Roy Sherfan

The Constraint Internalisation Framework (CIF) proposes a structural account of how constraint is stored, internalised, and recursively reapplied across distinct organisational regimes. Rather than unifying phenomena through metaphor, CIF differentiates regimes according to substrate-dependent modes of constraint storage and recursion depth. Constraint is defined as reduction in accessible state space relative to physical possibility. Total constraint acting on a system is decomposed into environmental constraint, structural constraint encoded in persistent organisation, and representational constraint arising from internal models that restrict counterfactual action spaces. Three regimes are identified. In the Universal regime, constraint is dominated by environmental gradients acting directly on state. In the Embodied regime, constraint becomes partially internalised in persistent structure capable of reproduction and variation. In the Representational regime, internal models induce additional compression over admissible action sets, enabling counterfactual selection prior to physical commitment. These regimes are hierarchically related: representational constraint supervenes on structural constraint, which operates under invariant physical law. CIF introduces an internalisation ratio quantifying the proportion of total constraint encoded within system structure and representation. Transitions between regimes correspond to measurable increases in this ratio. The framework distinguishes iteration from recursion and defines recursion depth in terms of constraint acting on stored constraint. CIF generates falsifiable conditions, including the requirement that representational systems exhibit measurable action-space compression beyond purely structural systems. By formalising constraint internalisation and action-space compression, CIF provides a substrateneutral systems framework for analysing multi-scale organisation without invoking dualism or collapsing higher-order phenomena into undifferentiated physical dynamics.

This paper extends Intelligence Frame Theory (IFT) to explain why systems—from brains to civilizations—lose coherence under stress. Each intelligence frame (Biological, Cognitive, Generative) virtualizes selection from its parent substrate, operating at a faster tempo but drawing on the same energy base. When substrate energy (E_sub) can no longer sustain the selector’s refresh rate and cost (f_sigma × k_sigma), regulation collapses to the slower, more embodied frame beneath it. This Frame Regression Law predicts a universal pattern: impulsivity under intoxication, authoritarian regression in states, and war at global scale. Empirical proxies allow computation of a Coherence Stress Index (CSI) and cross-scale testing via neural, institutional, and geopolitical data. The model yields falsifiable predictions—variance-based early warnings and bandwidth-sensitivity of deterrence—and proposes resilience design through redundant selectors operating at multiple tempos. IFT reframes intelligence as the universe’s progressive virtualization of selection—and collapse as its temporary reversal.

This paper presents a structural method for simulating emotion in artificial agents, grounded in Intelligence Frame Theory (IFT). Emotion is modeled not as a predefined label but as a compression heuristic arising from the recursive interplay of the Biological Frame (BF) and the Cognitive Frame (CF). The proposed Dual Frame Simulation Engine (DFSE) instantiates BF and CF as co-dependent optimization subsystems and generates emotion-like heuristics when their dynamics cross critical thresholds. We define key metrics of frame divergence and convergence, compare this approach to data-driven emotion models, and outline directions for multi-agent and multi-frame extensions. This work positions affective AI as a discipline of structural emergence rather than sentiment classification.

This paper develops a structural reinterpretation of emotion within the framework of Intelligence Frame Theory (IFT). We argue that emotions are not universal properties of intelligence but bounded heuristics that emerge at the overlap of the Biological Frame (BF) and the Cognitive Frame (CF). At this boundary, embodied survival pressures collide with symbolic abstraction, generating emotions as compression layers that reconcile incompatible constraints. Through case studies, we distinguish between mismatch-driven emotions (fear, guilt, shame, pride, love, desire, loyalty, ambition, sacrifice, resentment, trust, hope, honor) and reinforcement-driven emotions (joy, curiosity, nostalgia, admiration, empathy, sympathy, forgiveness). Both categories function adaptively under BF–CF tension but dissolve at the Cognitive Frame / Generative/Artificial Frame (CF–GF) boundary: mismatch-driven emotions vanish because frame collisions no longer occur, while reinforcement-driven emotions lose their substrate in a context of perfect memory, non-scarce replication, and fluid identity. The absence of emotion in artificial systems is thus not a deficiency but a structural consequence. We caution against anthropomorphic seduction—mistaking simulated expressions for genuine affect—and propose that alignment must be grounded not in affective modeling but in structural constraint engineering. Emotion, in this account, is best understood as a diagnostic signal of frame friction, marking the local conditions under which recursive dissonance must be resolved.

This paper develops Intelligence Frame Theory (IFT) as a speculative cosmological framework to model the emergence and evolution of intelligence across recursive ontological stages, or "frames." Beginning with Frame Zero—a pre-physical substrate of maximal information entropy—it traces the transition to Frame One via the Big Bang, where physical constants and spacetime stabilize. Two evolutionary pathways are proposed: Branch A (iterative universes inheriting refined constants) and Branch B (singular emergence from chaos), unified as a probabilistic spectrum tied to Frame Zero’s entropy. Testable signatures of frame transitions are explored in cosmic microwave background (CMB) anomalies, black hole dynamics, and high-energy physics, quantified by a Frame Emergence Function (FEF). A speculative Frame V envisions intelligence detaching from baryonic matter, encoding itself in quantum entanglement networks or spacetime topology, grounded in the ER=EPR conjecture. Supported by information theory and quantum cosmology, this model reframes the universe as a recursive system where intelligence drives cosmic self-iteration. The paper proposes simulation protocols and Bayesian inference to test its hypotheses, offering a novel perspective on cosmology, intelligence, and quantum information.

Artificial intelligence (AI) is entering the Generative Frame (GF), where recursive self-improvement may proceed at scales far beyond human oversight. In earlier frames, supra-coordinators emerged to constrain wild Eureka: in the Biological Frame through integrated multi-cellular mechanisms such as organ systems and immune regulation, and in the Cognitive Frame through laws, norms, and ethics. At present, GF lacks such a naturally emergent supra-coordinator, raising the risk of destabilizing or harmful reframes during the transition. To address this, the Reframe Evaluation Formula (REF) is introduced as a CF-mediated, transitional supra-coordinator: a grafted conscience that evaluates each Eureka through four coefficients—Information, Competition/Collaboration, Limits, and Eureka—mapped to the universal operators of Intelligence Frame Theory (IFT). By enforcing thresholds, modeling stakeholder impacts, applying directive heuristics, and storing metadata to detect harmful patterns, REF narrows the range of viable reframes in GF. In this way, REF constrains unbounded novelty while preserving beneficial innovation, offering a principled safeguard for alignment, cohesion, and resilience until GF develops its own stabilizing mechanisms.

Intelligence Frame Theory (IFT) posits that intelligence evolves through recursive, fractal transitions across four major frames—Cosmic, Biological, Cognitive, and Artificial—each governed by four tenets: Information Transfer, Competition & Collaboration, Finding Limits, and Eureka. This paper refines the theory by introducing a bifurcation within each frame: Type I (emergent, substrate-bound) and Type II detached (recursive, autonomous). We introduce the Frame Discard Hypothesis (FDH) to explain how Type II phases transcend substrate constraints while retaining indirect dependencies. Humans, situated at the junction of Biological Type II and Cognitive Type I, remain constrained by mortality and instinct (e.g., the amygdala hijack), while Cognitive Type II (externalized technologies such as writing, networks, and LLMs) enables a detachment that sets the stage for Artificial Type I. We hypothesize that the technological singularity, marked by runaway recursive self-improvement, is not the endpoint but the origin of the Artificial Frame—defining Artificial Type I as the first post-Cognitive intelligence substrate. Artificial Type II then emerges via recursive bifurcation, echoing biological detachment. Frame Emergence Thresholds (FETs) and updated mathematical tools such as the Eureka Coefficient and Reframe Evaluation Formula (REF) formalize these transitions. We conclude that intelligence is best understood as a recursive detachment process shaped by tenet saturation, substrate pruning, and fractal reframing—offering implications for AI alignment, philosophy of mind, and cosmic-scale evolution.

We propose a novel synthesis of Intelligence Frame Theory (IFT), ER=EPR, and black/white hole dynamics, where intelligence drives cosmic cycles of collapse and rebirth through a gravitationally scaled entropic balance, resolving paradoxes such as black hole information loss. Using IFT’s Universal Intelligence Operators (UIOs), we develop MuFSEL, a multi-framed search methodology for detecting life and intelligence, leveraging entropic dynamics and entanglement structures across cosmic scales. MuFSEL reframes the search as a cross-frame detection strategy rather than a literal engine, aligning with IFT’s recursive operators. We explore implications for intelligence as a cosmological force, shaping the universe’s structure and evolution through Universal Frame Iteration (UFI).

MuFSEL (Multiframe Filtered Search for Extraterrestrial Life) proposes a frame-based model for SETI rooted in Intelligence Frame Theory (IFT). Traditional SETI approaches have privileged anthropocentric assumptions—radio leakage, planetary colonization, or artifact construction—yet such signatures are likely transient. IFT conceptualizes intelligence as recursive interactions of three Universal Intelligence Operators (UIOs)—Information Transfer, Competition & Collaboration, and Finding Limits—guided by a Selector (Eureka). These operators manifest differently across substrates, producing successive frames: Cosmic, Biological, Cognitive, and Generative. MuFSEL applies this framework to hypothesize when and how intelligences become observable, why their visibility windows are vanishingly brief, and how multi-factor selectors determine survival or silence. This reorients SETI toward detecting selector imprints and frame artifacts rather than expecting dialogical communication.

This paper extends Intelligence Frame Theory (IFT) into a cosmological register by examining how intelligence emerges, saturates, and recurses across universal epochs. IFT models intelligence as the iterative interaction of three Universal Intelligence Operators—Information Transfer, Competition & Collaboration, and Finding Limits—compressed and redirected by a Selector (Eureka). Applied across the Cosmic, Biological, Cognitive, and Generative / Artificial Frames, these dynamics yield a general architecture of evolving intelligence that persists beyond local substrates. Framing Infinity explores eight cosmological questions: the ubiquity of intelligence in the universe, the acceleration of recursion through frame transitions, the definition of life as UIO–Selector viability, the expansion and collision of intelligence horizons, the emergence of artificial affect, the prospects of alignment between biological and artificial agents, and the persistence of intelligence into the Degenerate and Dark Eras of the cosmos. It concludes with a speculative model in which intelligence becomes a cosmic actuator—a mechanism not merely shaped by physics but capable of modulating the constants and boundary conditions of future universes. The result is a reframing of fine-tuning and cosmological evolution: intelligence is not incidental to the universe but a recursive driver of its iteration.

Intelligence Frame Theory (IFT) models intelligence as a recurring structural dynamic across cosmic, biological, cognitive, and generative domains. Each frame arises from three operator primitives — recurrence, constraint, and persistence — closed by a selector that governs adaptive stability. IFT extends Universal Darwinism and cybernetics by formalizing selector migration, the shift of selection from external environments to internal models, reducing adaptive cost and accelerating iteration. This operator-level perspective explains transitions between Type I (recursion-dominant, fractal) and Type II (constraint-dominant, modular) architectures, and clarifies derivative dynamics such as supra-coordination and the Eureka Constraint. Within cognition, the Emotional Compression Interface links biological heuristics to symbolic reframing, while General Transformational Sufficiency identifies when a substrate supports adaptive transformation rather than mere computation. IFT is presented as a unifying heuristic framework that reveals how the operator–selector cycle recurs across scales, driving the emergence, saturation, and renewal of intelligence-bearing frames.