Conceptual model of a multi-agent innovatNeuropsychological architecture of a general-purpose artificial intelligence agent

Z.V. Nagoev

---

Abstract: The object of the study is the neuropsychological architecture of the brain.
Aim. The study is to develop theoretical foundations for the creation of a universal artificial intelligence agent.
Research methods. An approach to the ontogenetic formation of universal control systems is used. The essence of these systems lies in the simulation modeling of the growth and development of natural autonomous intelligent agents, equipped with a basic cognitive architecture possessing structural and functional similarities to the cognitive architecture of the brain, in a real social environment. Specifically, a key hypothesis is advanced that such similarities are possessed by so-called neurocognitive architectures, which represent a metaphor for designing an intelligent system for controlling the behavior of an autonomous agent immersed in a real environment using sensors and effectors. This architecture is based on concepts regarding the composition and interaction modes of functional units–the so-called invariants of the organization of the cognitive architecture of decision-making.
Results. A structural and functional diagram of neuropsychological architecture has been developed, motivated by its systemic purpose, enabling neurocognitive transformations of the problem-defining space through the coordinated work of neurocognitive components implementing cascades of n-functions for dynamic transitions between the behavioral space, local problem-defining spaces, and the mental space.
Conclusion. A concept for the neuropsychological architecture of an autonomous behavioral control system for a universally intelligent agent has been developed. It is shown that the universal nature of the range of problems that such agents are capable of handling is ensured by a methodological and algorithmic framework for ontologization, identification, and problem solving based on the synthesis of the agent’s behavior within the agent-environment system by the neuropsychological architecture.

Keywords: general artificial intelligence, neuropsychological architecture, neurocognitive architecture, neurocognitive functions and mappings, multi-agent systems and functions

For citation. Nagoev Z.V. Neuropsychological architecture of a general-purpose artificial intelligence agent. News of the Kabardino-Balkarian Scientific Center of RAS. 2025. Vol. 27. No. 6. Pp. 186–208. DOI: 10.35330/1991-6639-2025-27-6-186-208

References

1. http://www.robotspodcast.com/podcast/2012/10/robots-rethink-robotics/#transcript
2. Russell S., Norvig P. Artificial Intelligence: Modern approach (AIMA). 2nd ed. Moscow: Williams, 2007. 1424 p.
3. Nagoev Z.V., Nagoeva O.V. Obosnovaniye simvolov i mul’tiagentnyye neyrokognitivnyye modeli semantiki yestestvennogo yazyka [Justification of symbols and multi-agent neurocognitive models of natural language semantics]. Nalchik: Izdatel’stvo KBNTS RAN, 2022. 150 p. (In Russian)
4. Nagoev Z.V., Anchekov M.I., Bzhikhatlov K.Ch. et al. Ontoepisociophylogenetic development of general artificial intelligence systems based on multi-agent neurocognitive
   architectures. News of the Kabardino-Balkarian Scientific Center of RAS. 2022. No. 6(110). Pp. 61–75. DOI: 10.35330/1991-6639-2022-6-110-61-75. (In Russian)
5. Nagoev Z.V., Nagoeva O., Gurtueva I. Multi-Agent neurocognitive models of semantics of spatial localization of events. Cognitive Systems Research. 2020. 59. Pp. 91–102. DOI:
   10.1016/j.cogsys.2019.09.015.
6. Nagoev Z.V. Multiagent recursive cognitive architecture. Biologically inspired cognitive architectures 2012. Proceedings of the Third Annual Meeting of the BICA Society, in Advances in Intelligent Systems and Computing Series, Springer. 2012. Pp. 247–248.
7. Nagoev Z.V. Intellektika, ili Myshleniye v zhivykh i iskusstvennykh sistemakh [Intellectics, or Thinking in Living and Artificial Systems]. Nalchik: Izdatel’stvo KBNTS RAN, 2013. 232 p. (In Russian)
8. Nagoev Z.V. Basic principles of neurocognitive modeling of the consciousness of a universal artificial intelligence agent. News of the Kabardino-Balkarian Scientific Center of RAS. 2025. Vol. 27. No. 1. Pp. 152–170. DOI: 10.35330/1991-6639-2025-27-1-152-170. (In Russian)
9. Nagoev Z.V., Pshenokova I.A., Nagoeva O.V. et al. Simulation model of a neurocognitive control system for an autonomous software agent performing cooperative behavior for the purpose of automatic ontology replenishment. News of the Kabardino-Balkarian Scientific Center of RAS. 2023. No. 6 (116). Pp. 226–234. DOI: 10.35330/1991-6639-2023-6-116-226-234. (In Russian)
10. Nagoev Z.V. Multi-agent existential mappings and functions. News of the KabardinoBalkarian Scientific Center of RAS. 2013. No. 4(54). Pp. 63–71. EDN: QZTFLX. (In Russian)
11. Nagoev Z., Nagoeva O., Anchokov M. et al. The symbol grounding problem in the system of general artificial intelligence based on multi-agent neurocognitive architecture. Cognitive Systems Research. 2023. Vol. 79. Pp. 71–84. DOI: 10.1016/j.cogsys.2023.01.002
12. Nagoev Z.V. Multi-agent existential mappings and functions. News of the KabardinoBalkarian Scientific Center of RAS. 2013. No. 4(54). Pp. 63–71. EDN: QZTFLX. (In Russian)
13. Chalmers D.J. The conscious mind: In search of a fundamental theory. 1 edition. Oxford: Oxford University Press, 1996. 432 p. ISBN 978-0195117899.
14. John R. Searle, “Minds, brains, and programs”. Behavioral and Brain Sciences 3 (3):417-457.
15. Nagoev Z.V. Ontoneuromorphogenetic modeling. News of the Kabardino-Balkarian Scientific Center of RAS. 2013. No. 4(54). Pp. 56–63. EDN: QZTFLN. (In Russian)
16. Apshev A.Z., Atalikov B.A., Kankulov S.A. et al. Ontophylogenetic algorithms for synthesizing phenotypes of intelligent software agents for use in multi-generation optimization
    problems of control neurocognitive architectures. News of the Kabardino-Balkarian Scientific Center of RAS. 2022. No. 6(110). Pp. 76–91. DOI: 10.35330/1991-6639-2022-6-110-76-91. (In Russian)

Information about the authors

Zalimkhan V. Nagoev, Candidate of Technical Sciences, General Director of the Kabardino-Balkarian Scientific Center of the Russian Academy of Sciences;
2, Balkarov street, Nalchik, 360010, Russia;
Leading Researcher, Department “Multi-Agent Systems”, Institute of Computer Science and Problems of Regional Management – branch of the Kabardino-Balkarian Scientific Center of the Russian Academy of Sciences;
37-a, I. Armand street, Nalchik, 360000, Russia;
zaliman@mail.ru, ORCID: https://orcid.org/0000-0001-9549-1823, SPIN-code: 6279-5857