pith. sign in

arxiv: 2510.03881 · v1 · submitted 2025-10-04 · 🧬 q-bio.NC

Intrinsic cause-effect power: the tradeoff between differentiation and specification

William G. P. Mayner , William Marshall , Giulio Tononi This is my paper

Pith reviewed 2026-05-18 10:27 UTC · model grok-4.3

classification 🧬 q-bio.NC
keywords integrated information theorycause-effect powerdifferentiationspecificationintrinsic existencemacro unitsneural systems
0
0 comments X

The pith

A tradeoff between differentiation and specification is necessary for intrinsic cause-effect power in macro systems.

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

The paper shows that substrate units must make a broad repertoire of cause-effect states available while also favoring one specific state to satisfy the requirements for intrinsic and specific cause-effect power. These two requirements are quantified separately through intrinsic differences from maximal specification and from maximal differentiation. When the same requirements are applied to larger macro units and systems of such units, as in neural systems, the two demands stand in tension and a balance between them is required. A sympathetic reader would care because the argument supplies an operational test for whether a physical substrate can support the existence property that the theory links to consciousness.

Core claim

To have cause-effect power, to have it intrinsically, and to have it specifically, substrate units in their actual state must both ensure the intrinsic availability of a repertoire of cause-effect states and increase the probability of a specific cause-effect state. Requirement (ii) can be assessed by the intrinsic difference of a state's probability from maximal differentiation, while requirement (i) can be assessed by the intrinsic difference from maximal specification. When applied to macro units and systems of macro units such as neural systems, a tradeoff between differentiation and specification is a necessary condition for intrinsic existence, i.e., for consciousness.

What carries the argument

The intrinsic difference, which quantifies how far the probability of the actual state deviates from the extremes of maximal differentiation and maximal specification to meet both availability and favoring requirements.

If this is right

  • Macro-level units in a system must exhibit the tradeoff to satisfy the existence requirements.
  • The tradeoff becomes visible only when the analysis moves from micro units to macro units.
  • Integrated information for neural systems depends on the presence of this balance between repertoire availability and specific favoring.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same measures could be applied to artificial networks to check whether they meet the minimal conditions for intrinsic existence.
  • Systems that achieve only one extreme without the other would be predicted to lack the existence property at the macro scale.
  • This framing connects abstract properties of experience to measurable cause-effect relations without assuming additional mechanisms.

Load-bearing premise

That the operational requirements for intrinsic cause-effect power can be fully captured by measuring intrinsic differences from maximal differentiation and maximal specification in the actual state of substrate units.

What would settle it

A concrete calculation on a system of macro units that yields high intrinsic cause-effect power while showing no tradeoff between differentiation and specification would settle the claim by showing the tradeoff is not necessary.

Figures

Figures reproduced from arXiv: 2510.03881 by Giulio Tononi, William G. P. Mayner, William Marshall.

Figure 1
Figure 1. Figure 1: Intrinsic information captures the tradeoff between determinism and indeterminism. [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Example 1: (A) a single-unit system (a monad). The unit stays in the current state with probability p, and switches to its other state with probability 1 − p. (B) The transition probability matrix when p = 0.9. (C) The intrinsic differentiation and intrinsic specification of the monad as a function of p. The intrinsic information (and thus integrated information) is maximized at p = 0.744, where φs = 0.427… view at source ↗
Figure 3
Figure 3. Figure 3: Example 2: Exploring the behavior of intrinsic information as a function of system size [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Example 3: Intrinsic units [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
read the original abstract

Integrated information theory (IIT) starts from the existence of consciousness and characterizes its essential properties: every experience is intrinsic, specific, unitary, definite, and structured. IIT then formulates existence and its essential properties operationally in terms of cause-effect power of a substrate of units. Here we address IIT's operational requirements for existence by considering that, to have cause-effect power, to have it intrinsically, and to have it specifically, substrate units in their actual state must both (i) ensure the intrinsic availability of a repertoire of cause-effect states, and (ii) increase the probability of a specific cause-effect state. We showed previously that requirement (ii) can be assessed by the intrinsic difference of a state's probability from maximal differentiation. Here we show that requirement (i) can be assessed by the intrinsic difference from maximal specification. These points and their consequences for integrated information are illustrated using simple systems of micro units. When applied to macro units and systems of macro units such as neural systems, a tradeoff between differentiation and specification is a necessary condition for intrinsic existence, i.e., for consciousness.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper extends Integrated Information Theory (IIT) by operationalizing intrinsic cause-effect power through two requirements on substrate units in their actual state: (i) ensuring intrinsic availability of a cause-effect repertoire, assessed via the intrinsic difference from maximal specification, and (ii) increasing the probability of a specific cause-effect state, assessed via the intrinsic difference from maximal differentiation. These are illustrated on simple micro-unit systems, with the conclusion that a tradeoff between differentiation and specification is a necessary condition for intrinsic existence (hence consciousness) when the framework is applied to macro units and neural systems.

Significance. If the macro-unit extension holds, the work refines IIT's criteria for existence by identifying a necessary tradeoff, potentially clarifying how intrinsic cause-effect power manifests in composite systems. A strength is the parameter-free character of the intrinsic-difference derivations, which avoids introducing new free parameters beyond the existing IIT formalism.

major comments (2)
  1. [Abstract and concluding section] Abstract and concluding section: the central claim that 'a tradeoff between differentiation and specification is a necessary condition for intrinsic existence, i.e., for consciousness' when applied to macro units and systems of macro units such as neural systems is asserted after illustrations confined to micro-unit systems. No derivation, formula adjustment for coarse-graining, or concrete macro-unit example is supplied to confirm that the same operational requirements and tradeoff continue to hold after the substrate units become composites.
  2. [Section introducing the intrinsic difference from maximal specification] Section introducing the intrinsic difference from maximal specification: the claim that this measure fully captures requirement (i) (repertoire availability) is load-bearing for the tradeoff conclusion, yet the manuscript does not demonstrate that the measure remains well-defined and independent of the differentiation measure once the units are themselves macro-scale.
minor comments (2)
  1. [Definitions of intrinsic differences] Clarify in the text whether the intrinsic-difference quantities are computed on the actual state only or also involve counterfactuals over possible states, to avoid ambiguity in the definitions.
  2. [Introduction] Add explicit cross-references to the prior IIT papers that supply the cause-effect power formalism, so readers can distinguish the new measures from the established axioms.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments and for noting the parameter-free character of the intrinsic-difference derivations. We address each major comment below and indicate the revisions that will be incorporated.

read point-by-point responses

  1. Referee: [Abstract and concluding section] Abstract and concluding section: the central claim that 'a tradeoff between differentiation and specification is a necessary condition for intrinsic existence, i.e., for consciousness' when applied to macro units and systems of macro units such as neural systems is asserted after illustrations confined to micro-unit systems. No derivation, formula adjustment for coarse-graining, or concrete macro-unit example is supplied to confirm that the same operational requirements and tradeoff continue to hold after the substrate units become composites.

    Authors: The operational requirements and the resulting tradeoff are formulated at the level of any substrate units in their actual state, with cause-effect repertoires determined by the units' transition probabilities. Because the intrinsic-difference measures are derived directly from these repertoires, the same logic applies without modification when units are coarse-grained into macro units; the effective transition probabilities of the macro units simply replace those of the micro units. We agree that the manuscript would be strengthened by an explicit statement of this generality. We will revise the abstract and concluding section to note that the requirements and tradeoff hold for macro units and systems of macro units by the same definitions. revision: yes

  2. Referee: [Section introducing the intrinsic difference from maximal specification] Section introducing the intrinsic difference from maximal specification: the claim that this measure fully captures requirement (i) (repertoire availability) is load-bearing for the tradeoff conclusion, yet the manuscript does not demonstrate that the measure remains well-defined and independent of the differentiation measure once the units are themselves macro-scale.

    Authors: The intrinsic difference from maximal specification quantifies repertoire availability by measuring the deviation of the actual cause-effect probability distribution from a maximally specified distribution. By construction this is independent of the intrinsic difference from maximal differentiation, which instead quantifies the concentration on a particular state. Both quantities are defined solely on the probability distributions over cause-effect states for the given units; nothing in the definitions presupposes micro-scale units. Consequently the measures remain well-defined and independent when the units are macro-scale composites whose states and transitions are obtained by coarse-graining. We will add a clarifying paragraph in the relevant section to make this independence and scale invariance explicit. revision: yes

Circularity Check

0 steps flagged

No significant circularity in the paper's derivation

full rationale

The paper extends the IIT framework by introducing an assessment for the intrinsic availability of a repertoire using the intrinsic difference from maximal specification, complementing the previously shown assessment for increasing the probability of a specific state. These are demonstrated in simple micro-unit systems. The statement that a tradeoff is necessary for macro units and neural systems follows logically from the combined operational requirements for intrinsic cause-effect power. The derivation does not reduce to a self-referential definition or fitted input presented as prediction; the self-citation to prior work is for one component, while the current contribution provides new analysis and illustrations, making the overall argument self-contained within the theoretical framework without circular reduction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper relies on IIT's foundational assumptions about consciousness properties and cause-effect power without introducing new free parameters or invented entities in the provided abstract.

axioms (1)
  • domain assumption IIT starts from the existence of consciousness and characterizes its essential properties operationally in terms of cause-effect power of a substrate of units.
    Invoked in the opening sentences to ground the operational requirements.

pith-pipeline@v0.9.0 · 5724 in / 1223 out tokens · 31370 ms · 2026-05-18T10:27:59.732882+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Quantifying Spacetime Integration across a Partition with Synergy

    cs.IT 2026-04 unverdicted novelty 7.0

    Synergy-based measures from partial information decomposition are found more suitable than current practice for quantifying integration in simple deterministic networks for the Information Integration Theory of Consciousness.

  2. Quantifying Spacetime Integration across a Partition with Synergy

    cs.IT 2026-04 unverdicted novelty 7.0

    Synergy-based measures of spacetime integration outperform current IIT practice when tested on simple deterministic networks.

  3. Quantifying Spacetime Integration across a Partition with Synergy

    cs.IT 2026-04 unverdicted novelty 6.0

    Introduces four synergy-based measures of spacetime integration from partial information decomposition and finds them more suitable than current IIT practice for simple deterministic networks.

Reference graph

Works this paper leans on

15 extracted references · 15 canonical work pages · cited by 1 Pith paper

  1. [1]

    M., Montabes de la Cruz, B

    Albantakis, L.; Barbosa, L.; Findlay, G.; Grasso, M.; Haun, A.M.; Marshall, W.; Mayner, W.G.; Zaeemzadeh, A.; Boly, M.; Juel, B.E.; et al. Integrated information theory (IIT) 4.0: Formulating the properties of phenomenal existence in physical terms.PLOS Computational Biology2023, 19, e1011465. https://doi.org/10.1371/journal.pcbi.1011465

    work page doi:10.1371/journal.pcbi.1011465

  2. [2]

    System integrated information.Entropy,25, 334

    Marshall, W.; Grasso, M.; Mayner, W.G.P .; Zaeemzadeh, A.; Barbosa, L.S.; Chastain, E.; Findlay, G.; Sasai, S.; Albantakis, L.; Tononi, G. System integrated information.Entropy,25, 334. https://doi.org/10.3390/e25020334

    work page doi:10.3390/e25020334

  3. [3]

    An information integration theory of consciousness

    Tononi, G. An information integration theory of consciousness.BMC Neuroscience2004,5, 42. https://doi.org/10.1186/1471-2202-5-42

    work page doi:10.1186/1471-2202-5-42

  4. [4]

    Integrated Information in Discrete Dynamical Systems: Motivation and Theoretical Framework.PLOS Computational Biology2008,4, e1000091

    Balduzzi, D.; Tononi, G. Integrated Information in Discrete Dynamical Systems: Motivation and Theoretical Framework.PLOS Computational Biology2008,4, e1000091. https://doi.org/10 .1371/journal.pcbi.1000091

    work page

  5. [5]

    From the Phenomenology to the Mechanisms of Consciousness: Integrated Information Theory 3.0

    Oizumi, M.; Albantakis, L.; Tononi, G. From the Phenomenology to the Mechanisms of Con- sciousness: Integrated Information Theory 3.0.PLOS Computational Biology2014,10, e1003588. https://doi.org/10.1371/journal.pcbi.1003588

    work page doi:10.1371/journal.pcbi.1003588

  6. [6]

    IIT — an online resource on Integrated Information Theory

    IIT Wiki. IIT — an online resource on Integrated Information Theory. https://www.iit.wiki. Accessed: 2025-09-25

    work page 2025

  7. [7]

    Intrinsic Units: Identifying a System’s Causal Grain.bioRxiv2024

    Marshall, W.; Findlay, G.; Albantakis, L.; Tononi, G. Intrinsic Units: Identifying a System’s Causal Grain.bioRxiv2024. https://doi.org/10.1101/2024.04.12.589163

    work page doi:10.1101/2024.04.12.589163 2024

  8. [8]

    A measure for intrinsic infor mation

    Barbosa, L.S.; Marshall, W.; Streipert, S.; Albantakis, L.; Tononi, G. A measure for intrinsic information.Scientific Reports2020,10, 18803. https://doi.org/10.1038/s41598-020-75943-4

    work page doi:10.1038/s41598-020-75943-4

  9. [9]

    Pearl, J.Causality: Models, Reasoning, and Inference, 2nd ed.; Cambridge University Press, 2009

    work page 2009

  10. [10]

    PyPhi: A Toolbox for Integrated Information Theory.PLoS computational biology,14, e1006343, [30048445]

    Mayner, W.G.P .; Marshall, W.; Albantakis, L.; Findlay, G.; Marchman, R.; Tononi, G. PyPhi: A Toolbox for Integrated Information Theory.PLoS computational biology,14, e1006343, [30048445]

    work page

  11. [11]

    Collapse Models: A Theoretical, Experimental and Philosoph- ical Review.Entropy2023,25, 645

    Bassi, A.; Dorato, M.; Ulbricht, H. Collapse Models: A Theoretical, Experimental and Philosoph- ical Review.Entropy2023,25, 645. https://doi.org/10.3390/e25040645. Version October 7, 2025 submitted toEntropy 14 of 14

    work page doi:10.3390/e25040645 2025

  12. [12]

    Potentiality Realism: A Realistic and Indeterministic Physics Based on Propensities.European Journal for Philosophy of Science2023,13, 1–16

    Santo, F.D.; Gisin, N. Potentiality Realism: A Realistic and Indeterministic Physics Based on Propensities.European Journal for Philosophy of Science2023,13, 1–16. https://doi.org/10.1007/ s13194-023-00561-6

    work page

  13. [13]

    25 years of criticality in neuroscience — established results, open controversies, novel concepts

    Wilting, J.; Priesemann, V . 25 years of criticality in neuroscience.Current Opinion in Neurobiology 2019,58, 105–111. https://doi.org/10.1016/j.conb.2019.08.002

    work page doi:10.1016/j.conb.2019.08.002 2019

  14. [14]

    , author Plenz, D

    Beggs, J.M.; Plenz, D. Neuronal avalanches in neocortical circuits.Journal of Neuroscience2003, 23, 11167–11177. https://doi.org/10.1523/JNEUROSCI.23-35-11167.2003

    work page doi:10.1523/jneurosci.23-35-11167.2003 2003

  15. [15]

    Integrated information as a common signature of dynamical complexity.Chaos: An Interdisciplinary Journal of Nonlinear Science2022,32, 013115

    Mediano, P .A.; Seth, A.K.; Barrett, A.B. Integrated information as a common signature of dynamical complexity.Chaos: An Interdisciplinary Journal of Nonlinear Science2022,32, 013115. https://doi.org/10.1063/5.0063380

    work page doi:10.1063/5.0063380