Friston, K., Kilner, J., & Harrison, L. · Journal of Physiology-Paris
Introduced the free energy principle — the core theoretical basis for modeling how self-models form and update, which Clarity operationalizes through its belief extraction API.
22 references · 7 themes
The Science Behind Clarity
Clarity's approach to computational self-models is grounded in 60 years of neuroscience, cognitive science, and philosophy of mind. These are the foundational papers and books — from Friston, Levin, Clark, Seth, Metzinger, and others — that inform our API design.
Free Energy Principle
The mathematical framework proposing that biological systems minimize surprise by maintaining predictive models of their environment — the theoretical foundation for Clarity's belief extraction engine.
Friston, K. & Kiebel, S. · Philosophical Transactions of the Royal Society B
Unified predictive coding with free energy minimization, providing the mathematical link between prediction errors and belief updates that Clarity's confidence scoring is built on.
Friston, K. · Nature Reviews Neuroscience
The landmark review establishing free energy as a unifying principle for perception, action, and learning — validating the theoretical basis for a single API that captures belief dynamics.
Life as we know it ↗ 2013
Friston, K. · Journal of the Royal Society Interface
Extended the free energy principle to all living systems, supporting Clarity's premise that self-modeling is a fundamental property of adaptive systems, not just brains.
Predictive Processing
The cognitive science framework describing brains as prediction machines that constantly generate and update models of reality — the computational paradigm behind Clarity's real-time belief tracking.
Rao, R. P. N. & Ballard, D. H. · Nature Neuroscience
The foundational predictive coding paper demonstrating how the brain generates top-down predictions and propagates bottom-up errors — the computational architecture Clarity adapts for belief extraction.
Clark, A. · Behavioral and Brain Sciences
Clark's influential synthesis arguing the brain is fundamentally a prediction machine, providing the cognitive science grounding for Clarity's approach to modeling belief formation and revision.
Clark, A. · Oxford University Press
The comprehensive account of predictive processing as a unified theory of mind, demonstrating how prediction and error-correction drive all cognition — the paradigm Clarity makes computationally explicit.
Consciousness & Interoception
Research on how subjective experience arises from the brain's models of its own internal states — informing how Clarity captures the self-referential aspects of belief systems.
Seth, A. K. · Trends in Cognitive Sciences
Proposed that selfhood emerges from interoceptive predictions about the body, informing Clarity's approach to capturing embodied, felt dimensions of belief — not just propositional content.
Seth, A. K. & Friston, K. J. · Philosophical Transactions of the Royal Society B
Bridged active inference with emotional processing, showing how the brain's self-model integrates body and emotion — the theoretical basis for Clarity's multi-dimensional belief representation.
Seth, A. K. · Dutton/Penguin
Seth's accessible synthesis of consciousness as a 'controlled hallucination' built from predictive models, validating Clarity's core premise: selfhood is a model that can be made explicit and measured.
Multi-Scale Cognition
Michael Levin's research on how intelligence and goal-directed behavior emerge at every biological scale — from cells to organisms — expanding the scope of what self-models can represent.
Levin, M. & Martyniuk, C. J. · BioSystems
Demonstrated that bioelectric signaling serves as a computational medium for morphogenesis, supporting Clarity's multi-scale approach to self-models that operate across biological levels.
Levin, M. · Frontiers in Psychology
Argued that cognitive boundaries are not fixed but dynamically scale — a key insight for Clarity's architecture, which models beliefs as emerging from nested, multi-scale processes.
Levin, M. · Frontiers in Systems Neuroscience
Levin's framework for detecting and characterizing intelligence across substrates, directly informing Clarity's design goal of making self-models measurable regardless of their implementation.
Self-Models
Philosophical and computational theories of how organisms construct models of themselves — the conceptual heart of what Clarity's API extracts and tracks.
Perlis, D. · Journal of Consciousness Studies
Early computational treatment of consciousness as self-referential function, prefiguring Clarity's approach to self-models as recursively structured belief systems that reference their own operation.
Metzinger, T. · MIT Press
The definitive philosophical account of selfhood as a transparent self-model, providing the conceptual framework for Clarity's central claim: the self is a model that can be extracted and analyzed.
Metzinger, T. · Basic Books
Made self-model theory accessible, arguing that subjective experience is a 'tunnel' constructed by the brain's self-model — reinforcing why making these models explicit (as Clarity does) is both possible and valuable.
Epistemic Intelligence
Research on how agents actively seek information to reduce uncertainty and improve their world models — the theoretical basis for Clarity's alignment measurement and confidence scoring.
Kirsh, D. & Maglio, P. · Cognitive Science
Distinguished actions taken to change the world (pragmatic) from actions taken to change what you know (epistemic), providing the conceptual foundation for Clarity's epistemic intelligence metrics.
Friston, K., Rigoli, F., Ognibene, D., Mathys, C., Fitzgerald, T., & Pezzulo, G. · Cognitive Neuroscience
Formalized how biological agents balance exploiting known rewards with exploring to reduce uncertainty, directly informing Clarity's measurement of how beliefs guide information-seeking behavior.
Parr, T. & Friston, K. J. · Journal of the Royal Society Interface
Detailed how epistemic actions reduce expected uncertainty in generative models, providing the mathematical grounding for Clarity's confidence scoring and belief alignment measurement.
Bayesian Cognition
The broader research program treating cognition as probabilistic inference — the mathematical language Clarity uses to represent beliefs, update confidence, and measure alignment.
Knill, D. C. & Pouget, A. · Trends in Neurosciences
Established that neural computation is fundamentally Bayesian, encoding and manipulating probability distributions — the neuroscience basis for Clarity's probabilistic belief representation.
Tenenbaum, J. B., Kemp, C., Griffiths, T. L., & Goodman, N. D. · Science
Showed how Bayesian inference over structured representations explains human concept learning and theory formation — the cognitive science parallel to Clarity's hierarchical belief extraction.
Parr, T., Pezzulo, G., & Friston, K. J. · MIT Press
The comprehensive textbook unifying active inference, free energy, and Bayesian cognition into a single framework — the definitive reference for the theoretical foundations underlying Clarity's entire approach.
These foundations inform every aspect of Clarity — from belief extraction and confidence scoring to alignment measurement. Want to see how the theory becomes code?