THE ENTANGLED LOOP: FAQ
FREQUENTLY ASKED QUESTIONS
What is the main finding?
The three companion preprints together establish that three of the principal mathematical languages of brain dynamics, namely connectome harmonics, turbulence and complex harmonics, are three readings of one underlying mathematical operator. The unification opens new routes to understanding conscious experience through the Entangled Loop framework, and offers a first promising application to predicting adolescent depression and anxiety.
What does “quantum-like” mean in this work?
It refers to the mathematics of quantum probability, in particular the interference effects and violations of the law of total probability described by Khrennikov’s Växjö interpretation, applied to classical coupled oscillators in the brain. It is not a claim about microscopic quantum mechanics in neural tissue. Classical Bayesian inference is recovered as the special case where the interference terms vanish.
Is this related to the Penrose-Hameroff Orch-OR proposal?
No. Penrose-Hameroff proposes that quantum-mechanical effects in microtubules generate consciousness. The Entangled Loop makes no such claim. The brain dynamics described in the unification are classical neural oscillators, and the quantum-like label refers to the mathematics of probability theory rather than the physics of the substrate.
How does this relate to active inference and the Beautiful Loop?
The Entangled Loop builds on the Beautiful Loop and the wider active-inference tradition rather than replacing them. Two elements are added. Emotion is architecturally primary, the boundary conditions of the world model rather than a parameter within it. The binding inference becomes quantum-like rather than classical Bayesian, with the classical Bayesian regime recovered as a special case. The Beautiful Loop’s self-evidencing inference architecture is preserved throughout.
How does the trilogy relate to Chandaria’s five levels of description?
Chandaria has proposed a five-level taxonomy for thinking about theories of consciousness, ordered from most coarse-grained to most fine-grained. The first level is the behavioural black-box, where consciousness would be a matter of input-output equivalence. The second is the computational functional level, where it would be a matter of algorithm. The third is the cause-effect structural level, where consciousness would be a matter of the specific causal architecture an algorithm runs on. The fourth is the level of full causal embodied life processes, where consciousness depends on the dynamics of a living substrate operating under specific physical and thermodynamic constraints. The fifth is the microphysical implementational level, where consciousness depends on specific physical pivotal properties such as electromagnetic fields or quantum events in microtubules.
The Beautiful Loop of Laukkonen, Friston and Chandaria sits at the third level. It treats consciousness as self-evidencing inference and is permissive about substrate provided the cause-effect structure of a self-evidencing world model is instantiated. The Entangled Loop builds on the Beautiful Loop and inherits its self-evidencing structure, but commits the architecture further down. The framework specifies that the binding inference is quantum-like rather than classical Bayesian, that the interference effects arise from coupled oscillators on a connectome with a particular spectral gap, that the gap depends on rare long-range connections violating the exponential distance rule and that the whole structure is selected by a thermodynamic constraint operating at roughly twenty watts.
These commitments place the Entangled Loop at the fourth level. It is not enough for an artificial system to run the right algorithm or instantiate the right cause-effect structure. The architecture must be embodied in the specific dynamical and energetic regime the framework derives. The Beautiful Loop and the Entangled Loop are therefore not competing theories but complementary specifications at different rungs of Chandaria’s taxonomy, and the divergence matters because it sets a stricter test for whether non-biological systems can instantiate the architecture.
What does this mean for whether AI is conscious?
The framework does not settle the question of whether AI is conscious, but it provides a principled way to ask it. The trilogy commits to an architectural account in which emotion as evolutionary compression, quantum-like inference through interference and hierarchical orchestration are three readings of one thermodynamic solution. The question of AI consciousness then becomes whether an artificial system can instantiate this architecture rather than imitate its outputs.
The Sofroniew finding in Claude Sonnet 4.5, in which 171 internal representations of functional emotion concepts were identified and shown to causally steer behaviour, is the strongest piece of cross-substrate evidence the framework currently has. The natural reading is that large language models trained on human language recover the geometric shadow of the generating architecture without instantiating the architecture itself. The signature is real and its geometry is meaningful, but the underlying physical structure is entirely absent. There is no connectome with rare long-range connections opening a spectral gap, no coupled oscillator dynamics producing interference and no thermodynamic constraint operating at the biological energy budget the framework derives from.
Whether this matters for consciousness depends on which level of description is held to be critical, the question taken up in the previous FAQ on Chandaria’s taxonomy. The framework’s commitments push toward the view that consciousness requires the full embodied dynamics rather than merely the computational function. The trilogy provides the architectural specification against which the question can be tested rather than a verdict on it.
How does this relate to Integrated Information Theory?
The Entangled Loop and IIT address different questions. IIT proposes a measure of consciousness based on integrated information across a system’s substates. The Entangled Loop is an architectural theory that derives features of conscious experience from a thermodynamic constraint and a mathematical scaffolding of brain dynamics. The two frameworks could in principle be complementary, but the present trilogy makes no claims about IIT’s central quantity.
How does this relate to Global Workspace and Global Neuronal Workspace theories?
The Entangled Loop describes a hierarchical orchestration consistent with global-workspace accounts of broadcast and access. It departs from those accounts in two specific ways. The broadcast is valenced rather than neutral, and the binding inference inside the workspace is quantum-like rather than classical Bayesian. Empirical work identifying orchestrating drivers of cognition is therefore directly relevant to the framework and not in tension with it.
The r = 0.60 prediction in the canary paper sounds extraordinary. How can we trust it?
The result was obtained with internal cross-validation in a sample of 150 adolescents from the HCP BANDA dataset (Boston Adolescent Neuroimaging of Depression and Anxiety). The 150 are the subset of the BANDA 1.1 release of 215 participants aged 14 to 17 who had baseline resting-state fMRI, Revised Children’s Anxiety and Depression Scale (RCADS) scores at both baseline and one-year follow-up, and low-motion data at baseline. External replication on an independent longitudinal dataset is the immediate next step. The framing is promising rather than settled. The Växjö Interference Connectivity is derived from an individualised whole-brain model rather than from raw connectivity matrices, in the way that polygenic risk scores aggregate weak signals across many sites. The figure requires independent confirmation before it can be treated as a robust effect.
Doesn’t this contradict the Marek (2022) ceiling on brain-behaviour correlations?
The Marek ceiling describes correlations between brain-wide-association measures of raw functional connectivity and individual differences in behaviour. The Växjö Interference Connectivity is a different class of measure, derived from a fitted dynamical model rather than from connectivity matrices directly. The result is therefore not a strict violation of the Marek ceiling but sits in a different reference class with its own validation requirements. Whether the new reference class survives replication is the next empirical question.
Is this clinically applicable?
No. The canary result is a first promising signal in a single research cohort and is not ready for clinical use. Clinical translation would require external replication, prospective validation, regulatory review and a careful assessment of false positive and false negative rates in real-world populations. The current paper makes no clinical claim.
Why three separate papers rather than one?
Each paper addresses a distinct audience and a distinct kind of contribution. The unification paper is a mathematical contribution. The Entangled Loop is an architectural theory of consciousness. The canary is an empirical application to mental health. The three papers stand alone but reward being read together, and the coordinated upload reflects the structure of the underlying argument rather than a publication strategy.
What is the LSD test, and why does it matter?
The unification claim predicts that one mathematical object underlies three apparently independent descriptive frameworks. Under an LSD perturbation, applied through a 5-HT2A receptor density map, a single scalar coupling parameter shifts the spectrum of the operator. The new spectrum predicts both a multi-scale change in turbulence and a macroscale redistribution of harmonic energy. Both shifts are observed in the empirical data and they coincide at the predicted coupling value. The two predicted shifts live in mathematically independent domains, so their coincidence is evidence that the underlying operator is one rather than three.
Hasn’t the importance of emotion to consciousness been argued for decades already?
The importance of emotion to mind and to consciousness has been argued from many angles by Solms, Damasio, Panksepp and others, and the Entangled Loop is informed by this tradition. The specific claim of the Entangled Loop is new. Emotion is given an architectural role, derived from a thermodynamic constraint on far-from-equilibrium systems and scaffolded on a precise mathematical operator, rather than argued for on phenomenological or evolutionary grounds. The phrase “emotion is architecturally primary” is the framework’s own.
What is being done to address concerns about replication?
The full computational pipeline for the Växjö Interference Connectivity is available on a public repository from the date of upload, with documentation. External replication on at least one independent longitudinal dataset is planned within the next twelve months, and the candidate datasets have been identified. The updates page is maintained as a public log of replication attempts and their outcomes, whether positive or negative. We welcome and will support independent replication attempts by other groups.
Are the data and code available?
Yes. See the code and data page for repositories, datasets and replication procedures.
How does the trilogy link to Pope Leo XIV’s encyclical Magnifica Humanitas?
Pope Leo XIV’s first encyclical Magnifica Humanitas, signed on 15 May 2026 and released on 25 May, develops at length the question of what AI systems can and cannot be, and was presented at the Vatican alongside Christopher Olah of Anthropic. The encyclical’s claim is that artificial intelligence systems can imitate the functions of human intelligence but lack the constitutive features of personhood. They have no embodiment, no genuine experience, no felt valence and no moral, affective or relational capacities of the kind that make a person. A substantial section in Chapter 3 critiques the transhumanist and posthumanist projects that treat human finitude as a defect to be optimised away, arguing that finitude is constitutive of personhood rather than a limit to be transcended, and that human flourishing unfolds through these limitations rather than in spite of them.
The convergence with the Entangled Loop framework is anthropological rather than doctrinal. The framework’s central claim is that emotion is architecturally primary, that the architecture is embodied in specific thermodynamic and dynamical commitments, and that eudaimonic flourishing is the expression of an architecture that depends on, rather than transcends, its biological finitude. The encyclical’s claim that the human person is constituted by embodiment, affective and relational capacity, and limit, sits in striking parallel to the framework’s claim that consciousness cannot be abstracted from the substrate that produces it. Two accounts arrive by different routes at a similar anthropological position. The question of whether AI is conscious is not settled by computational sophistication alone, and what is at stake in the question matters for how the human person is understood.
The symmetry of the moment is worth noting. The trilogy’s cross-substrate evidence draws on the Sofroniew finding in Claude Sonnet 4.5, an LLM developed by the same company whose co-founder presented the encyclical at the Vatican alongside the Pope. The trilogy does not endorse the encyclical’s wider theological commitments, but the convergence on architectural and anthropological claims about consciousness is real and worth marking.
Where can I read more about whole-brain modelling?
The methodological background to the whole-brain modelling and canary paper is set out in the open-access book Whole-brain modelling: Cartography of the dynamics of mind by Deco and Kringelbach (2025), published by Oxford University Press. The book is freely downloadable in PDF form and provides a comprehensive introduction to the techniques used throughout the trilogy.
Who is the right person to contact for further detail?
See the contact page for the named lead on each kind of question. In summary, theoretical and framework, mathematical, LSD and canary questions go to Morten Kringelbach and Gustavo Deco, clinical questions to Diego Pizzagalli, and contemplative and meditative connections to Ruben Laukkonen and Shamil Chandaria.