Felt Cognition

A functional model of the stream of thought, built on the verified neuro chain. This is a frontier paper: almost every claim is model-level, the open register is large, and no link reaches causal evidence — the honest signature of the subject. The hard problem of experience is stated, and left open.

The thesis

A thought is many parallel ionic eddies igniting in γ (winner-take-most); the basal-ganglia loop neuro verified selects one; a dopamine error shapes which eddies are laid down next (learning what to think); and serial selection, bound within a θ frame, is the stream. What people call a “field of consciousness” is named here as band-structured phase-coherence on ionic spikes and synaptic currents — communication-through-coherence — not a physical field. The embodied real-time loop is proposed as felt cognition; whether it is experience is open.

This release emerges the substrate instead of assuming it. An in-package engine builds the brain’s organs from 4D-DNA, lets the cortex radiate a real electromagnetic brainwave at the speed of light, and stores memory as a physical attractor clocked by the emerged θ rhythm — then runs the functional model on top of that. The numbers in every chapter are now produced by shipped, deterministic code (seed = 19), closing the reproduction gap the old version had. The new physical layer is §3 organs, §4 the brainwave, and §5 memory; the consciousness claims remain hypotheses on top of that simulation, and the hard problem stays open.

Physical mediator (the bright line)

Every “eddy/field” names what carries it: the ion spike and the synaptic current along real connections, gated by the phase of classified low-frequency oscillations (δ/θ/γ). The oscillation does not carry the message; it sets when the message gets through. The earlier rejection of the electromagnetic “vortex field” cited a coherence length short by ~10⁹× — but that is the quantum decoherence length, and §4 emerges that the classical brainwave is in fact coherent across the brain (emitted at speed c — verified). So the coherence objection is corrected: what is validly retired is the quantum carrier, while whether the classical field is a medium is left open, decided by field strength, not coherence. The paper’s working mechanism is the ionic process and its θ/γ clock either way; existence and brain-scale coherence verified, the medium question open.

The arc

• Not a field — the retired EM field as a medium, and the named mechanism that replaces it (ions + band-phase gating); and the real emission that is kept.
• Brain organs from 4D-DNA — the cerebrum, cerebellum, hypothalamus and hippocampus emerged from measured master-gene γ; the developmental order falls out. verified
• The EM brainwave — E/I populations make θ and γ; the current radiates a real field whose front travels at c. verified
• How memory remembers — engram cells as bistable attractors; Hebbian write, cue completion, and θ-phase protection — the physics that drives the brainwave. verified
• Parallel eddies — many local ionic γ-assemblies, winner-take-most; γ sets ignitability. model
• Selection — one eddy chosen by neuro’s basal-ganglia loop; γ predicts the delay. model
• Learned field — dopamine RPE updates what gets thought next; γ read-only. model
• The stream — serial selection, conflict, memory feedback, θ–γ binding. model
• The felt loop — body + brain as one real-time loop; the hard problem open. model
• Hemispheres & AI — graded asymmetry; why computing thought is not feeling it. model
• Inter-organ coordination — eight organs coupled only by the measured ephaptic near-field settle into partial, metastable coordination; the in-silico cancel-vs-augment test shows the field causally contributes. verified
• Sensory inflow — eight sensory streams (nine modalities) couple to their anatomical relays at the same measured ephaptic strength; input loads onto the central field without seizing, and cross-modal binding rides only the shared field. verified
• Light to memory — a brainwave is emerged light; brainwave ⊕ sensory EM superpose and angle-rectify (α = 2/π) into a sign-surviving information bit that clears the engram fold and writes a persisting memory; a downstream neuron feels the rolled field. verified
• What a thought is — the closing definition: a thought is a union of faculties, not an abstract umbrella — serial selection among parallel ionic eddies on a substrate emerged from 4D-DNA. Closed on the mechanism axis; whether any of it is felt stays open. verified
• The faculties of mind — the atlas behind the definition: thought decomposed into fourteen faculties (ten built and bit-reproducible, four honestly owed), the partition census-checked against the standard neurocognitive domains so it cannot be gerrymandered. forced

Part II — Disorders of the mind

The same emerged cerebrum, now broken in specific, named ways. Autism is read not as one lesion but as three separable faults, and each downstream chapter follows one consequence: what a chemical can reach, what only a wearable theta carrier can reach, how that carrier must be operated and whether it is buildable, how ADHD separates, and what a synthetic population says. The closing chapters then extend the same axes into a transdiagnostic atlas — schizophrenia as the over-ignition mirror of autism, and epilepsy as the over-synchronisation pole — re-cutting conditions by shared mechanism rather than symptom list, and finally add the plasticity layer the static engine never had: a phase-correlation Hebbian update that makes consolidation representable, resolves the open dosing question, and supplies the reversible→chronified switch the temporal disorders need (depression as the first of them). A state-switching layer then reads the R19 cell over time — hysteresis, a latency that diverges at the fold (closing the epilepsy ictal time-course owed to it), and the barrier as switching threshold — and bipolar disorder uses it: two poles on one valence axis, episodes as bistable switches, kindling as the accumulating trace. Every result is an in-silico coupling state — efficacy = 0, not medical advice.

• Autism in three axes — autism modeled on the emerged cerebrum as three separable faults (threshold E-I, output gain, long-range wiring), each with a distinct, bit-reproducible (ΔPAC, ignition) fingerprint; which fault any individual's autism is, stays open. model
• What a chemical reaches — a threshold-lowering (gain) chemical fully reverses the E-I fault, partly the output fault, and only masks the wiring fault by over-synchronisation; the engine's account of why a stimulant relieves some autism and not others. model
• The θ-cap pacemaker — a wearable theta carrier is the only handle on the wiring axis no chemical reaches, but only as an external pacemaker (not a benign lane): it forces coordination in a narrow window, is cleanly removable, and is molecularly safe below the spinodal fold. model
• Operating principle and feasibility — the dynamics force one mode (minimum-effective, deficit-matched, continuous), paired with a physical-feasibility review: every component (theta-tACS, closed-loop phase-locking, multi-electrode montages) exists today, the specific assembly does not, and feasibility is not benefit. model
• ADHD, and why it differs — an explicit gene-grounded ADHD substrate (gain/arousal axes, intact wiring): a stimulant restores it where it only partly reaches autism, the cap is redundant for ADHD, and on an AuDHD substrate the two compose. ADHD model validity open. model
• The virtual trial — across eighty emerged cerebra, responders are gain-dominated and non-responders wiring-dominated and partly cap-rescuable; an honest refutation shows the residual is a dose-cap/stiffness limit, not a severe-wiring tail. model
• Schizophrenia, the mirror — the over-ignition pole of the same R19 axis autism-T sits on: a disinhibitory bias lowers the fold so irrelevant assemblies ignite (aberrant salience), and its positive, negative and cognitive symptoms map onto the threshold, output and wiring axes — a gain-reducing antipsychotic reverses the positive domain only, which is why dopamine blockade spares the rest. model
• Epilepsy, the over-sync pole — the framework's own over-synchronisation failure mode made the object of study: an excitatory bias drives the order parameter to the over-sync ceiling and the selective gate collapses (all assemblies ignite — the ictal state); autism-T < health < schizophrenia < epilepsy on one axis, and an anticonvulsant-class push reverses it. model
• The plasticity layer — the consolidation layer the static atlas never had: a phase-correlation Hebbian update of the ephaptic kernel (form forced, rate swept). It makes after-effects representable, resolves the open dosing question (spaced beats massed for retained structure — the cap repairs, pacing beats holding), and turns a reversible excursion into a chronified one (η=0 reverts, η>0 retains); η=0 reproduces the frozen engine bit-for-bit. The substrate the temporal disorders will stand on. model
• Depression & treatment resistance — the first temporal disorder, built on the plasticity layer (it imports the connectome rather than re-deriving it). Depression as the chronification of a low-coordination operating point: the engine's own HPA/stress axis drives a sustained withdrawal that lowers coordination below health (acute, reversible), and with plasticity the excursion writes a retained structural trace that does not revert — the reversible→chronified switch. Antidepressant delayed onset is a consolidation timescale, treatment resistance is the depth of the trace; no new constant, η=0/stress=0 reproduces the frozen engine bit-for-bit. model
• The state-switching layer — the fast counterpart to the plasticity layer: the R19 bistable cell read over time, with no new machinery. Sweeping the field gives hysteresis (a loop of width twice the fold — a state, once switched, resists switching back); the transition latency diverges at the fold (critical slowing), which closes the §25 ictal time-course owed to it; and the barrier sets the switching threshold (shallow wells flip, deep wells hold). Constant drive reproduces the engine bit-for-bit. The layer the mood-cycle disorders stand on. model
• Bipolar disorder — the second temporal disorder and the first to use the switching layer (it imports both E2 and the plasticity layer). Mania and depression are read as two operating poles on one valence axis (manic above health, depressive below); an episode is a bistable switch that persists past its trigger; kindling is the plasticity trace accumulating across episodes, so each switch gets easier; and a mood stabiliser is barrier-raising. η=0 reproduces the frozen engine bit-for-bit. model

Method and discipline

Every claim carries a status — verified / model / forced / open — and a failure log; retired claims are irreversible. The substrate chapters are now emerged by an in-package engine (repro/mind/_engine): the numbers reproduce from shipped, deterministic code (seed = 19, single-threaded, bit-for-bit), closing the C1 reproducibility gap the old release had, where only frozen results shipped without their generator. A model never certifies a model. Two extra rules hold here: a physical mediator is named for every eddy/field (no abstract synchrony), and the dependency is one-way — mind cites neuro, neuro does not depend on mind. The standing limits, the open register, and the retired register are in The open problem of experience; the scope is set in the Constitution.

Branch

This paper is a jamming branch of the vacuum-elasticity program and the companion of the neural paper: every mechanism here rests on neuro’s verified chain (ions, rhythm, the θ/γ code, memory, value, the basal-ganglia loop), and on the same R19 switch read across the Jamming Physics papers. neuro is the settled substrate; this paper is its open frontier — the dependency never runs the other way.

Contents

• §1Constitution and scopeWhat this paper reads (the functional stream of thought, present-tense), the physical-mediator discipline, the one-way dependency on neuro, and the hard problem held open.
• §2The retired field, and what replaces itThe retired EM field, why it fails physically, and the named mechanism that replaces it: ion spikes and synaptic currents, gated by classified low-frequency phase (CTC). The coherence-length rejection is corrected — it was the quantum case; the classical field is coherent across the brain, so its medium role is open.
• §3Brain organs from 4D-DNAThe substrate is emerged, not assumed: each organ’s master gene sets a measured bistable γ, and from γ alone the engine predicts the commit order — hypothalamus, cerebellum, cerebrum, hippocampus. Absolute size stays open.
• §4The EM brainwave, emergedExcitatory and inhibitory populations make a theta–gamma local field that is radiated through the wave equation; the emitted front travels at the wave speed c and the field is coherent across the brain — so the old coherence-length rejection (a quantum number) fails. Whether it is the computational medium turns on field strength and stays open.
• §5How memory remembersMemory is a physical attractor: a Hebbian write deepens an engram cell’s well until it completes from a tenth of a cue; the theta brainwave is the clock, and writing and retrieving on opposite phases measurably protects old memory.
• §6Parallel micro-eddiesA thought is many local ionic γ-assemblies laid down in parallel; the competition is winner-take-most, γ sets ignitability, and absolute size and consciousness are open.
• §7Selection: the closed loopAmong the parallel eddies, one is selected by the basal-ganglia action-selection loop of the neuro chain; the carried (most-ignited) eddy wins, γ predicts the delay.
• §8The laid-down field is learnedWhich eddies get laid down is itself learned: a selection outcome makes a dopamine reward-prediction error that updates which eddies appear next — the system learns what to think.
• §9The stream of thoughtThe stream is serial selection — one eddy then the next, with conflict, memory feedback, and a global γ synchrony binding the contents; the 'large field' is θ–γ phase-coherence.
• §10The embodied feeling loopBody and brain form one loop; the real-time feeling loop proposes consciousness is this integrated eddy process — but whether the loop is experience, or only correlates with it, is open.
• §11Hemispheres, and why thought is not feelingHemispheric asymmetry is graded, not the folk myth; and a system can run the stream’s computation yet lack the embodied real-time loop and experience — an architectural hypothesis.
• §12The open problem of experienceThe standing reference: the honest negative on the access marker, the open register (access, the hard problem, the quale), and the irreversible retired register.
• §13Inter-organ coordination by the ephaptic fieldEight organs — each a measured master-gene identity with a cited rhythm — coupled only by the measured ephaptic near-field (∼1/r³, at threshold; neuro §18–§19) settle into partial, metastable coordination. The in-silico cancel-vs-augment test shows the field causally contributes coordination, opens CTC windows, carries θ–γ coupling, and supports a traveling wave. Whether cognition uses it stays open.
• §14Sensory inflow — what the measured field bindsEight sensory streams covering the nine modalities (one skin organ carries touch, warmth and high-threshold pain), each with a master-gene γ read verbatim from the neuro chain, couple to their anatomical central relay at the same measured ephaptic strength (κ = 0.55, no new constant). Sensory drive loads onto the central field without seizing it (anchor R = frozen M9); distinct senses bind cross-modally only through the shared field (cancel < measured < augment). Whether cognition uses it stays open.
• §15Light to memory — how emerged light writes an engramOn the same vacuum lattice a brainwave is emerged light. Brainwave and sensory EM superpose, and geometric angle rectification (α = 2/π single, δ = 1/π² double, with 2π = α/δ) turns their signed, self-cancelling phase overlap into a sign-surviving information bit. A bound pair rectifies to ¼, an unbound pair to the floor δ; the contrast clears the R19 engram fold and writes a persisting memory under a θ write-clock. Six informations roll through γ slots in one θ frame, all recovered, and a downstream neuron feels the rolled field. Verified in-silico mechanism; biological use stays open.
• §16What a thought is — the definition that closes the paperThe closing chapter defines the paper’s central word. A thought is the union of fourteen mental faculties, not an abstract umbrella — mechanically, serial selection among parallel ionic γ-eddies bound within a θ frame, running on a substrate emerged from 4D-DNA (organs, brainwave and memory generated, not assumed). The everyday felt “stream” is recovered as a fast chain of discrete selections. The definition is closed on the mechanism axis; whether any of it is felt is a separate, orthogonal question left open.
• §17The faculties of mind — thought decomposed into fourteen facultiesThe atlas behind the closing definition. Thought is partitioned into fourteen faculties: ten (perception, attention, the serial stream, memory, learning, the affect mechanism, arousal/sleep, dreaming, large-scale binding, pathology) are built, frozen and bit-reproducible; four (language, volition, social cognition, metacognition) are honestly owed with their external inputs named. The partition is run against the standard neurocognitive domains so it cannot be a set gerrymandered to look complete, and the dream worked example shows why “thought” is the umbrella and “cognition” one room under it. Closed on the mechanism axis; the felt axis stays orthogonal.

Part II — Disorders of the mind

• §18Autism as a three-axis fault — threshold, gain, and wiringAutism read on the emerged cerebrum as three separable faults — a threshold/E-I fault (T, PAC down, ignition fold raised), an output/gain fault (O, PAC down, fold normal), and a long-range wiring fault (W, PAC unchanged, with a local-over/long-range-under imbalance). A 17-gene moderate-ASD cohort (γ via SantaLucia 1998) partitions T/O/W; the pair (ΔPAC, ignition) uniquely fingerprints each and reproduces under SEED = 19. Which fault any individual's autism is, is held open. efficacy = 0; not a diagnosis.
• §19What a chemical can and cannot reach — mask versus correctionOn the three-axis model, a scalar threshold-lowering (gain) operator — the mechanism by which the catecholaminergic stimulant class raises excitability — fully reverses the T fault (R back to health), partly helps O, and cannot correct W: brute gain leaves the locality imbalance exactly invariant and reaches health only by over-synchronisation (a mask). A selective tri-lever buys a safety margin, not new efficacy. The engine's account of why a stimulant relieves some autism and not others. efficacy = 0; no drug-treats-autism claim.
• §20The θ-cap — an external pacemaker for the wiring axisThe keystone disorder chapter. On the wiring-faulted cohort only a coherence-forcing external clock routes (inj ≈ 0.08–0.10; over-sync past 0.15); every passive additive coupling is inert (broadcast relay, far-pair relay, pure superposition), so the benign additive-lane reframing is REFUTED. With the cap on, far-pair coherence is restored (0.213 ≥ health 0.170) without over-sync; across OFF/ON cycling there is no rebound and no acquired dependence — the substrate has no plasticity, so it paces, not repairs. At the R19 switch (λ = 2g, Ω = 0.018) there are zero irreversible flips below the spinodal fold at any cycle count; the binding constraint is circuit over-sync, not molecular wear.
• §21The θ-cap operating principle — and whether it is physically buildableIf a θ-cap is to supply the wiring-axis function at all, the dynamics pin the operating mode on all sides: below the window it is inert, above it over-syncs, intermittent dosing reverts to the deficit, and nothing is banked — so the only mode is minimum-effective, deficit-matched amplitude, applied continuously (a wearable drug with no half-life past removal). Paired with a physical-feasibility review: every component (theta-tACS, closed-loop phase-locked EEG-tACS, multi-electrode phase-shifted montages that alter long-range connectivity, individualised MRI-optimised targeting, wearable home delivery) exists in research form today, but the specific assembly does not, there is no in-vivo readout of the wiring deficit, and the plasticity sign is phase-dependent. Feasibility of the apparatus is not evidence of benefit. efficacy = 0.
• §22ADHD — and why it separates from autismAn explicit gene-grounded ADHD substrate — not a relabelled autism cohort — emerged on the same engine: six output/gain genes (DRD4, SLC6A3, COMT, SNAP25, DBH, TH), two arousal/threshold genes (ADRA2A, SLC6A4), wiring = none, with axis-ambiguous genes pre-registered excluded. A stimulant restores ADHD to health in both synchrony and θ–γ coupling where the same stimulant only partly reaches autism; the cap's long-range benefit is 3.3× larger where the wiring is broken; and on an explicit AuDHD substrate the stimulant fixes the gain axis and the cap the wiring axis without interference. Why severe ADHD can look like autism, and why it is not. ADHD model validity open.
• §23The virtual trial — who the model says responds, and who does notA heterogeneous synthetic population of eighty emerged cerebra (seventy-six affected) with sampled fault mixes and stiffness dispersion, run under stimulant, cap, and both. Stimulant-responder fraction falls monotonically with wiring share (r = −0.60); adverse over-sync rises with cap amplitude and is minimised at the window; non-responders are wiring-dominated and the cap rescues about a third. An honest refutation, promoted to a finding: the cap-unrescued residual is not the severe-wiring tail but a dose-cap/stiffness limit, arguing for amplitude matched to the deficit. Every fraction is an in-silico coupling state. efficacy = 0.
• §24Schizophrenia — the over-ignition mirror of autismOn the one shared R19 ignitability axis (the fold = spinodal(g) = 0.3849), schizophrenia is the opposite pole from autism-T: a disinhibitory/excitatory bias LOWERS the fold (ignition threshold 0.245 vs health 0.395) so weak, in-health-sub-fold assemblies ignite — aberrant salience with nothing relevant lost. The pair (ignition-direction, aberrant-vs-lost) separates HEALTH / AUTISM-T / SCHIZOPHRENIA uniquely; a gain-reducing antipsychotic restores selectivity and worsens autism-T, the stimulant the reverse. Its positive, negative and cognitive symptoms sit on the threshold, output and wiring axes, and one gain-reducing operator reverses POSITIVE only — why dopamine blockade spares the negative and cognitive domains (an axis-structured, not dose-structured, response). Which pole a given psychosis is, is owed. efficacy = 0.
• §25Epilepsy — the over-synchronisation poleThe framework's own over-sync failure mode — the ceiling the θ-cap must stay below — made a primary module. An excitatory/disinhibitory E/I bias drives the global order parameter R monotonically to the over-sync ceiling (0.422 vs health 0.390); past a critical bias (+0.3) the selective single-winner gate collapses and every candidate assembly ignites — the ictal state, a loss of gating (Axis-A firewall; consciousness_claim = 0). On one synchrony axis: autism-T (under-ignited) < health (selective) < schizophrenia (aberrant, some irrelevant) < epilepsy (all ignited). An inhibitory / threshold-raising anticonvulsant-class push moves R back down and restores the gate (raises the seizure threshold). The static susceptibility is characterised; the ictal time-course is owed to a state-switching layer (E2). efficacy = 0.
• §26The plasticity layer — consolidation and the reversible→chronified switchThe plasticity / consolidation layer the structural atlas never had. The frozen engine has no variable that changes with use — which is why the θ-cap (§20–21) could only PACE, not repair, and why its plasticity sign was OPEN. This adds a slow phase-correlation Hebbian update to the ephaptic kernel, W_ij ← max(0, W_ij(1+η·C_ij)) with C_ij = <cos(θ_j−θ_i)>; the form is forced (standard phase-STDP, no free constant, row-stochastic so the ~1/r³ locality is kept), the rate η is open and every sign holds over an η sweep (anti-tuning), and the coupling-vs-bias map is the SAME one the schizophrenia/epilepsy modules use — no new tuned constant. Four results: driving the healthy point under plasticity then removing the drive leaves R at or above baseline (consolidation / after-effects); the same total cap dose delivered SPACED leaves a larger retained structural trace than MASSED (a spacing effect from pure phase-plasticity, so the cap REPAIRS and pacing beats holding — resolving the §20–21 open question); without plasticity a faulted excursion fully reverts when the bias is removed (the §20 'paces not repairs' result, now shown to be a consequence of the plasticity-free substrate), while with plasticity it leaves a retained trace — plasticity is the reversible→chronified switch (Axis-A firewall: a mechanism boundary, not a claim about the felt quality of chronic illness; consciousness_claim = 0); and η=0 reproduces the frozen M9 anchor bit-for-bit, a pure add-on. The layer, not a disorder — depression, bipolar and addiction import its substrate and are owed to later modules. efficacy = 0.
• §27Depression and treatment resistance — the chronification of a low-coordination operating pointThe first temporal disorder, built on top of the plasticity layer (it imports PlasticConnectome rather than re-deriving the rule). Major depression as the chronification of a low-coordination operating point. The handle is the engine's own HPA/stress axis — the M18 cortisol cascade and the M17 valence geometry, where cortisol sits on the withdrawal pole — mapped to a sustained withdrawal bias through the SAME coupling map the schizophrenia/epilepsy/E0 modules use (no new tuned constant; only the stress→withdrawal sign is asserted, magnitude open, signs hold over a sweep). Four results: a sustained withdrawal lowers the global order parameter below health (R 0.390→0.331 at severe withdrawal) — the acute, reactive depressed operating point, fully reversible on the static substrate; without plasticity (η=0) the excursion reverts exactly when the stressor lifts (reactive low mood), with plasticity (η>0) it leaves a retained structural trace that does not revert and deepens with exposure (‖ΔW‖ 0.075→0.151→0.227→0.338) — the reversible→chronified switch on the HPA handle (the robust signal is the structural trace; post-removal R is reported, not asserted below baseline); a coordination-restoring push moves the chronified structure only slowly — the restoring movement accumulates over epochs and is small after one (0.018 vs 0.138 at eight) with a therapeutic R direction — antidepressant delayed onset as a consolidation timescale, not a pharmacokinetic delay; and at a fixed restoring budget the fraction of the depressive trace neutralised decreases as the trace deepens — treatment resistance is the depth of the chronified trace. η=0/stress=0 reproduces the frozen M9 anchor bit-for-bit (Axis-A firewall: a retained trace is a mechanism boundary, not a claim about the felt quality of depression; consciousness_claim = 0). Bipolar (E2+E0) and addiction (T3a) import this substrate and are owed. efficacy = 0; not medical advice.
• §28The state-switching layer — hysteresis, the fold, and the ictal time-courseThe fast counterpart to the plasticity layer: the R19 bistable cell ṡ = g·s − s³ + h (the supercritical pitchfork the whole framework rests on) read OVER TIME, with no new equation and no free constant — g = 1.0 universal scale, the fold taken from the engine's own spinodal(g), all grids swept stimulus probes (anti-tuning). Four results: (E2.1) sweeping the tilting field up then down, the up/down transitions sit on opposite sides of zero, enclosing a hysteresis loop of width 2·spinodal ≈ 0.77 predicted as twice the fold — a state, once switched, resists switching back; (E2.2) the transition latency diverges as the drive approaches the fold (critical slowing) and falls monotonically as it overshoots (73.4→1.36) — a slow run-up then an abrupt jump — which CLOSES the ictal time-course the epilepsy chapter (§25) explicitly owed to this layer; (E2.3) the spinodal rises monotonically with the well depth, so the same handle that sets the barrier sets the switching threshold (shallow wells flip under a fixed drive, deep wells hold); (E2.4) constant-drive integration reproduces the engine's own settle bit-for-bit with the fold read from the engine, so the layer adds without altering (engine e61083ae…, tree 0fbf4988…, byte-unchanged). Axis-A firewall: a bistable transition is a mechanism boundary, not a claim about the felt quality of a mood state (consciousness_claim = 0). The layer, not a disorder — bipolar and the cycle disorders import it. efficacy = 0; not medical advice.
• §29Bipolar disorder — two poles on one axis, episodes as bistable switchesThe second temporal disorder and the first to use the state-switching layer: it IMPORTS both the plasticity layer (E0, the slow trace) and the switching layer (E2, the fast switch) and re-derives neither. Mania and depression are read as two operating poles on ONE valence axis — the engine's own M17 approach/avoidance geometry (dopamine on the approach pole, the M18 cortisol cascade on the withdrawal pole) — mapped through the SAME coupling map the schizophrenia/epilepsy/E0 modules use; no new constant. Five results: (B1) an approach bias raises the order parameter above health (R = 0.422 > 0.390), a withdrawal bias lowers it below (0.367 < 0.390), euthymia is the anchor between, ordering depressive < euthymic < manic monotone — not two diseases but two excursions on one axis; (B2) an episode is a bistable transition inheriting E2's hysteresis and its latency falling with overshoot, so it is entered at one fold and only left at the other (it persists past its trigger); (B3) kindling is the E0 trace accumulating across alternating episodes (‖ΔW‖ 0.06→0.33 over episode count), and a deeper trace lowers the barrier so each subsequent switch needs less drive — history makes the next episode easier; (B4) the mood-stabiliser sign is whatever raises the barrier — raising the well depth raises the flip threshold (0.38→0.64) so both manic and depressive switches become harder to enter; (B5) η = 0/no-bias reproduces the frozen M9 anchor R = 0.38961455156044245 bit-for-bit, W identical (engine e61083ae…, tree 0fbf4988…, byte-unchanged). Axis-A firewall: a bistable transition and a retained trace are mechanism boundaries, not a claim about the felt quality of a mood state (consciousness_claim = 0); the clinical heterogeneity is locked. Addiction (T3a) is owed. efficacy = 0; not medical advice.