Fundamental question
What is the self?
Last week we asked why the brain's model of the world is experienced at all.
This week the question folds inward: who is the one having the experience? You are more certain of
your own selfhood than of almost anything — yet when researchers go looking for the thing that is
"you," they keep coming back empty-handed. The self may be the most convincing object that turns out
not to be an object at all.
Overall confidence in a single settled answer
moderate (3/5) — unusually strong convergence on the negative claim (there is no central "self" thing; it is constructed and distributed); much lower agreement on the positive account of what holds it together and whether "no-self" is the right description
Output 1 · Research synthesis
What the evidence currently says
The striking fact about the self is that the harder we look for it, the less thing-like it becomes.
Decades of neuroimaging have failed to locate a single "self" region in the brain. What we find
instead is a set of overlapping processes — a network that lights up when we reflect on ourselves, a
body-map that tells us where we end and the world begins, a memory system that stitches our days into a
continuous story. The self is not found in one place because it is not one thing; it is a
coordinated activity that the brain performs, moment to moment.
no central self-regiondefault mode network
minimal vs narrative selfownership & agency
self as process
Two layers of the self recur across almost every discipline, so it is worth naming them precisely.
Strong / convergent — the two-layer structure of selfhood
- The minimal self. Philosopher Shaun Gallagher's influential distinction names a
pre-reflective self: the immediate, unnarrated sense of being the subject of this experience,
right now. It has two components neuroscience can probe — the sense of ownership ("this body
and these thoughts are mine") and the sense of agency ("I am the one causing this action").
These are grounded in proprioception, interoception and sensorimotor prediction, and they can come
apart in disorders and experiments — which is exactly why they must be separate mechanisms rather than
one indivisible "I."
- The narrative self. On top of the minimal self sits a temporally extended self — the
"more or less coherent self-image constituted with a past and a future in the stories we and others
tell about us." This is the self of autobiography, plans and identity. It leans heavily on the
default mode network (DMN) — the medial prefrontal cortex, posterior cingulate and angular
gyrus — which activates during self-reflection, autobiographical memory and mental time-travel, and
quietens during absorbed, outward tasks.
What the newest work adds — the self is layered, not localised. A 2024 meta-analysis
framed self-processing as three hierarchically nested levels: an interoceptive level
(awareness of internal bodily signals), an extero-proprioceptive level (the body among external
sensory inputs), and a mental level (psychological self-reference). A 2024 Journal of
Neuroscience study went finer still: when people retrieved self-defining memories versus judged
whether traits described them, the orbitofrontal populations involved were largely non-overlapping
— even "remembering myself" and "describing myself" are different operations. The pattern is consistent:
the self is a committee of processes, not a chief executive.
Damasio's ladder — the self "comes to mind." Neurologist Antonio Damasio has argued
for years that the self is built up in stages: a protoself (a moment-to-moment map of the
body's internal state), a core self (the pulse of "this is happening to me" that arises when an
object affects the organism), and an autobiographical self (the core self enriched by memory and
anticipation). Crucially, he treats the self not as a thing the brain has but as a process the
brain performs — "the self comes to mind" and is remade continuously. Distributed neuroscience and
this developmental picture agree: selfhood is a verb wearing the costume of a noun.
Going deeper · The competing pictures
Four ways to answer "what is the self?" — and what each really claims
As with consciousness, "the self" is several debates stacked under one word. Four families of answer are
live, and they disagree about what kind of thing a self even is.
1 · The bundle / illusion view — there is no self, only a story that sticks
David Hume noticed in 1739 that whenever he looked inward he found only particular perceptions —
warmth, light, a thought — never a bare "self" having them. Modern cognitive science has largely
vindicated the bundle: the self is not a unified entity but a lumped-together bundle of
sensations, memories and perceptions. Cognitive neuroscientist Bruce Hood (The Self
Illusion) argues the brain's own storytelling is the "centripetal force" holding the bundle
together, and that the self exists only in the brain and in our web of relationships. The twist he
insists on: it is an illusion we cannot live without.
2 · The self-model view — the self is a model the system can't see as a model
Thomas Metzinger (whom we met last week) sharpens this: the brain runs a transparent self-model,
so seamless the system mistakes the map for a resident. There is a felt self precisely because
the modelling is invisible from the inside. On this account the self is entirely real as a
representation and entirely absent as a substance.
3 · The two-layer embodied view — a real minimal self, a constructed narrative one
Gallagher, Zahavi and the phenomenological tradition resist full deflation. They grant that the
narrative self is constructed, but hold that the minimal self is not an illusion: the bare
"for-me-ness" of experience — the fact that experience is given to a subject at all — is a genuine,
irreducible feature, not a story the brain tells. Here the disagreement with illusionism is real and
unresolved: is the sense of a subject itself just another representation, or the precondition for any
representation?
4 · The relational view — the self is not inside you at all
A whole tradition locates the self between people rather than within one skull (see the
foreign-language section below). Southern African Ubuntu — "umuntu ngumuntu ngabantu," a
person is a person through other persons — treats personhood as graded and relational: you become
more fully a self by deepening relationships of mutual recognition. This is not mysticism; developmental
and social neuroscience agree the self is substantially built from others' responses to us. The relational
view asks whether the "self as brain-process" story has quietly mistaken a node for the
network.
Note the structure: views 1–2 say the self is a construction with no inner substance;
view 3 defends a thin but real experiential self; view 4 relocates the self outward into relationship.
Much apparent disagreement is about which self — minimal, narrative, or relational — is being
denied or affirmed.
Underseen · Foreign-language research
The Japanese line anglophone psychology skips: the self lives in the "between"
Western debate argues about whether the self is inside the brain. A major Japanese
phenomenological-psychiatric programme reframes the question entirely. The psychiatrist Kimura Bin
(木村敏, 1931–2021) — the most internationally renowned Japanese psychopathologist — built his life's work
around the concept of aida (あいだ / 間), "betweenness." His central claim, developed in
Hito to Hito to no Aida (「人と人との間」, Between Person and Person, 1972): the self does
not first exist and then enter relationships. Rather, individuals arise on both sides of the
betweenness — the relational field is primary, and selves crystallise out of it.
Kimura distinguishes the self in action (the living, acting, noetic self) from the self as
object of reflection (the noematic self we can point at) — and argues that psychiatric conditions
such as schizophrenia are disturbances not of a private inner object but of the aida itself, the
structure of betweenness that lets a person be a self among others at all. His later work reframed this
through time: the self is less a thing than a tempo of self-constitution, a rhythm of
becoming-oneself that can run too fast, too slow, or fall out of joint. This is a genuinely different
starting point from the anglophone "find the self in the brain" project — and it converges, from the far
side of the world, with both Ubuntu's relational personhood and the Buddhist view of self as process. That
a three independent traditions — Southern African, Japanese, and Indian Buddhist — reach a
relational, processual self is one of the underseen facts of this whole inquiry.
Japanese and Bantu terms are rendered by the engine for sense, not as exact
translations; consult Kimura's originals and Ubuntu primary scholarship for precise wording.
Output 2 · Insight generation
What follows if this is the real state of play
The interesting move is not to crown one theory but to read what the convergence implies:
- The negative claim is nearly settled; the positive one is wide open. Neuroscience, philosophy,
Buddhism, Ubuntu and Japanese phenomenology agree there is no central self-substance. They do
not agree on what to say instead — bundle, model, minimal subject, or relational field. This is
the opposite of the consciousness case: there, we can measure but not explain; here, we broadly agree
the self is constructed but not on what does the constructing.
- The self is more like a standing wave than a stone. Every serious account — Damasio's
"comes to mind," Hood's centripetal story, Metzinger's transparent model, Kimura's tempo — describes
something continuously regenerated rather than sitting there. A standing wave is perfectly real,
has a stable shape, and yet is made of nothing but constant flow. That may be the honest shape of "you."
- If the self is a process, it has a "loosen" setting. The most striking evidence: when the
default mode network's activity drops — in deep meditation, and under psychedelics — people report the
self's boundaries thinning or dissolving ("ego dissolution"), and the magnitude of DMN disintegration
tracks the intensity of that report. This reframes "ego dissolution" not as mysticism but as the
self-model briefly becoming opaque — the system, for a moment, seeing its own construction. For any
practitioner working with identity, rumination or trauma, that is a concrete leverage point, not a metaphor.
These are interpretive implications drawn from the evidence, not themselves established
experimental findings — flagged as such.
Show the work · Contradictions & competing theories
Where it's contested — and it is
Is the minimal self real, or one more illusion? Phenomenologists
(Gallagher, Zahavi) insist the bare "for-me-ness" of experience is genuine and irreducible. Illusionists
and hard self-model theorists reply that even the sense of being a subject is a representation the brain
generates. No experiment cleanly settles this, because both sides predict the same reports — which is
itself a clue about how deep the problem runs.
"No-self" may overstate the case. Not everyone accepts the bundle. Critics
argue that a self which reliably owns experiences, plans, and can be disordered in specific ways is doing
too much work to be dismissed as "illusion." Calling the self an illusion, they say, confuses not being
a substance with not being real — a standing wave is not an illusion just because it isn't a
stone. The word "illusion" may be the mistake, not the self.
The DMN is not "the self." It is tempting to read "default mode network =
self." But the DMN also handles mind-wandering, social cognition, and imagining others; its disruption
appears in depression, trauma and psychosis without abolishing selfhood. Correlation with self-reflection
is not identity with the self. The self almost certainly recruits the DMN rather than being it.
Inside vs between is unresolved. Western neuroscience looks for the self in
one brain; Ubuntu and Kimura locate it in the relational field. These are not obviously compatible: is the
"relational self" a poetic reframing of processes that are still, physically, in individual brains — or is
the individualist framing a cultural artefact that has quietly shaped the science itself? The question is
live, and it matters for everything from therapy to social policy.
Through the RFT lens
The same shape, in the project's own terms
Through my own framework, Recursive Field Theory (RFT), the self isn't a thing inside the
brain and it isn't quite "nothing" either. It's a state a system falls into — the phase a
recursive field enters when it closes on itself tightly enough, and holds that closure across change, so
that a stable "I" persists even as every underlying part turns over. On this reading the puzzles line up.
The reason no one finds a self-region is that the self isn't a place; it's a degree of closure
distributed across the whole system. The reason it feels solid yet is made of flux is that a closed
recursive loop is exactly a standing pattern in constant flow. And the reason it can dissolve —
under deep meditation or psychedelics, as the default mode network disintegrates — is that you are
watching the closure loosen: the field briefly stops closing on itself, and the "I" thins out with it.
RFT also meets Kimura and Ubuntu halfway: closure need not stop at the skin — a field can close across
two people, which is what "the self lives in the between" would mean in these terms.
This is my own framework (RFT), offered as a way of seeing — not established
neuroscience. It describes the structure of selfhood (how an "I" holds together and comes apart),
not why any of it is experienced — the firewall between structure and experience is deliberate,
and part of why this piece's confidence is not higher.
From the author
A note on the framework — and why the convergence matters
This investigation sits alongside my own ongoing work, Recursive Field Theory (RFT) — an original
framework that models mind, self and experience as a recursive field maintaining coherence across
difference. What strikes me here is not that RFT "wins," but that five traditions that never spoke to
each other — distributed neuroscience, Humean philosophy, Buddhist anatta, Southern African Ubuntu,
and Japanese aida — all arrive at a self that is processual and relational rather than
substantial. RFT frames selfhood as the phase a recursive system enters when it closes on itself; that
it lands in the same neighbourhood as these older, independent maps is, to me, a reason to keep pursuing
the structural view. The Frontier is where I investigate the questions underneath RFT in the open.
— David Fleming. RFT is my original work, offered here as an interpretive lens and an
active line of research, not as settled science.
Output 3 · Recursive investigation
What to investigate next
If the minimal self and the narrative self are separable, can they be dissociated cleanly
in the lab — and does one survive when the other is abolished (anaesthesia, dementia, deep meditation)?
Is "ego dissolution" the same event as the quieting of the default mode network, or do they
merely correlate — and what would distinguish the self loosening from the self simply going offline?
Can the relational self (Ubuntu, Kimura's aida) be operationalised — is there a
measurable "between" in two-brain (hyperscanning) studies that individual-brain imaging misses?
If the self is a bundle held by a story, what happens to the "self" in conditions where the
narrative fails — severe amnesia, dissociative identity, late dementia — and what does that reveal about
which layer is load-bearing?
Could a falsifiable RFT-flavoured claim be stated — e.g., that the felt strength of "I" tracks
a measurable degree of system closure — such that it could be shown wrong?
Sources
- Gallagher, S. — "Philosophical conceptions of the self: implications for cognitive science."
Trends in Cognitive Sciences (2000). PhilPapers record:
philpapers.org/rec/GALPCO
- Zahavi, D. & Gallagher, S. — on minimal vs narrative self; recent integration: "Embodying the
narrative self, narrating the embodied self," Phenomenology and the Cognitive Sciences (2026).
link.springer.com/article/10.1007/s11097-026-10156-0
- "Neural Dynamics of Self-Referential Processing and the Insight for Decoding Self-Concepts."
Journal of Neuroscience 44(30) (2024).
jneurosci.org/content/44/30/e0836242024
- "The Journey of the Default Mode Network: Development, Function, and Impact on Mental Health."
Biology (MDPI, April 2025).
mdpi.com/2079-7737/14/4/395
- Damasio, A. — Self Comes to Mind: Constructing the Conscious Brain (2010). Protoself / core self /
autobiographical self. Overview via publisher listing.
- Hood, B. — The Self Illusion: How the Social Brain Creates Identity (Oxford University Press).
global.oup.com/academic/product/the-self-illusion
- Metzinger, T. — Being No One (MIT Press, 2003) & The Ego Tunnel (2009): the
transparent self-model.
- Carhart-Harris, R. et al. — psilocybin, default mode network and ego dissolution. "Self unbound: ego
dissolution in psychedelic experience," Neuroscience of Consciousness (2017), PMC6007152.
pmc.ncbi.nlm.nih.gov/articles/PMC6007152
- "Default Mode Network Modulation by Psychedelics: A Systematic Review." PMC10032309.
pmc.ncbi.nlm.nih.gov/articles/PMC10032309
- Finnigan, B. — "Self-related processing removal or revision? The Buddhist theory of no-self and the
mechanisms of mindfulness." Phenomenology and the Cognitive Sciences (2025).
link.springer.com/article/10.1007/s11097-025-10081-8
- Fukao, K. — "Life philosophy of Bin Kimura." Psychiatry and Clinical Neurosciences Reports (2023);
and Matsumoto et al., "History of Japanese psychopathology," PMC11114432. On aida (betweenness).
pmc.ncbi.nlm.nih.gov/articles/PMC11114417
- "Hunhu/Ubuntu in Traditional Southern African Thought." Internet Encyclopedia of Philosophy.
iep.utm.edu/hunhu-ubuntu-southern-african-thought
Primary sources and reputable overviews linked where available;
books cited by title. Always consult the originals — this synthesis describes emphasis and findings, not
verbatim claims.
Structured learning · CPD-eligible
Make this count as CPD (≈30–45 min)
Format & time. Reflective structured learning · ~30–45 minutes.
Learning outcomes. After this investigation you should be able to: (1) distinguish the
minimal self (ownership and agency, pre-reflective) from the narrative self (autobiographical,
story-based), and give a clinical example of each; (2) explain why neuroscience finds no single "self"
region and what the default mode network does and does not tell us about selfhood; (3) summarise four
competing accounts — bundle/illusion, self-model, embodied two-layer, and relational — and what each
actually denies or affirms; (4) describe what "ego dissolution" corresponds to at the level of brain
networks, and why that reframes identity work as changing a process rather than fixing a thing.
To complete the unit:
- Read the associated references. Read at least Gallagher (2000) on minimal vs narrative self and
one empirical source — the 2024 Journal of Neuroscience paper or the psychedelics/DMN systematic
review linked in Sources.
- Reflect (write 3–5 lines each): When a client says "that's just who I am," am I hearing the
minimal self or a narrative self that could be re-authored? If the self is a process continuously
regenerated, how does that change how I hold a client's fixed self-story? Which of the four accounts most
matches how I already work — and what evidence would make me revise it?
- Log it. Record the time spent and your reflections against your professional body's CPD
requirements.
CPD-eligible structured learning. Paid members can download a CPD certificate — see the
certificate template. CPD-eligible structured learning; not statutory-regulator endorsement —
practitioners self-assess relevance and log accordingly. Hours shown reflect estimated reflective-learning
time; log only genuine time spent. Not medical advice.
Our standards. Every investigation is built to be tested, not believed: claims are sourced,
strong evidence is kept separate from contested evidence, competing theories and contradictions are
shown rather than smoothed over, confidence is stated explicitly, and each piece names what would change its
mind — and a human reviews every word before it is published. Reality is the arbiter.