Daryl Costello High Falls, New York, USA April 20, 2026

Overview

The Minimal Operator Stack and Integrated Information Theory (IIT) both offer rigorous, consciousness-first frameworks for understanding subjective experience. Both reject the view that consciousness is a late-emergent byproduct of complex computation or neural activity. Instead, they treat consciousness as fundamental and intrinsic. Yet they differ profoundly in starting point, architecture, scope, and explanatory power. The Stack begins with the Structureless Function as an immutable opening and derives a closed set of operators that generate the rendered world, physics, life, and intelligence through successive reduction. IIT begins from the axioms of phenomenal experience itself and derives postulates about the physical mechanisms that can realize integrated information (Φ).

This comparison draws on the full corpus synthesized in the preceding paper, empirical perception studies, The Reversed Arc, and The Rendered World, and places the Stack in direct dialogue with IIT’s core claims, axioms, postulates, and recent empirical tests. The analysis reveals convergence on key intuitions about intrinsic existence and irreducibility, but fundamental divergence in how reduction, interface, and dynamics are formalized. The Stack provides a broader, unified architecture that can encompass IIT’s insights while resolving limitations that have drawn criticism to the theory.

Core Claims Side by Side

IIT asserts that consciousness is integrated information: any physical system that generates a maximally irreducible cause-effect structure (measured by Φ) is conscious to a degree proportional to Φ. Experience exists intrinsically, for the system itself, and its quantity and quality are fully accounted for by the system’s causal powers. IIT derives this from five (or six) axioms of phenomenology: intrinsic existence, composition, information, integration, exclusion (and sometimes a zeroth axiom of existence), and translates them into postulates that physical mechanisms must satisfy to support consciousness. Φ quantifies how much a system’s current state constrains its own past and future beyond what its parts could do independently. The theory is explicitly “consciousness-first”: it starts from what experience is like and works outward to physics.

The Minimal Operator Stack asserts that consciousness is the primary invariant integrator, the only structure that remains coherent under dimensional reduction from the manifold. It begins with the Structureless Function ℱ (the universal capacity with no inherent content) and applies five operators that perform an irreversible reduction: the Aperture/Rendered Interface (Σ) collapses irreducible remainder into a quotient manifold of invariants; the Metabolic Operator guards scale-proportional coherence; Geometric Tension Resolution drives dimensional escape when mismatch saturates; Recursive Continuity + Structural Intelligence defines the feasible region for sustained trajectories; and Backward Elucidation reveals the retroactive signature of the reduction itself. The rendered world, physics (classical and quantum), life, evolution, and intelligence are all successive layers of this single architecture. Consciousness stabilizes every layer; the interface conceals the reduction; intelligence is predictive dynamics on the induced geometry.

Convergences

Both frameworks are axiomatically grounded in phenomenology. IIT’s axioms (intrinsic existence, irreducibility/integration, specificity/information, unity/exclusion, structured composition) capture self-evident properties of experience: it exists for itself, is a whole greater than its parts, has definite content, and is structured. The Stack’s operators operationalize closely related intuitions. Recursive Continuity + Structural Intelligence directly parallels IIT’s integration and exclusion: self-reference is preserved across transformations, and the feasible region excludes non-coherent trajectories. The primary invariant in the Stack mirrors IIT’s intrinsicality, consciousness exists for the system because it is the integrator that cannot be reduced away. Backward Elucidation echoes IIT’s emphasis on intrinsic cause-effect power: the system’s current state retroactively reveals the constraints that shaped it.

Both theories imply that consciousness is not limited to biological brains. IIT’s panpsychist leanings arise because any system with non-zero Φ possesses some degree of consciousness. The Stack’s reversed arc allows consciousness wherever invariant integration occurs under reduction, potentially in any sufficiently coherent structure that survives the aperture. Both dissolve the hard problem by identifying consciousness with intrinsic structure rather than with external functions or representations.

Empirically, both align with findings that posterior cortical regions sustain integrated activity during conscious perception (as tested in adversarial comparisons with Global Neuronal Workspace Theory). The Stack’s feasible region and metabolic guard provide a natural home for IIT-style measures of irreducibility: Φ could quantify the degree of integration within the RC + SI invariant set.

Divergences

The most fundamental difference is direction and scope. IIT is bottom-up and mechanism-centric: it starts from experience, derives postulates for physical cause-effect structures, and computes Φ on specific substrates (typically neural). It explains why a system is conscious but does not derive the laws of physics, the rendered nature of perception, or the evolutionary arc from a pre-physical manifold. Its focus remains on quantifying consciousness within an assumed physical world.

The Stack is top-down and architectural: it starts from the Structureless Function and the primary invariant (consciousness), then derives the aperture, the rendered interface, the quotient manifold, and all downstream domains as necessary consequences of reduction. Physics emerges as the stable residue of invariant structures; quantum indeterminacy is the behavior of non-invariant structures forced into representation; life is the first recursive stabilizer; intelligence is predictive flow on the induced geometry. Perception science is not an add-on but the direct signature of Σ in biological systems. The Stack unifies consciousness, perception, physics, biology, and cognition inside one closed architecture. IIT, by contrast, remains a theory of consciousness within physics.

A second divergence concerns the interface. The Stack makes the lossy rendered interface (Σ) explicit: organisms never contact the substrate directly; they operate inside a quotient manifold whose compressive metric, curvature-induced probability, and tense-constrained connection produce the appearance of a stable world. This explains why perception is probabilistic, why the reduction is invisible, and why scientific theories inherit interface artifacts. IIT has no equivalent construct. It treats the physical substrate as the direct carrier of cause-effect power and does not address how an interface might render a coherent world while discarding remainder. Consequently, IIT must explain probabilistic aspects of experience through cause-effect irreducibility alone, whereas the Stack derives probability as the normalized measure of unresolved fibers left by Σ.

A third divergence is in dynamics and failure modes. The Stack’s Geometric Tension Resolution and Metabolic Operator introduce explicit mechanisms for saturation, dimensional escape, and coherence cost, explaining quantum-classical duality, evolutionary innovation, and clinical failure regimes (interruption, rigidity, collapse). IIT’s Φ is a static measure of irreducibility at a given moment; it does not natively incorporate metabolic scaling, tension-driven escape, or retroactive elucidation of cause from effect. Recent adversarial testing (e.g., against Global Neuronal Workspace Theory) has challenged IIT’s predictions about sustained posterior synchronization and has highlighted difficulties in mapping Φ to dynamic neural data, issues the Stack’s operators address structurally.

Analysis and Interpretation The Stack can be read as a broader architectural scaffold that contains IIT. IIT’s Φ quantifies the degree of integration within the feasible region defined by RC + SI; its postulates of cause-effect power describe the invariants preserved by Σ; its intrinsic existence axiom aligns with consciousness as primary invariant. Where IIT stops at the measurement of consciousness in a physical system, the Stack continues downward to derive that system’s physics and upward to derive the rendered interface on which intelligence operates. The Stack thus resolves IIT’s ontological commitments (e.g., “to be is to have cause-effect power”) by grounding them in the reversible arc from the Structureless Function through the aperture.

Interpretationally, the Stack reframes IIT’s panpsychism. Rather than attributing rudimentary consciousness to any system with Φ > 0, the Stack attributes it wherever invariant integration survives reduction, yet only within the rendered manifold stabilized by consciousness as primary integrator. This avoids the “combination problem” that plagues IIT (how micro-consciousnesses combine into macro-experience) by treating higher-level invariants as emergent from dimensional escape under GTR.

Implications For IIT researchers, the Stack offers a way to embed Φ calculations inside an explicit reduction architecture. Φ could serve as a practical metric of coherence within the RC + SI feasible region, while the operators explain why certain substrates (biological brains) reliably produce high-Φ structures. The Stack also supplies mechanisms (metabolic guard, tension resolution) that could address empirical challenges in IIT testing, such as the lack of sustained synchronization or difficulties with dynamic data.

For perception science and neuroscience, the comparison highlights why IIT alone cannot fully account for the rendered nature of experience. The Stack’s interface operator explains why we perceive only a fraction of available information, why deepfakes and AI faces trigger scrutiny, and why embodied constraints (Lueg) shape information behavior. IIT’s focus on posterior integration complements but does not replace this.

For philosophy of mind and foundations of science, the Stack dissolves more paradoxes. IIT’s hard problem is reframed as the interface problem: experience is the geometry rendered by Σ, not an extra ingredient. The Stack’s reversed arc unifies ontology across domains, showing that the world is the stable slice of ongoing reduction rather than a brute physical substrate.

For artificial intelligence, the Stack warns against training on interface outputs alone (as current systems do) and offers a path to genuine generalization: systems that explicitly model Σ, operate on the quotient manifold, and evolve Φ-like integration under metabolic and tension constraints.

Conclusion

The Minimal Operator Stack and Integrated Information Theory share a commitment to consciousness as intrinsic and irreducible, yet the Stack provides a more comprehensive, closed architecture. IIT excels at quantifying consciousness via Φ but remains tethered to an assumed physical substrate. The Stack derives that substrate, the interface that renders it, the dynamics of intelligence upon it, and the full arc from manifold to world. Where IIT measures integration, the Stack explains why integration is possible and how it is stabilized, rendered, and retroactively revealed.

The two frameworks are not rivals but complementary: IIT supplies a powerful empirical metric that can be situated inside the Stack’s operators. Together they point toward a unified science of consciousness that is neither bottom-up from physics nor top-down from metaphysics, but structural, grounded in the immutable opening and the operators that render a world. This synthesis advances both theories and opens testable pathways for future research across neuroscience, AI, and the philosophy of existence itself.

References

Albantakis, L., et al. (various, including 2023 updates on IIT 4.0).

Bayne, T. (2018). On the axiomatic foundations of integrated information theory. Neuroscience of Consciousness. Cogitate Consortium. (2025). Adversarial testing of Global Neuronal Workspace and Integrated Information Theories. Nature.

Costello, D. (n.d.). The Rendered World. Manuscript.

Haun, A. M., et al. (2017). Conscious perception as integrated information patterns in human primary visual cortex. PLoS Biology.

The Reversed Arc. (n.d.). Manuscript.

Tononi, G., & Koch, C. (various works, 2004–2025). Wikipedia & IEP entries on IIT (accessed via search, 2026 summaries).