Pith. sign in

REVIEW 3 major objections 4 minor 25 references

Mapping Neural Theories of Consciousness onto the Common Model of Cognition

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Four neural theories of consciousness converge on two shared architectural ingredients.

desk verdict A clear but preliminary mapping of four consciousness theories onto the CMC; the IIT leg is weak enough that the 'four jointly' claim is not yet supported. read the letter →

arxiv 2506.12224 v1 pith:2MRY4LVC submitted 2025-06-13 q-bio.NC cs.AI

Paul S. Rosenbloom , John E. Laird , Christian Lebiere , Andrea Stocco This is my paper
classification q-bio.NCcs.AI
keywords consciousnessCommonModelofCognitioncognitivearchitectureglobalworkspacetheoryintegratedinformationrecurrentprocessingpredictiveworkingmemory
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper attempts an early mapping of four neural theories of consciousness—Global Neuronal Workspace Theory, Integrated Information Theory, Recurrent Processing Theory, and Predictive Processing/Neurorepresentationalism—onto the Common Model of Cognition (CMC), a consensus abstraction of cognitive architectures. The authors argue that, when translated into CMC terms, the four theories converge on two structural ingredients: recurrent local processing inside modules, and a global working memory plus cognitive cycle whose states are complex, possibly multimodal, interacting representations. They further argue that this convergence shows where access consciousness (reportability) and phenomenal consciousness (qualia) each live in the architecture: access consciousness arises through the global working memory and cycle, while phenomenal consciousness arises through recurrent processing or complex representations. The mapping is deliberately a beginning, not a completed translation; the authors flag at least one unresolved point in the IIT mapping. If the mapping holds, it offers a single architectural vocabulary for relating neural theories of consciousness and for extending them to non-human and artificial agents.

What carries the argument

The central machinery is the Common Model of Cognition (CMC), a consensus abstraction of a cognitive architecture: a working memory hub for inter-module communication, procedural and declarative long-term memories with buffers, perception and motor modules, learning mechanisms, and a roughly 50 ms cognitive cycle that drives sequential decisions. The mapping proceeds by identifying each theory's central construct with a CMC component: GNWT's global workspace with working memory, IIT's physical system with a problem space of states and operators, RPT's local feedback with within-module recurrence and its global feedback with working-memory-based recurrence, and PP/NREP's predictive hierarchy with multimodal working-memory states plus implicit prediction in procedural and declarative metadata. Problem spaces—a construct from two CMC instantiations—do the load-bearing work for IIT, since they provide both a distinct current state and a space of possible states with operators that change states.

What would settle it

A decisive test would be to implement each of the four theories in a full CMC instantiation and check whether every implementation requires both recurrent local module processing and a global stateful working-memory cycle; showing that IIT can support consciousness without the cycle, or RPT without local recurrence, would refute the paper's convergence claim.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that four prominent neural theories of consciousness, despite their different commitments, jointly require the same two capabilities from an abstract cognitive architecture: (1) local processing within modules that may be recurrent, and (2) a global working memory and a cognitive cycle that operate on states with complex, possibly multimodal, interacting representations. GNWT maps its global workspace onto the working memory and its spotlight onto the cognitive cycle; IIT maps a physical system's self-cause-effect power onto a problem space whose operators drive state changes through the cycle; RPT maps local feedback to within-module recurrence and global feedback to working memory; and PP/NREP maps its predictive hierarchy onto the same modules, working memory, and cycle. Since these capabilities are already present in the CMC, the authors conclude that the CMC provides a useful level of abstraction for relating neural theories of consciousness and for applying them beyond the brain.

Load-bearing premise

The load-bearing premise is that each neural theory's core constructs can be translated into CMC components without distortion; if, for example, IIT's cause-effect structures cannot be represented as problem-space symbol structures, or if RPT's local recurrence cannot be captured in the CMC's currently unspecified perception and motor modules, the claimed convergence would be an artifact of the mapping.

Editorial extensions

If this is right

  • Access consciousness in a CMC-based system should depend on the global working memory and cognitive cycle, because GNWT, RPT, and PP/NREP all place reportability at the global level.
  • Phenomenal consciousness should be expected to arise from local recurrent processing in modules or from the complexity of working-memory representations, not from a single dedicated consciousness module.
  • IIT-style consciousness, if the translation is faithful, would be tied to problem-space structure: a distinct current state plus operators that give the system cause-effect power over itself.
  • The working memory module is the point of convergence: it simultaneously plays the role of the global workspace, the global recurrent signal, the highest-level multimodal representation, and the integrated state of IIT.
  • The paper's mapping implies that any CMC-based artificial agent could in principle host these neural theories of consciousness, provided its modules support local recurrence and its working memory supports complex states.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A stronger test of the convergence claim would be to implement each theory inside a full CMC instantiation and see whether their predicted behavioral markers of access and phenomenal consciousness actually diverge or coincide.
  • The unresolved IIT question—whether cause-effect structures are relations within states or across states—could be settled by defining operators as explicit cause-effect relations between problem-space states, making the structure checkable in a CMC agent.
  • If the convergence is real, disagreements among the four theories may be less about the underlying architecture and more about where each theory places the conscious locus within that architecture.
  • The mapping suggests a testable design rule: a CMC-based agent aiming at consciousness would need both recurrent local circuits and a multimodal stateful hub, and removing either should remove the corresponding conscious capacity.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper proposes an initial mapping of four neural theories of consciousness—Global Neuronal Workspace Theory, Integrated Information Theory, Recurrent Processing Theory, and Predictive Processing/Neurorepresentationalism—onto the Common Model of Cognition (CMC). It argues that these theories jointly depend on two CMC features: (1) local processing within modules that may be recurrent, and (2) a global working memory and cognitive cycle operating on complex, possibly multimodal, interacting representations. The paper also maps the access/phenomenal consciousness distinction onto these components and discusses how the CMC can accommodate the spotlight mechanism of GNWT and direct intermodal communication in PP/NREP. The authors are explicit that the mapping is incomplete and that Dendritic Integration Theory is excluded.

Significance. If the mapping succeeds, it would provide a rare bridge between neural theories of consciousness and computational cognitive architecture, with potential consequences for understanding artificial consciousness and for testing consciousness theories in implemented systems. The paper is clearly written, engages recent literature, and, importantly, is candid about unresolved aspects (notably the IIT mapping). However, the central result is a qualitative interpretive hypothesis rather than a demonstrated theorem or empirical finding. The main value lies in organizing existing ideas under one framework, but the conclusion that all four theories jointly depend on the CMC's global working memory and cognitive cycle is only as strong as the faithfulness of each individual mapping, and at least one mapping (IIT) faces a serious unresolved difficulty.

major comments (3)
  1. [Section 4, footnote 2]
  2. [Section 5]
  3. [Section 7]
minor comments (4)
  1. [Section 2, Figure 1]
  2. [Section 4, equation (1)]
  3. [Section 5]
  4. [Section 1 / References]

Circularity Check

2 steps flagged · score 6.0 of 10

Section 7's 'joint dependence' on WM and the cognitive cycle is a restatement of the mapping choices made in Sections 3–6, not an independent result.

  1. self definitional [Section 4 (IIT mapping); echoed in Section 7 conclusion]
    "The essence of a physical system, as found in axiom/postulate pair (1) above, maps straightforwardly onto a problem space, with both a distinct current state (3) – part of WM – and a space of all possible states, with operators that move among the states. The cause-effect power on itself arises via cognitive cycles during which operators, implemented by PM, drive state changes."

    IIT's central axiom is that a physical system's consciousness is its intrinsic cause-effect power on itself, computed from the system's own transition probabilities. The paper converts that power into 'cognitive cycles during which operators, implemented by PM, drive state changes,' with the current state 'part of WM.' The Section 7 claim that IIT depends on a global WM and cognitive cycle is therefore not derived from IIT; it is the mapping itself. The attached footnote concedes the authors 'have not been able to figure out' how cause-effect structures relate across states, so the mapping is admitted to be incomplete and cannot independently support the shared-dependence conclusion.

  2. renaming known result [Section 7 (Conclusion)]
    "To conclude, it is worth noting how so many parts of the neural theories of consciousness discussed here are reflected in aspects of the CMC. For example, the omni-directional and multimodal nature of the WM module is compatible with the global workspace of the GNWT, the global feedback signal of the RPT, the highest-level representations of the PP/NREP, and – in as much as it supports problem space representations – the most integrated representations in the IIT."

    Each compatibility listed here was established earlier as a mapping stipulation: GNWT's GW is mapped to WM, RPT's global feedback is assigned to WM, PP/NREP's high-level representations are 'generated in WM,' and IIT's current state is 'part of WM.' Once the defining construct of every theory is identified with WM, the observation that all four are compatible with WM is not an emergent result but a definitional consequence. The Section 7 'joint dependence' claim is the same identification restated as a finding.

full rationale

The paper is candid that it is 'beginning to explore what can be learned by mapping' the four theories onto the CMC, and it does not claim to have run experiments or made quantitative predictions. The central deficiency is structural: the Section 7 conclusion that the theories 'depend on the CMC supporting: (1) local processing within modules that may be recurrent; and (2) a global WM and cognitive cycle' restates the mapping choices made in Sections 3–6. GNWT's workspace is 'naturally mapped' to WM, IIT's current state is declared 'part of WM' and its cause-effect power is implemented as PM-driven cognitive cycles, RPT's global feedback is assigned to WM, and PP/NREP's high-level multimodal representations are 'generated in WM.' Each mapping installs the conclusion it later reports. The IIT section is the clearest instance: the mapping replaces IIT's intrinsic, transition-probability-based cause-effect power with an inter-module cognitive-cycle mechanism, and a footnote admits the authors 'have not been able to figure out' how IIT's cause-effect structures relate across states. The paper's compatibility observations are therefore true by construction, not discoveries about the neural theories. Self-citations such as [11] and [12] are part of the framework, but the circularity would remain even if CMC were externally authored; the problem is that the 'joint dependence' is built into the mapping. Still, because the paper is an explicit interpretive mapping rather than a claimed empirical prediction, the circularity is partial, not total—score 6.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The paper introduces no free parameters and no invented entities. Its central claim rests on the assumption that CMC is a valid consensus model, that the four theories are faithfully represented, and that neural concepts can be translated into CMC components (especially problem spaces and module recurrence). These are domain assumptions and ad hoc mapping assumptions, not derived results.

assumptions (4)
  • domain assumption The Common Model of Cognition is a valid consensus abstraction of human cognition.
    The paper adopts CMC as the target model based on the authors' prior work (refs 11,12) without independent validation.
  • domain assumption The four neural theories are accurately represented by their cited formulations and by the integrative characterization in Storm et al. (ref 1).
    The mapping relies on the authors' interpretation of each theory; no competing interpretations are considered.
  • ad hoc to paper Neural concepts such as global workspace, integrated cause-effect structure, and local recurrence can be meaningfully mapped to CMC components (WM, problem spaces, module-internal loops).
    This is the central assumption of the paper; without it, the mapping does not hold. It is asserted in Sections 3-6 rather than derived.
  • ad hoc to paper CMC's perception and motor modules may have recurrent processing, despite the model being silent on their internals.
    Invoked in Section 5 to make RPT compatible; it is a compatibility assumption, not a known property of CMC.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Mapping Neural Theories of Consciousness onto the Common Model of Cognition." pith.science (2026). https://pith.science/paper/2MRY4LVC

@misc{pith2026250612224,
  author       = {Pith},
  title        = {Pith review of: Mapping Neural Theories of Consciousness onto the Common Model of Cognition},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2MRY4LVC}},
  note         = {Machine review of arXiv:2506.12224}
}
read the original abstract

A beginning is made at mapping four neural theories of consciousness onto the Common Model of Cognition. This highlights how the four jointly depend on recurrent local modules plus a cognitive cycle operating on a global working memory with complex states, and reveals how an existing integrative view of consciousness from a neural perspective aligns with the Com-mon Model.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

25 extracted references · 24 canonical work pages

  1. [1]

    F., Klink, P

    Storm, J. F., Klink, P. C., Aru, J., Senn, W., Goebel, R., Pigorini, A., Avanzini, P., Wanduf-fel, W., Roelfsema, P. R., Massimini, M., Larkum, M. E., Pennartz, C. M. A. An integrative, multiscale view on neural theories of consciousness. Neuron 112, 1531-1552 (2024)

    work page 2024

  2. [2]

    J.: A Cognitive Theory of Consciousness

    Baars, B. J.: A Cognitive Theory of Consciousness. Cambridge University Press, Cam-bridge, UK (1988)

    work page 1988

  3. [3]

    Neuron 70(2), 200-227 (2011)

    Dehaene, S., Changeux, J-P.: Experimental and Theoretical Approaches to Conscious Pro-cessing. Neuron 70(2), 200-227 (2011)

    work page 2011

  4. [4]

    Nature Reviews Neuroscience 17, 450–461 (2016)

    Tononi, G., Boly, M., Massimini, M., Koch, C.: Integrated information theory: From con-sciousness to its physical substrate. Nature Reviews Neuroscience 17, 450–461 (2016)

    work page 2016

  5. [5]

    Lamme, V. A. F.: Towards a true neural stance on consciousness. Trends in Cognitive Sci-ences 10(11), 494–501 (2006)

    work page 2006

  6. [6]

    Philosophy and the Mind Sciences 1(II)

    Hohwy, J., Seth, A.: Predictive processing as a systematic basis for identifying the neural correlates of consciousness. Philosophy and the Mind Sciences 1(II). (2020)

    work page 2020

  7. [7]

    Pennartz, C. M. A.: What is neurorepresentationalism? From neural activity and predictive processing to multi-level representations and consciousness. Behavioural Brain Research 432, 113969 (2022)

    work page 2022

  8. [8]

    E.: Cellular mechanisms of conscious processing

    Aru, J., Suzuki, M., Larkum, M. E.: Cellular mechanisms of conscious processing. Trends in Cognitive Sciences 24(10), 814–825 (2020)

    work page 2020

Show all 25 references
  1. [9]

    K., Bayne, T.: Theories of consciousness

    Seth, A. K., Bayne, T.: Theories of consciousness. Nature Reviews Neuroscience 23, 439–452 (2022)

    2022

  2. [10]

    L.: A landscape of consciousness: Toward a taxonomy of explanations and impli-cations

    Kuhn, R. L.: A landscape of consciousness: Toward a taxonomy of explanations and impli-cations. Progress in Biophysics and Molecular Biology 190, 28–169 (2024)

    2024

  3. [11]

    E., Lebiere, C., Rosenbloom, P

    Laird, J. E., Lebiere, C., Rosenbloom, P. S.: A Standard model of the mind: Toward a com-mon computational framework across artificial intelligence, cognitive science, neurosci-ence, and robotics. AI Magazine 38(4), 13-26 (2017)

    2017

  4. [12]

    Common Model of Cognition

    Stocco, A., Steine-Hanson, Z., Koh, N., Laird, J. E., Lebiere, C., Rosenbloom, P. S.: Anal-ysis of the human connectome data supports the notion of a “Common Model of Cognition” for human and human-like intelligence. Neuroimage 235, 118035 (2021)

    2021

  5. [13]

    Trends in Cognitive Sciences 9(2), 46–52 (2005)

    Block, N.: Two neural correlates of consciousness. Trends in Cognitive Sciences 9(2), 46–52 (2005)

    2005

  6. [14]

    K.: 40 years of cognitive architectures: Core cognitive abilities and practical applications

    Kotseruba, I., Tsotsos, J. K.: 40 years of cognitive architectures: Core cognitive abilities and practical applications. Artificial Intelligence Review 53(1), 17-94 (2020)

    2020

  7. [15]

    S., Barto, A

    Sutton, R. S., Barto, A. G.: Reinforcement Learning: An Introduction. 2nd edn. MIT Press, Cambridge, MA (2018)

    2018

  8. [16]

    R., Bothell, D., Byrne, M

    Anderson, J. R., Bothell, D., Byrne, M. D., Douglass, S., Lebiere, C., Qin, Y.: An integrated theory of the mind. Psychological Review 111(4), 1036-1060 (2004) Mapping Theories of Consciousness onto the CMC 7

    2004

  9. [17]

    S.: A Cognitive odyssey: From the Power Law of Practice to a general learning mechanism and beyond

    Rosenbloom, P. S.: A Cognitive odyssey: From the Power Law of Practice to a general learning mechanism and beyond. Tutorials in Quantitative Methods for Psychology, 2(2), 43-51 (2006)

    2006

  10. [18]

    S., Laird, J

    Rosenbloom, P. S., Laird, J. E., Lebiere, C., Stocco, A., Granger, R. H., Huyck, C.: A pro-posal for extending the Common Model of Cognition to emotion. In Proceedings of the 22nd International Conference on Cognitive Modeling (2024)

    2024

  11. [19]

    E., Lebiere, C., Rosenbloom, P

    Laird, J. E., Lebiere, C., Rosenbloom, P. S., Stocco: A proposal to extend the Common Model of Cognition with metacognition. arXiv:2506.07807 (2025)

    2025 arXiv

  12. [20]

    J., Geld, N., Kozma, R.: Global workspace theory (GWT) and prefrontal cortex: Recent developments

    Baars, B. J., Geld, N., Kozma, R.: Global workspace theory (GWT) and prefrontal cortex: Recent developments. Frontiers in Psychology 12(749868) (2021)

    2021

  13. [21]

    Scholarpedia on Integrated Information Theory, http://www.scholarpedia.org/article/Inte-grated_information_theory, last accessed 2025/4/27

    2025

  14. [22]

    K.: A science of consciousness beyond pseudo-science and pseudo-consciousness

    Gomez-Marin, A., Seth, A. K.: A science of consciousness beyond pseudo-science and pseudo-consciousness. Nature Neuroscience 28(4) (2025)

    2025

  15. [23]

    R., Laird, J

    Newell, A., Yost, G. R., Laird, J. E., Rosenbloom, P. S., Altmann, E.: Formulating the prob-lem-space computational model. In: Rashid, R. F. (ed.) CMU Computer Science: A 25th Anniversary Commemorative, pp 255-293. ACM Press/Addison-Wesley, New York, NY (1991)

    1991

  16. [24]

    E.: The Soar Cognitive Architecture

    Laird, J. E.: The Soar Cognitive Architecture. MIT Press, Cambridge, MA (2012)

    2012

  17. [25]

    S., Demski, A., Ustun, V.: The Sigma cognitive architecture and system: Towards functionally elegant grand unification

    Rosenbloom, P. S., Demski, A., Ustun, V.: The Sigma cognitive architecture and system: Towards functionally elegant grand unification. Journal of Artificial General Intelligence 7(1), 1–103 (2016)

    2016

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.