RCM
Foundational architecture
Current page
Top
RCM
The Reflexive Coherence Model architecture of reflexive coherence.
The Reflexive Coherence Model interprets consciousness as a possible regime of integrated, self-modeling, temporally stable information dynamics.
Model snapshot
RCM v1.2 Architecture Reflexive coherence
A bridge framework for identifying candidate structural and dynamical conditions associated with reflexive phenomenological organization, without treating them as proof of consciousness.
RCM v1.2 treats reflexive coherence as an operational research target rather than an ontological verdict. It asks whether a system exhibits integrated information dynamics, operational self-modeling, reflexive causal coupling, and temporal stability across a finite window.
The model does not identify consciousness by declaration. It organizes the conditions under which stronger hypotheses about reflexive phenomenological organization could become scientifically meaningful.
This page lays out RCM v1.2 as a bridge framework: theoretical enough to organize questions about consciousness, but operational enough to state what would have to be observed before stronger claims were warranted.
What RCM claims
RCM proposes that consciousness can be studied as a possible regime of integrated, self-modeling, temporally stable information dynamics. It is not offered as proof of consciousness, but as a way to identify candidate structural and dynamical conditions associated with reflexive phenomenological organization.
The model is therefore a bridge framework. It connects philosophical questions about subjective organization with operational questions about what could be observed, compared, and falsified in biological or artificial systems.
v1.2 framing
- Cautious: candidate conditions, not certainty.
- Operational: emphasis on observables and finite windows.
- Structural: organization matters more than substrate.
- Bridge-like: compatible with multiple theories of consciousness.
What RCM does not claim
These boundaries are part of the model. RCM is meant to make stronger claims harder to make casually, not easier.
Non-claim
RCM does not prove that a system is conscious.
Non-claim
High RCI does not imply personhood or moral status by itself.
Non-claim
RCI is not a direct measure of subjective experience.
Non-claim
Current AI systems should not be described as conscious on this basis.
Non-claim
Consciousness need not resemble human experience in every candidate system.
Non-claim
RCM does not resolve the ontological hard problem.
Core components
RCM v1.2 treats reflexive coherence as a conjunction of four components. None is sufficient alone; the framework becomes informative when they appear together and remain stable over a specified window.
Informational integration
Relevant internal variables must be coordinated rather than merely co-present. Integration is necessary, but not sufficient.
Operational self-modeling
The system maintains a state-linked model of its own organization that can participate in control, inference, or regulation.
Reflexive causal coupling
The self-model and the modeled dynamics influence one another in non-trivial feedback loops. This replaces stronger language of absolute causal closure.
Temporal stability across a finite window
The organization must persist across an explicit observational window rather than appearing as a one-step correlation.
Contextual vs systemic stability
Temporal stability is not a vague appeal to persistence. RCM separates short-window stability within a task context from broader stability across the system's operating conditions.
Contextual temporal stability
Reflexive organization remains coherent across a bounded episode, task, or interaction window.
Systemic temporal stability
Reflexive organization remains robust across wider state changes, perturbations, and repeated windows.
RCI as an operational proxy
The Reflexive Coherence Index is best read as an operational proxy for reflexive coherence. It is intended to track observable organization: integration, operational self-modeling, reflexive causal coupling, and temporal stability across a finite window.
High RCI should be interpreted cautiously. It indicates reflexive informational organization, not proof of consciousness and not a direct measurement of subjective experience.
Observable handles
- Integration: coordinated internal variables.
- Self-modeling: state-linked operational model.
- Coupling: bidirectional, non-trivial causal influence.
- Stability: persistence within an explicit window.
Why this matters for AI and consciousness research
RCM gives researchers a cautious vocabulary for discussing candidate reflexive organization without prematurely attributing consciousness. This is especially important for AI systems, where fluent behavior can invite stronger interpretations than the evidence supports.
The value of the framework is comparative: it helps ask which architectures exhibit integrated self-modeling, which forms of coupling are operational rather than decorative, and which patterns remain stable across finite windows.
Connections to other frameworks
RCM is not proposed as a replacement for existing theories, but as a structural constraint layer that can coexist with other approaches. It asks what makes integration, broadcast, or inference become reflexively stabilised into an internal perspective.
IIT
Emphasises integration; RCM adds explicit self-modeling, coupling, and temporal-stability constraints.
GWT
Emphasises broadcast; RCM asks what makes broadcast self-referentially stabilised.
Predictive Processing
Emphasises inference; RCM focuses on when inference becomes reflexive and coherent.
Explore the framework
The three pages form a connected route through architecture, temporal dynamics, and artificial-system interpretation.
Next
Keep exploring the ecosystem: definitions, development history, and primary materials.