The DOI matches its Crossref record; 235 of the 257 references it cites were found in a registry or an index; 49 of the 119 sources its claims cite were checked for support.
Consciousness in active inference: Deep self-models, other minds, and the challenge of psychedelic-induced ego-dissolution
Open accessAbstract
Predictive processing approaches to brain function are increasingly delivering promise for illuminating the computational underpinnings of a wide range of phenomenological states. It remains unclear, however, whether predictive processing is equipped to accommodate a theory of consciousness itself. Furthermore, objectors have argued that without specification of the core computational mechanisms of consciousness, predictive processing is unable to inform the attribution of consciousness to other non-human (biological and artificial) systems. In this paper, I argue that an account of consciousness in the predictive brain is within reach via recent accounts of phenomenal self-modelling in the active inference framework. The central claim here is that phenomenal consciousness is underpinned by ‘subjective valuation’—a deep inference about the precision or ‘predictability’ of the self-evidencing (‘fitness-promoting’) outcomes of action. Based on this account, I argue that this approach can critically inform the distribution of experience in other systems, paying particular attention to the complex sensory attenuation mechanisms associated with deep self-models. I then consider an objection to the account: several recent papers argue that theories of consciousness that invoke self-consciousness as constitutive or necessary for consciousness are undermined by states (or traits) of ‘selflessness’; in particular the ‘totally selfless’ states of ego-dissolution occasioned by psychedelics or meditative absorption. I argue that the present active inference account is uniquely positioned to address this worry, by demonstrating how phenomenal self-models can dissolve while phenomenal consciousness is preserved. This points towards a more general taxonomy of minds, artificial and biological, within the predictive processing paradigm.
Overview
CoChat Summary
## TL;DR
The paper argues that phenomenal consciousness in active inference is grounded in “subjective valuation”: a deep inference about how well an organism is self-evidencing across multiple timescales. It also argues that psychedelic ego-dissolution does not refute this view, because even selfless states retain affective evaluation and precision-based control.
## Key findings
- The central claim is that consciousness is underpinned by hierarchically deep self-models, specifically “subjective valuation” or precision estimation on the action model across multiple levels of the generative model.
- Active inference is presented as a formalization of allostasis in which organisms minimize expected free energy; this is linked to prior preferences, epistemic value, and adaptive policy selection.
- The paper argues that psychedelic-induced ego-dissolution can be explained by the REBUS model as a relaxation of high-level priors and a failure of sensory attenuation, producing unpredictable action and thought outcomes and a reduced sense of agency.
- It claims ego-dissolution does not undermine an active inference theory of consciousness because affective experience and subjective valuation remain present even when ordinary self-consciousness is diminished or absent.
## Methods
- Conceptual/theoretical paper rather than an empirical study; it develops an argument within the active inference and predictive processing framework.
- Uses prior literature and illustrative examples from perception, action, affect, meditation, depersonalization, schizophrenia, and psychedelic experience; no new sample or original data are reported.
- Main analysis is philosophical-computational: it maps consciousness, selfhood, and ego-dissolution onto hierarchical generative models, precision weighting, sensory attenuation, and expected free energy.
## Limitations
- The paper does not present new empirical data or a formal model test; this is evident from the text.
- It explicitly describes some linked accounts as “preliminary” or under development, especially the REBUS account as applied to psychedelics.
- The paper notes terminological ambiguity around “self-consciousness” and acknowledges that whether ego-dissolution counts as “totally selfless” depends on how self-consciousness is defined.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
37 Valid · 4 Not valid · 39 Not checked · 69 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“largely due to the fact that predictive processing is not exclusively concerned with conscious processing (But see: Hohwy 2012; Dolega and Dewhurst 2015; Friston 2018; Solms and Friston 2018; Wiese 2018; Williford et al. 2018; Clark 2019; Clark et al. 2019; Kirchhoff et al. 2019; Whyte 2019; Friston et al. 2020; Hohwy and Seth 2020; Ramstead et al. 2020b; Safron 2020; Solms 2019, 2021; Whyte and S”
§ Consciousness in active inference: Deep self-models, other minds, and the challenge of psychedelic-induced ego-dissolution > Introduction
What the sources say
- Supportedcites Hohwy 2012 source
In the “Biased competition” section, Hohwy says predictive-coding-related precision processing and attention may occur outside consciousness. The Concluding Remarks likewise say that attention does not necessarily produce conscious perception. These passages support the claim that predictive processing is not exclusively concerned with conscious processing. (LLM, conf=0.91)
- Not matchedcites Dolega and Dewhurst 2015
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Supportedcites Friston 2018 source
The Conclusion says that predictive-coding formulations do not necessarily imply conscious processing. It adds that the same principles used to explain visual perception can model pattern formation in single cells, supporting the claim that predictive processing is not exclusively about consciousness. (LLM, conf=0.94)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“The free energy principle (Friston et al. 2010; Badcock et al. 2019) thus casts control theoretic ‘essential variables’ in terms of high precision prior expectations.”
§ Consciousness in active inference: Deep self-models, other minds, and the challenge of psychedelic-induced ego-dissolution > The active inference framework
What the sources say
- Partly supportedcites Friston et al. 2010 sourceExperimental — accuracy not yet measured
The Introduction and §7.1 describe active inference as replacing control-theoretic optimal policies with prior expectations about state trajectories. Section 5.3 says greater relative precision makes those expectations more effective in generating action. The paper does not explicitly equate control-theoretic ‘essential variables’ with high-precision prior expectations; its stated comparison is with policies and state transitions. (LLM, conf=0.78)
- Partly supportedcites Badcock et al. 2019 sourceExperimental — accuracy not yet measured
In “The variational free-energy formulation” (page unavailable), the paper describes maintaining functional states within physiological bounds as avoiding surprising states, and says inherited adaptive priors guide organisms toward unsurprising states. This supports the proposed link between homeostatic regulation and prior expectations, but the passage does not identify control-theoretic ‘essential variables’ as high-precision priors; its discussion of precision concerns prediction-error signals. (LLM, conf=0.79)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Action initiation involves ‘systematic misrepresentation’ (Wiese 2017)—whereby proprioceptive evidence that, for instance, my arm is not moving, is attenuated to allow the system to bring about the desired movement (Adams et al. 2013; Brown et al. 2013).”
§ Consciousness in active inference: Deep self-models, other minds, and the challenge of psychedelic-induced ego-dissolution > Consciousness in active inference > Active inference and the phenomenal self
What the sources say
- Not checked — no full text of the sourcecites Wiese 2017 source
- Partly supportedcites Adams et al. 2013 sourceExperimental — accuracy not yet measured
In “Active inference, predictive coding and reflexes” (page unavailable), the paper says a mismatch between predicted and actual proprioceptive input generates an error that activates motor neurons and produces movement. “Active inference and sensory reafference” (page unavailable) discusses attenuation of sensory reafference during movement, but does not say that evidence of the body’s initial stillness is attenuated to initiate movement. The phrase “systematic misrepresentation” is attributed to Wiese, not to Adams et al. (LLM, conf=0.84)
- Supportedcites Brown et al. 2013 source
The Abstract says self-generated movement requires predictions to override sensory evidence that the body is not moving, and that attenuating the precision of sensory evidence resolves this conflict. The simulations section, “The permissive role of sensory attenuation in action,” shows that reducing attenuation suppresses movement. The phrase “systematic misrepresentation” is attributed to Wiese, not to Brown et al. (LLM, conf=0.94)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Every source the paper cites for this claim was checked and supports it.
In the paper
“by tracking regularities over time it can track its control of sensory outcomes more generally, where expectations of the downstream consequences of action inform policy selection (Friston 2018).”
§ Consciousness in active inference: Deep self-models, other minds, and the challenge of psychedelic-induced ego-dissolution > Consciousness in active inference > Active inference and the phenomenal self
What the sources say
- Supportedcites Friston 2018 source
In “Thick and deep models” (page unavailable), Friston says a system changes the flow of sensations through action and, with a temporally extended model, can infer the consequences of its actions. The same section says such a system selects actions by their expected outcomes; “Phenomenal transparency and the counterfactual me” describes choosing among possible futures to determine a course of action. This supports the paraphrase as a claim about organisms capable of modeling consequences over time. (LLM, conf=0.86)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Recall that, under active inference, lower levels of the predictive hierarchy track regularities that are unfolding on shorter timescales and higher levels track regularities unfolding on longer timescales (Kiebel et al. 2008; Friston et al. 2017b).”
§ Consciousness in active inference: Deep self-models, other minds, and the challenge of psychedelic-induced ego-dissolution > Consciousness in active inference > Active inference and the phenomenal self
What the sources say
- Partly supportedcites Kiebel et al. 2008 sourceExperimental — accuracy not yet measured
The Introduction and Conclusion state that lower levels represent fast, short-lived environmental trajectories while higher levels represent slowly changing context. The Methods describe this as perceptual inference, however, rather than establishing the claim specifically under active inference. (LLM, conf=0.84)
- Not matchedcites Friston et al. 2017b
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
88 more claims · 18 with their cited sources checked
Create a free account to see all 149 claims and to verify the 5 references still open.
- It may state two separate claims as one. 25
- Probably not a substantial claim. 41
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Self-check results appear here once an independent check has confirmed them.
Cited works that were retracted
Works this paper cites that were later retracted, from the retraction notices publishers deposit. Major when the notice came out before this paper was published.
A note, not an issue: for example, a cited work that was retracted only after this paper was published, which its authors could not have known.
In the paper
“Although a preliminary account, the REBUS model is growing in empirical support (Alamia et al. 2020; Dupuis 2020; Girn et al. 2020; Herzog et al. 2020).”
Cited work 10.1038/s41598-020-74060-6 · retracted 2022-09-15 · notice 10.1038/s41598-022-20093-y · this paper published 2021-01-01
Citations
Cites
1 retracted
RETRACTED ARTICLE: A mechanistic model of the neural entropy increase elicited by psychedelic drugs
Herzog, R. A., Mediano, P. A. M., Rosas, F. E., Carhart-Harris, R. L., Perl, Y. S., Tagliazucchi, E., & Cofré, R. (2020) · Scientific Reports · doi:10.1038/s41598-020-74060-6 · Cited by 50 · cited 1×
“Although a preliminary account, the REBUS model is growing in empirical support (Alamia et al. 2020; Dupuis 2020; Girn et al. 2020; Herzog et al. 2020).”
MethodRetractedRor2 signaling regulates Golgi structure and transport through IFT20 for tumor invasiveness
Nishita, M., Park, S.-Y., Nishio, T., Kamizaki, K., Wang, Z., Tamada, K., Takumi, T., Hashimoto, R., Otani, H., Pazour, G. J., et al. (2017) · Scientific Reports · doi:10.1038/s41598-016-0028-x · Cited by 15,947
The free-energy principle: a unified brain theory?
Friston, K. (2010) · Nature Reviews Neuroscience · doi:10.1038/nrn2787 · Cited by 8,120 · cited 2×
“Formally, precision is the inverse variance of a probability distribution (Feldman and Friston 2010), and optimization of precision-weighting is frequently equated to attention within predictive processing schemes (Feldman and Friston 2010; Clark 2013).”
BackgroundNagel, T. (1974) · The Philosophical Review · doi:10.2307/2183914 · Cited by 7,440
The feeling of what happens: body and emotion in the making of consciousness
(2000) · Choice Reviews Online · doi:10.5860/choice.37-6553 · Cited by 5,998
Whatever next? Predictive brains, situated agents, and the future of cognitive science
Clark, A. (2013) · Behavioral and Brain Sciences · doi:10.1017/s0140525x12000477 · Cited by 5,924 · cited 3×
“…by utilizing ‘priors’ (probability distributions about hidden states of the world) and incoming sensory data (‘prediction errors’) to arrive at a posterior estimate: a ‘best guess’ of the hidden causes of sensory signals (Clark 2013, 2015; Hohwy 2013; Aitchison and Lengyel 2017; Nave et al. 2020).”
BackgroundMethodHumans integrate visual and haptic information in a statistically optimal fashion
Ernst, M. O., & Banks, M. S. (2002) · Nature · doi:10.1038/415429a · Cited by 4,844
Rubber hands ‘feel’ touch that eyes see
Botvinick, M., & Cohen, J. (1998) · Nature · doi:10.1038/35784 · Cited by 4,410 · cited 2×
“For instance, in the ‘rubber hand illusion’ (Botvinick and Cohen 1998), visual information about the location of the hand is deemed to be precise due to the corroborating synchronous stroking pattern on the rubber hand, while the conflicting proprioceptive input suggestive of the hand’s real location is down-weighted…”
BackgroundThe Feeling of What Happens: Body and Emotion in the Making of Consciousness
Panksepp, J., & Modell, A. H. (2000) · Neuropsychoanalysis · doi:10.1080/15294145.2000.10773287 · Cited by 3,393
Origins and functions of positive and negative affect: A control-process view.
Carver, C. S., & Scheier, M. F. (1990) · Psychological Review · doi:10.1037/0033-295x.97.1.19 · Cited by 2,726
Research works printed but not linked: 2
- Girn, 2020, LSD flattens the functional hierarchy of the human brain, BioRxiv
- Hohwy, 2017, The Subject’s Matter
Cited by
Consciousness in Artificial Intelligence: Insights from the Science of Consciousness
Butlin, P., Long, R., Elmoznino, E., Bengio, Y., Birch, J., Constant, A., Deane, G., Fleming, S. M., Frith, C., Ji, X., et al. (2023) · arXiv · doi:10.48550/arxiv.2308.08708 · Cited by 196
Self-citationExpecting some action: predictive processing and the construction of conscious experience
Nave, K., Deane, G., Miller, M., & Clark, A. (2022) · Review of Philosophy and Psychology · doi:10.1007/s13164-022-00644-y · Cited by 50
Self-citationMetzinger, T. K. (2024) · The MIT Press eBooks · doi:10.7551/mitpress/15196.001.0001 · Cited by 100
Buchborn, T., Kettner, H., Kärtner, L., & Meinhardt, M. W. (2023) · Frontiers in Neuroscience · doi:10.3389/fnins.2023.1232459 · Cited by 34
“Efforts to understand the neuronal substrates of ego dissolution (Carhart-Harris et al., 2016; Tagliazucchi et al., 2016; Stoliker et al., 2022), along with philosophical and phenomenological analyses (Letheby and Gerrans, 2017; Milliere, 2017; Deane, 2021), have been ongoing.”
BackgroundParvizi-Wayne, D., Sandved-Smith, L., Pitliya, R. J., Limanowski, J., Tufft, M. R. A., & Friston, K. J. (2024) · Frontiers in Psychology · doi:10.3389/fpsyg.2024.1354719 · Cited by 43
“…there would be no phenomenal experience (Frankfurt, 1988, p. 162; Goldman, 1970, p. 96; Sartre, 1956; Zahavi, 1999, 2005; Legrand, 2007a; Kriegel, 2009; Kiverstein, 2020; although see Letheby, 2020, Millière, 2020, Winter et al., 2020, Deane, 2021, Laukkonen and Slagter, 2021, Metzinger, 2024).”
BackgroundResultInfluential
Metadata
- Authors
- Deane, G.
- Affiliations
- University of Edinburgh (GB)
- Corresponding author
- Deane, George
- ORCID
- Deane, George
- Venue
- Neuroscience of Consciousness
- Publisher
- Oxford University Press (OUP)
- Published
- 2021-01-01
- Volume
- 2021
- Issue
- 2
- Pages
- niab024-niab024
- ISSN
- 2057-2107 (ISSN-L)
- PMC
- PMC8408766
- OpenAlex
- W3196481983
- Cited by
- 38
- References
- 257
- Citations by year
- 2026: 7 · 2025: 8 · 2024: 12 · 2023: 4 · 2022: 7
- FWCI
- 0.85
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- Neuroscience of Consciousness (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Edinburgh Research Explorer (University of Edinburgh) (submittedVersion, cc-by)
- Funders
- ERC XSPECT project (DLV-692739)
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › Embodied and Extended Cognition
- Social Sciences › Arts and Humanities › Philosophy › Philosophy and Theoretical Science
- Social Sciences › Arts and Humanities › Philosophy › Free Will and Agency
- Keywords
- active inference, predictive processing, phenomenal consciousness, self-model, subjective valuation, sensory attenuation, other minds, altruism, self-consciousness, computational psychiatry
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Consciousness in active inference: Deep self-models, other minds, and the challenge of psychedelic-induced ego-dissolution. https://app.cochat.ai/papers/10.1093/nc/niab024
- The paper
- Deane, G. (2021). Consciousness in active inference: Deep self-models, other minds, and the challenge of psychedelic-induced ego-dissolution. Neuroscience of Consciousness. https://doi.org/10.1093/nc/niab024
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.