T.L. further acknowledges the support of the FWO (grant G.0391.13N), the Service Public de Wallonie Recherche (grant 2010235–ARIAC) by DigitalWallonia4.ai and the Flemish Government through the AI Research Program. T.L. and F.C.S. also acknowledges the support of TAILOR, a project funded by the EU Horizon 2020 research and innovation program (grant 952215). J.M.P and F.C.S gratefully acknowledge funding from Funda{\c c}{\~a}o para a Ci{\^e}ncia e a Tecnologia Portugal through grants PTDC/MAT-APL/6804/2020 and PTDC/CCI-COM/366/2020. Conceptualization, T.L. M.S. J.M.P, and F.C.S.; methodology, T.L. M.S. J.M.P, and F.C.S.; software T.L. and M.S. ; formal analysis, T.L. M.S. J.M.P, and F.C.S. ; writing – original draft, T.L. M.S. J.M.P, and F.C.S.; writing – review and editing, T.L. M.S. J.M.P, and F.C.S.; visualization, T.L. M.S. J.M.P, and F.C.S; funding acquisition, T.L. J.M.P, and F.C.S.; project administration, T.L. J.M.P, and F.C.S. The authors declare no competing interests. Publisher Copyright: {\textcopyright} 2024 The Author(s)",

year = "2024",

month = feb,

day = "16",

doi = "10.1016/j.isci.2024.108862",

language = "English",

volume = "27",

journal = "iScience",

issn = "2589-0042",

publisher = "Cell Press",

number = "2",

}

TY - JOUR

T1 - Evolution of a theory of mind

AU - Lenaerts, Tom

AU - Saponara, Marco

AU - Pacheco, Jorge M.

AU - Santos, Francisco C.

N1 - Funding Information: T.L. thanks the Université Libre de Bruxelles and the F.R.S-FNRS for awarding a sabbatical leave between October 2021 and September 2022, providing the time to think about Theory of Mind and related topics, and to create the results reported in this article. T.L. is grateful to the Center for Human Compatible AI (UC Berkeley), the Group on AI for People and Society (Técnico Lisboa) and the Responsible and Social Aware AI groups (Umeå university) for hosting him in that period. The resources and services used in this work were provided by the VSC (Flemish Supercomputer Center), funded by the Research Foundation Flanders ( FWO ) and the Flemish Government . Funding Information: T.L. and M.S. gratefully acknowledges the research support of the F.R.S. - FNRS (grants 31257234 and 40007793 ). T.L. further acknowledges the support of the FWO (grant G.0391.13N ), the Service Public de Wallonie Recherche (grant 2010235–ARIAC ) by DigitalWallonia4.ai and the Flemish Government through the AI Research Program . T.L. and F.C.S. also acknowledges the support of TAILOR , a project funded by the EU Horizon 2020 research and innovation program (grant 952215 ). J.M.P and F.C.S gratefully acknowledge funding from Fundação para a Ciência e a Tecnologia Portugal through grants PTDC/MAT-APL/6804/2020 , PTDC/CCI-INF/7366/2020 , and UIDB/50021/2020 . Funding Information: T.L. thanks the Université Libre de Bruxelles and the F.R.S-FNRS for awarding a sabbatical leave between October 2021 and September 2022, providing the time to think about Theory of Mind and related topics, and to create the results reported in this article. T.L. is grateful to the Center for Human Compatible AI (UC Berkeley), the Group on AI for People and Society (Técnico Lisboa) and the Responsible and Social Aware AI groups (Umeå university) for hosting him in that period. The resources and services used in this work were provided by the VSC (Flemish Supercomputer Center), funded by the Research Foundation Flanders (FWO) and the Flemish Government. T.L. and M.S. gratefully acknowledges the research support of the F.R.S. - FNRS (grants 31257234 and 40007793). T.L. further acknowledges the support of the FWO (grant G.0391.13N), the Service Public de Wallonie Recherche (grant 2010235–ARIAC) by DigitalWallonia4.ai and the Flemish Government through the AI Research Program. T.L. and F.C.S. also acknowledges the support of TAILOR, a project funded by the EU Horizon 2020 research and innovation program (grant 952215). J.M.P and F.C.S gratefully acknowledge funding from Fundação para a Ciência e a Tecnologia Portugal through grants PTDC/MAT-APL/6804/2020 and PTDC/CCI-COM/366/2020. Conceptualization, T.L. M.S. J.M.P, and F.C.S.; methodology, T.L. M.S. J.M.P, and F.C.S.; software T.L. and M.S. ; formal analysis, T.L. M.S. J.M.P, and F.C.S. ; writing – original draft, T.L. M.S. J.M.P, and F.C.S.; writing – review and editing, T.L. M.S. J.M.P, and F.C.S.; visualization, T.L. M.S. J.M.P, and F.C.S; funding acquisition, T.L. J.M.P, and F.C.S.; project administration, T.L. J.M.P, and F.C.S. The authors declare no competing interests. Publisher Copyright: © 2024 The Author(s)

PY - 2024/2/16

Y1 - 2024/2/16

N2 - Even though the Theory of Mind in upper primates has been under investigation for decades, how it may evolve remains an open problem. We propose here an evolutionary game theoretical model where a finite population of individuals may use reasoning strategies to infer a response to the anticipated behavior of others within the context of a sequential dilemma, i.e., the Centipede Game. We show that strategies with bounded reasoning evolve and flourish under natural selection, provided they are allowed to make reasoning mistakes and a temptation for higher future gains is in place. We further show that non-deterministic reasoning co-evolves with an optimism bias that may lead to the selection of new equilibria, closely associated with average behavior observed in experimental data. This work reveals both a novel perspective on the evolution of bounded rationality and a co-evolutionary link between the evolution of Theory of Mind and the emergence of misbeliefs.

AB - Even though the Theory of Mind in upper primates has been under investigation for decades, how it may evolve remains an open problem. We propose here an evolutionary game theoretical model where a finite population of individuals may use reasoning strategies to infer a response to the anticipated behavior of others within the context of a sequential dilemma, i.e., the Centipede Game. We show that strategies with bounded reasoning evolve and flourish under natural selection, provided they are allowed to make reasoning mistakes and a temptation for higher future gains is in place. We further show that non-deterministic reasoning co-evolves with an optimism bias that may lead to the selection of new equilibria, closely associated with average behavior observed in experimental data. This work reveals both a novel perspective on the evolution of bounded rationality and a co-evolutionary link between the evolution of Theory of Mind and the emergence of misbeliefs.

KW - Behavioral neuroscience

KW - cognitive neuroscience

KW - game playing

UR - http://www.scopus.com/inward/record.url?scp=85183037798&partnerID=8YFLogxK

U2 - 10.1016/j.isci.2024.108862

DO - 10.1016/j.isci.2024.108862

M3 - Article

AN - SCOPUS:85183037798

VL - 27

JO - iScience

JF - iScience

SN - 2589-0042

IS - 2

M1 - 108862

ER -

Evolution of a theory of mind

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Stochastic Evolutionary Dynamic Code for Theory of Mind in the Constant-Size Centipede Game

Stochastic Evolutionary Dynamic Code for Theory of Mind in the Incremental Centipede Game

Moran simulation code for Theory of Mind in the Incremental Centipede Game

Cite this

Evolution of a theory of mind

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Datasets

Stochastic Evolutionary Dynamic Code for Theory of Mind in the Constant-Size Centipede Game

Stochastic Evolutionary Dynamic Code for Theory of Mind in the Incremental Centipede Game

Moran simulation code for Theory of Mind in the Incremental Centipede Game

Cite this