Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 60 of the 60 references it cites were found in a registry or an index; 24 of the 50 sources its claims cite were checked for support.

4E cognition and the coevolution of human–AI interaction

Open access
Noller, Jörg · 2025 · Discover Artificial Intelligence

Abstract

This paper examines the interaction between humans and large language models (LLMs) through the lens of 4E cognition, which encompasses embodied, embedded, enactive, and extended cognitive processes, to provide a comprehensive understanding of their relational dynamics. It argues that LLMs should not be conceived of in terms of objects but be understood as a processual and relational phenomenon that co-constitutes human agency in a shared sociotechnical environment. As such, the paper reframes AI as an interactive medium of enactive and extended human action. By focusing on the relational entanglement of social databases, dynamic patterns, and algorithmic structures, the paper proposes a 4E-compatible connectionist account of AI—one that understands AI not only as a technological artifact but as a co-evolving component of the extended cognitive ecology of human life, shaping and shaped by enactive practices, intentions, and norms. Finally, the paper discusses the limits of AI, discussing the problem of AI hallucination and collapse from a 4E cognition perspective.

Overview

CoChat Summary

TL;DR

This conceptual paper develops a 4E (embodied, embedded, enactive, extended) account of large language models, arguing that LLMs should be understood not as simulations or duplications of human intelligence, nor as mere tools, but as co-constitutive elements of extended human cognition and agency within a shared digital lifeworld. It reframes ethical and governance questions around human–AI coevolution rather than machine autonomy.

Key findings

No empirical findings are reported; the paper advances a theoretical framework.

  • LLMs are characterized as "linguistic enactors" rather than embodied agents: their participation in sense-making is grounded in dialogical and symbolic responsiveness, not sensorimotor coupling, which distinguishes them from classical cognitive artifacts and embodied human partners.
  • The paper proposes "extension relations" as a new category beyond Ihde's embodiment, hermeneutic, alterity, and background relations, arguing AI is co-constitutive of cognition and action rather than an external influence.
  • Drawing on Pedreschi et al., human–AI coevolution is defined as "a process in which humans and AI algorithms continuously influence each other," producing emergent effects such as filter bubbles, opinion polarization, model degradation, and behavioral conformism.
  • Hallucination and model collapse are framed as structural features of LLMs as associative next-token predictors (citing Shumailov et al. on recursive training and Kalai et al. on training regimes rewarding confident answers over abstention), not incidental failures.

Methods

Conceptual/philosophical analysis; the paper explicitly states "no datasets were generated or analysed during the current study."

  • The argument unfolds in six steps across Sections 2–7: a 4E agency account of LLMs, AI-extended agency, AI-extended action, human–AI coevolution, epistemic limits (hallucination and model collapse), and ethical conclusions.
  • It synthesizes phenomenological (Husserl, Ihde), enactivist/4E (Gallagher, Varela, Clark & Chalmers), and connectionist sources, alongside recent AI-ethics literature (Floridi, Buckner, Haraldsen, Slaby, Aagaard).

Limitations

The paper contains no dedicated limitations section.

  • Stated in the text: the author acknowledges using OpenAI's ChatGPT "to assist in linguistic revision and idea development," with all content and conclusions claimed as the author's own.
  • Evident from the text: the argument is entirely theoretical with no empirical validation, case studies, or data; the paper itself notes that AI systems "trained on static datasets lack access to the sensorimotor contingencies and interactive feedback loops that characterize human learning," an epistemic asymmetry it does not resolve empirically.
  • Evident from the text: the framework's central construct ("extension relations") is proposed rather than tested, and the paper leaves open how its normative recommendations for data curation, design, governance, benchmarking, and education would be operationalized.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1007/s44163-025-00595-0

Cites

60 of 60 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

20 Valid · 4 Not valid · 16 Not checked · 80 Own results and interpretations

41 more claims · 6 with their cited sources checked

Create a free account to see all 120 claims and to verify the 1 reference still open.

See all claims — free account
  • It may state two separate claims as one. 47
  • Probably not a substantial claim. 41

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 55 references

Research works printed but not linked: 16

  1. Agre PE. Toward a critical technical practice: lessons learned in trying to reform AI. In: Geoffrey CB, Susan LS, William T, Les G, editors. Social science, technical systems, and cooperative work. New York: Psychology Press; 1997.
  2. Buckner C. From deep learning to rational machines: what the history of philosophy can teach us about the future of artificial intelligence. Oxford: Oxford University Press; 2024.
  3. Dreyfus HL. What computers still can’t do: a critique of artificial reason. Cambridge: MIT Press; 1992.
  4. Dubber MD, Pasquale F, Das S, editors. The oxford handbook of ethics of AI. Oxford: Oxford University Press; 2020.
  5. Floridi L. The 4th revolution: how the infosphere is reshaping human reality. Oxford: Oxford University Press; 2014.
  6. Gallagher S. Decentering the brain: Embodied cognition and the critique of neurocentrism and narrow-minded philosophy of mind. Constr Found. 2018;14(1):8–21.
  7. Noë A. Action in perception. Cambridge: MIT Press; 2004.
  8. Norman DA. The design of everyday things: revised and expanded edition. New York: Basic Books; 2013.
  9. O’Neil C. Weapons of math destruction: how big data increases inequality and threatens democracy. New York: Crown Books; 2016.
  10. Pariser E. The Filter bubble: what the internet is hiding from you. New York: Penguin Books; 2011.
  11. Searle J. Minds, Brains and Science. London: Harvard University Press; 1984.
  12. Selwyn N. Should robots replace teachers?: AI and the future of education. Cambridge: Polity Press; 2019.
  13. Stalder F. The digital condition (trans: Pakis VA). Cambridge: Wiley; 2018.
  14. Suchman LA. Human-machine reconfigurations: plans and situated actions. 2nd ed. Cambridge: Cambridge University Press; 2007.
  15. Voenecky S, Kellmeyer P, Mueller O, Burgard W, editors. The Cambridge handbook of responsible artificial intelligence: interdisciplinary perspectives. Cambridge: Cambridge University Press; 2022.
  16. Wellner GP. When AI is gender-biased. Humana Mente. 2020;13(37):127.

Cited by

See all 15

Metadata

Authors
Noller, J.
Affiliations
Ludwig-Maximilians-Universität München (DE)
Corresponding author
Noller, Jörg
Venue
Discover Artificial Intelligence
Published
2025-11-13
Volume
5
Issue
1
ISSN
2731-0809 (ISSN-L)
OpenAlex
W4416187033
Cited by
15
References
55
Citations by year
2026: 15
FWCI
6.18
Type
journal article
Language
en
Open access
Licence
Locations
  • Discover Artificial Intelligence (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Ludwig-Maximilians-Universität München
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Embodied and Extended Cognition
  • Social Sciences › Arts and Humanities › History and Philosophy of Science › Cybernetics and Technology in Society
  • Social Sciences › Psychology › Social Psychology › Social Robot Interaction and HRI
Keywords
4E cognition, large language models, human-AI interaction, extended cognition, enactive cognition, embodied cognition, embedded cognition, human agency, coevolution, connectionism, AI hallucination, socio-technical environments

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-01, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-01 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: 4E cognition and the coevolution of human–AI interaction. https://app.cochat.ai/papers/10.1007/s44163-025-00595-0
The paper
Noller, J. (2025). 4E cognition and the coevolution of human–AI interaction. Discover Artificial Intelligence. https://doi.org/10.1007/s44163-025-00595-0

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.