Merge pull request 'research: mechanical depth gate for tranches (#409 )' (#410 ) from research-depth-gate into m10-derive-refresh

research: mechanical depth gate for tranches (issue #409 )
The overnight R1 run (epic #375) closed 16 of 31 domains with reduced-depth batched masters; the quality firewall was self-attested and bent under controller context pressure. Post-hoc review caught it; this change makes depth mechanically checkable so the failure class cannot recur silently. - scripts/research_gate_check.py: stage-file presence + substance floors + adversarial provenance + R0 reconciliation + source traceability; R1_FULL / R1_LITE grades, LITE requires registry pre-declaration (silent scope reduction = gate drift). - research_standard.md: five binding rules — mechanical gate, breadth-before-depth as the only legal trade, declared grades, adversarial independence, depth checkpointing between domains. - domain_registry.md: stale statuses corrected, P–AE marked R1_LITE retroactively, 39 grandfathered findings across 8 pre-gate domains recorded (C-cognitive-architectures passes R1_FULL clean as the calibration domain). Version intent: patch (tooling + research governance, no runtime behavior change). Fixes #409.
2026-09-20 11:36:06 +01:00 · 2026-09-20 11:31:08 +01:00 · 2026-09-20 11:20:41 +01:00 · 2026-09-20 06:28:19 +01:00 · 2026-09-20 06:26:54 +01:00
284 changed files with 22694 additions and 8 deletions
@@ -25,22 +25,22 @@
    "memory"
   ],
   "live_defect": false,
-   "last_revalidated": "2026-09-19",
-   "next_action": "none \u2014 defect no longer reproduces: zero MessageAbortedError in current logs (incl. active worker), six most recent session checkpoints full documents (111-396 lines); verified by present-absence 2026-09-19"
+   "last_revalidated": "2026-09-20",
+   "next_action": "none \u2014 defect FIXED_VERIFIED (present-absence 2026-09-19); stale bug label removed 2026-09-20 after #407 fix (PR #408 merged) stopped the workflow re-adding it"
  },
  {
   "number": 316,
   "title": "ADAPT-1 B3-v2 (RESEARCH-LIKE): Qdrant associative-memory projection spike (executes #162)",
-   "tracker_state": "open",
-   "reconciliation_state": "VALID_OPEN",
+   "tracker_state": "closed",
+   "reconciliation_state": "FIXED_VERIFIED",
   "labels": [
    "governance",
    "memory",
    "perception"
   ],
   "live_defect": false,
-   "last_revalidated": "2026-09-12",
-   "next_action": "B3-v2 Qdrant associative-memory projection spike \u2014 not started; ADAPT-1 sub-item under epic #284"
+   "last_revalidated": "2026-09-20",
+   "next_action": ""
  },
  {
   "number": 315,
@@ -1147,8 +1147,20 @@
   "live_defect": false,
   "last_revalidated": "2026-09-11",
   "next_action": ""
+  },
+  {
+   "number": 407,
+   "title": "Auto-labeler re-adds bug label to closed fixed issues, defeating reconciliation",
+   "tracker_state": "closed",
+   "reconciliation_state": "FIXED_VERIFIED",
+   "labels": [
+    "infrastructure"
+   ],
+   "live_defect": false,
+   "last_revalidated": "2026-09-20",
+   "next_action": "fixed by PR #408 (closed-issue guard + keyword fallback filing-time-only); verified live: bug label removed from #324 and no re-add after 75s across label events"
  }
 ],
- "last_reconciled": "2026-09-19",
- "reconciled_by": "fresh-session tranche 2026-09-19: template retrofit (spine+plain-English on all open issues), closures #158/#162 (SUPERSEDED) and #324 (FIXED_VERIFIED), research labels applied to #163/#164/#165/#166/#182/#316, stale bug labels removed from #163-#166; 2026-09-19 absorption pass: parked children absorbed into parents (#175<-#176..#179; #181<-#182,#183,#164..#166) per one-open-note-per-idea standard"
+ "last_reconciled": "2026-09-20",
+ "reconciled_by": "2026-09-20 overnight Atlas close-out: R1 complete (#376-#406 all R1_QUALIFIED), epic #375 closed; artifact updated for #316 (FIXED_VERIFIED spike concluded) and #324 (FIXED_VERIFIED, label blocker owned by #407); prior: fresh-session tranche 2026-09-19 (template retrofit, closures #158/#162 SUPERSEDED + #324, absorption pass)"
 }
@@ -0,0 +1,47 @@
+# 00 — ISSUE SCOPE (#376 R1)
+
+## Core question
+What does Complementary Learning Systems (CLS) actually establish about fast episodic acquisition, slower generalization, consolidation, replay, schemas, interference, reconsolidation, and retention — and which of those mechanisms should or should not inform Rune's Memory architecture?
+
+## Subquestions
+1. What did the original CLS theory actually claim?
+2. Which claims were empirical vs computational/theoretical?
+3. What has changed since the original theory?
+4. Which mechanisms survived later evidence?
+5. Which were weakened or substantially revised?
+6. How does schema learning change "fast vs slow"?
+7. What is known about replay selection?
+8. What is known about destructive interference?
+9. What is known about memory transformation during consolidation?
+10. What is known about reconsolidation?
+11. Does consolidation necessarily imply transformation?
+12. When is rapid integration possible?
+13. What evidence exists against a simple dual-store interpretation?
+14. What competing theories explain the same observations?
+15. What parts are biological constraints rather than general learning principles?
+16. Which properties transfer to software agents?
+17. Which do NOT transfer?
+18. Does Rune already implement analogous behavior?
+19. What would materially alter Rune's target architecture?
+20. What remains experimental?
+
+## Hypotheses under test (from R0, NOT accepted)
+H1: fast exact episodic storage + slow generalized knowledge is the right Rune split.
+H2: two stores / two writers is the software form.
+H3: schema-gated fast consolidation is justified.
+H4: consolidation should be utility-prioritized, not recency-based.
+H5: covering episodes must be retained after generalization.
+H6: reconsolidation has a software analogue in supersession.
+H7: independent-evidence corroboration should gate promotion.
+H8: destructive consolidation must be rejected (append-only).
+H9: evidence→episode→observation→belief→procedure is the right progression.
+
+## Falsifiers
+- F1: If deeper evidence shows no useful fast/slow separation outside biological constraints, the CLS architectural analogy is weakened or rejected.
+- F2: If schema-rapid-integration findings fail replication or are explained by simpler mechanisms, H3 falls.
+- F3: If replay prioritization evidence is task-specific (rodent navigation) with no general principle, H4 weakens to speculation.
+- F4: If reconsolidation's boundary conditions prove so biological that no software analogue preserves the useful property, H6 falls.
+- F5: If retained-episodes practice has no measured benefit in AI systems (consolidation without retention works as well), H5 is a preference, not a requirement.
+
+## Stopping criterion
+Evidence saturation: new sources repeat known mechanisms/criticisms, add only detail.
@@ -0,0 +1,32 @@
+# 01 — INWARD: Current Rune reality (evidence-first)
+
+_Evidence hierarchy: observed runtime > persisted canonical state > independent tests > source existence > design docs. Sources: T0 gates, T0.1 reconciliation, R0 artifact, direct store inspection (this tranche, read-only)._
+
+## What Rune already does (PROVEN, T0/T0.1)
+- **Fast exact capture:** every entry append-only with created_at; dedupe/supersede gate with receipts (T0 PROVEN_PRESENT). 924 canonical entries; supersession preserves prior rows (56 superseded retained).
+- **Slow derived layer:** knowledge.json consolidation (PROVEN to run; single inspectable writer `memory_consolidation.py`); source-attributed (memory/abstraction/neuron/graph provenance counts in T0 map).
+- **Trust lifecycle:** provisional/validated/locked + HMAC audit ledger (PROVEN — but T0.1: trust does NOT gate retrieval today).
+- **Dual-write discipline:** episodic writes synchronous via `memory_store.py add_entry`; consolidation runs offline. Two writers exist structurally (agent path vs consolidation path).
+- **Failure learning:** wash_count/use history, promotion_eligible flag (PROVEN fields; promotion gate semantics = design, not exercised at scale).
+
+## What is DESIGNED but not proven
+- `derived_from` links: ABSENT (T0.1/M: "what depended on X" unanswerable).
+- Utility-prioritized or schema-gated consolidation: ABSENT (consolidation = full sweep; trigger non-deterministic — T0.1 R5: no cron/timer exists, agent-invoked).
+- Reconsolidation loop: ABSENT (locked is terminal; retrieved beliefs are not re-litigated).
+- Case layer (episode→case→procedure): ABSENT.
+- Independent-evidence corroboration test: ABSENT (promotion_eligible driven by wash_count = frequency, not independence).
+
+## What is BROKEN/DORMANT
+- Sleep-time transcript mining: BROKEN (stale DB path, T0).
+- Warm tier: vestigial (all 924 rows tier='warm'; T0.1 R4).
+- Graph projection: dormant (last edge 2026-08-26).
+- 439/838 active entries unindexed (embedder silent failure — T0.1 R3): the fast layer itself is LEAKY.
+
+## R0 CLS claims (to reconcile, not accept)
+R0 (families/complementary_learning_systems.md) claimed CLS supports: H1-H9 above, classifications COVERED (two-store split), ADAPT (utility qualification, shadow-mode, schema fast-path, reconsolidation loop, utility-prioritized consolidation), REJECT (recency-based priority, destructive consolidation). T1-impact flag: YES (three missing mechanisms).
+
+## Where R0 made implementation leaps
+1. R0 said "utility (retrieved-and-used count)" as qualification signal — but Rune's wash_count IS a retrieved-use count and T0 evidence shows it was never validated as a quality signal (it's frequency, gameable, and the T0.1 store shows superseded entries with high wash counts).
+2. R0's "schema-gated fast path" assumed schema detection is tractable — no Rune mechanism for schema-matching exists (abstraction.db has 59 facts; no matching machinery).
+3. R0's reconsolidation analogue assumed retrieval-triggered lability maps to supersession — but Rune supersession is write-triggered, not retrieval-triggered; the mapping was asserted, not derived.
+4. R0 did not distinguish which CLS claims are empirical vs computational assumptions (e.g. d≈0.5 decay is a fit parameter, not a law).
@@ -0,0 +1,6 @@
+# 02 — OUTWARD (populated by the deep-research actor; verified by controller)
+
+Files: historical_foundations.md, modern_cls.md, consolidation_and_replay.md,
+schema_learning.md, reconsolidation.md, interference_and_forgetting.md,
+competing_models.md, criticism_and_limitations.md, software_and_ai_analogues.md
+ sources.md (ledger) + claims.json (machine-readable).
@@ -0,0 +1,8 @@
+# Competing theories (researched from critics, not just CLS advocates)
+
+- **MTT vs standard consolidation** (Nadel & Moscovitch 1997/2000 vs Squire & Alvarez 1995): retrograde amnesia gradients exist in BOTH directions depending on lesion/method; meta-analyses of patient data show inconsistent gradient temporality. Status: genuinely UNRESOLVED.
+- **Hippocampal indexing vs store-and-transfer:** Teyler & Rudy 2007 (DOI 10.1002/hipo.20350); Ryan/Tonegawa 2015 engram work (DOI 10.1126/science.aaa5542) favors trace-persistence + retrieval-dependence over physical transfer. 2025 reassessment preprint exists (Authorea).
+- **Consolidation skeptics / old-memory hippocampal engagement:** Nadel, Hupbach, Gomez & Newman-Smith 2012 (Neurosci Biobehav Rev) — hippocampus remains active for remote memories; Bontempi et al. 1999 (J Neurosci); Frankland et al. 2004, Science 304. DOI 10.1126/science.1094804 — remote memories shift cortical systems (ACC) but hippocampal damage still disrupts DETAILED remote recall; Squire camp counters deficits reflect detail/quality not necessity (Squire, Genzel, Wixted & Morris 2015, CSH Perspect Biol). **This is the field's fault line on transformation-vs-retention.**
+- **Trace Transformation Theory:** R0's "Winsneski/Bhatt" DOES NOT EXIST. Verified anchors: Moscovitch & Gilboa 2021 preprint (OSF, DOI 10.31234/osf.io/yxbrs) + Sekeres/Winocur/Moscovitch 2018.
+- **Bayesian/latent-cause:** Gershman, Blei & Niv 2010, Psych Review 117:197 — reconsolidation/extinction phenomena emerge from latent-cause inference; reconsolidation = belief-updating special case. This is the strongest competing FRAMEWORK for Rune: it explains the same data without a biological consolidation story.
+- **CLS v2 (McClelland et al. 2020)** absorbs schema effects and responds to MTT — the live theoretical front.
@@ -0,0 +1,13 @@
+# Consolidation & replay
+
+**Foster & Wilson 2006, Nature 440:680. DOI 10.1038/nature04587.** Reverse replay of behavioral sequences in awake hippocampal place cells.
+
+**Carr, Jadhav & Frank 2011, Nat Neurosci 14:147. DOI 10.1038/nn.2732.** Awake replay serves BOTH retrieval and consolidation — unresolved dual role.
+
+**Ambrose, Pfeiffer & Foster 2016, Neuron 91:1342. DOI 10.1016/j.neuron.2016.07.047.** Awake reverse replay gated by changing reward — precursor evidence for prioritization.
+
+**Mattar & Daw 2018, Nat Neurosci 21:1609. DOI 10.1038/s41593-018-0232-z.** Computational model: replay prioritized by EXPECTED FUTURE UTILITY (gain = value of replaying; need = state relevance), NOT recency; replays benefit FUTURE learning. Makes replay prospective/selective — biological analog of prioritized experience replay, and a direct constraint on CLS's replay-as-teacher loop.
+
+**Tanrıverdi et al. 2023, J Cogn Neurosci. DOI 10.1162/jocn_a_02021.** In humans, MORE awake hippocampal-cortical co-reactivation predicted MORE forgetting of details — co-reactivation may reorganize/gist-ify rather than preserve.
+
+**Method validity caveat:** "Evaluating hippocampal replay without a ground truth" (eLife 2023, DOI 10.7554/eLife.85635.sa0) + 2024 dopamine-signaling exchange (eLife): replay DECODING methods still contested — treat specific replay-content claims as method-dependent.
@@ -0,0 +1,10 @@
+# Negative evidence / limitations
+
+1. **Fenn et al. 2009** (DOI 10.1101/lm.1500808): sleep REDUCED false memories — direct counter-result to Payne 2009. The gist-consolidation direction is not settled.
+2. **Tanrıverdi et al. 2023** (DOI 10.1162/jocn_a_02021): replay-like co-reactivation associated with FORGETTING — consolidation-as-helping is not monotone.
+3. **RIF reliability** (European Psychologist 2011, DOI 10.1027/1016-9040/a000040; Frontiers 2014, DOI 10.3389/fpsyg.2014.01343; Camp/Pecher/Schmidt 2007 cue-independent tests): effect size unstable across labs/cues; inhibition interpretation often fails.
+4. **Reconsolidation human failures:** prediction-error boundary (Sevenster/Kindt) means many "reactivation" protocols fail to destabilize — numerous human nulls (Exton-McGuinness & Reichelt review).
+5. **Sharon/Moscovitch/Gilboa 2011** (PNAS, DOI 10.1073/pnas.1005238108): rapid acquisition of arbitrary associations WITHOUT hippocampus — counterevidence to strict hippocampal necessity for fast learning.
+6. **Ryan/Tonegawa engram line** (DOI 10.1126/science.aaa5542): retrograde amnesia without trace loss — behavioral "forgetting after consolidation disruption" can be RETRIEVAL failure, not trace destruction; weakens lesion-based consolidation arguments in BOTH directions.
+7. **Hoffman et al. 2024, Memory & Cognition. DOI 10.3758/s13421-024-01576-x:** "Awake targeted memory reactivation doesn't work" — a named negative result.
+8. **Two R0-brief references do not exist** (McClelland/Whyte 1994; Winsneski/Bhatt) — flagged and excluded (see sources.md).
@@ -0,0 +1,7 @@
+# Historical foundations (API-verified citations)
+
+**McClelland, McNaughton & O'Reilly 1995, "Why there are complementary learning systems in the hippocampus and neocortex," Psych Review 102(3):419-457. DOI 10.1037/0033-295X.102.3.419.**
+Claims verified: (1) catastrophic interference in backprop-style networks — new arbitrary associations sharing representations destroy old ones; (2) forces slow, interleaved, small-step statistical learning in neocortex; (3) structurally separate hippocampus with sparse representations provides fast episodic learning and later acts as TEACHER via replay; (4) episodes become neocortical gist. Computational assumptions: hippocampal sparseness (pattern separation), neocortical overlap (generalization), replay-driven consolidation.
+NOTE: the "McClelland/McNaughton/Whyte 1994 bitter-tracing paper" from the R0 brief DOES NOT EXIST (Crossref/OpenAlex/arXiv searched) — likely conflation of the 1995 review with animal rapid-learning studies. Flagged; excluded from ledger.
+
+**Squire 1992 (Archives of Neurology); Squire & Alvarez 1995 (Curr Opin Neurobiol).** Declarative-memory background; hippocampal involvement TEMPORARY; systems consolidation = neocortex gradually sufficient. This is the position CLS inherits — and the main target of MTT critics.
@@ -0,0 +1,9 @@
+# Interference / forgetting / generalization hazards
+
+**McCloskey & Cohen 1989 (Psychol Learn Motiv 24:109); Ratcliff 1990 (JEP:General 119:195).** Catastrophic interference — the motivating phenomenon for CLS; robust in simulation.
+
+**O'Reilly & Norman 2002, TiCS 6. DOI 10.1016/s1364-6613(02)02005-3; Norman & O'Reilly 2003, Psych Review 109. DOI 10.1037/0033-295x.110.4.611.** REC model: hippocampal pattern SEPARATION (prevents interference/false memories) vs pattern COMPLETION (retrieval); computational trade-off explains age-related and amnesic distortions. Robust computationally; hippocampal-subfield fMRI support (Lacy/Yassa/Stark line, e.g. 2011 Learn Mem, DOI 10.1101/lm.1971111).
+
+**False memories via gist:** Payne et al. 2009, Neurobiol Learn Mem. DOI 10.1016/j.nlm.2009.03.007 — sleep INCREASED DRM false memory (gist consolidation). **BUT Fenn et al. 2009, Learn Mem. DOI 10.1101/lm.1500808 found the OPPOSITE** (sleep reduced false memories). Flag: the gist-consolidation claim is directly contested at the effect level. Ledger records the PAIR.
+
+**Retrieval-induced forgetting:** Anderson, Bjork & Bjork 1994, JEP:LMC 20:1063. DOI 10.1037/0278-7393.20.5.1063. Meta-analytically real on average — Murayama et al. 2014, Psych Bull. DOI 10.1037/a0037505 — but strong cue-dependence; the INHIBITION interpretation contested (interference accounts; Raaijmakers-school critiques); reliability concerns documented (European Psychologist 2011, DOI 10.1027/1016-9040/a000040; Frontiers in Psychology 2014, DOI 10.3389/fpsyg.2014.01343). Effect real; mechanism and cross-lab reliability of the inhibitory version unsettled.
@@ -0,0 +1,13 @@
+# Modern CLS (CLS v1 → v2)
+
+**Nadel & Moscovitch 1997 (Hippocampus 7:217); Nadel et al. 2000 (Psych Review 107:66).** MTT: each retrieval creates a new hippocampal trace; episodic memories stay hippocampus-dependent FOREVER; only semanticized gist becomes neocortex-dependent. Direct challenge to standard consolidation and CLS's teacher-transfer picture.
+
+**Teyler & DiScenna 1986; Teyler & Rudy 2007 (Hippocampus, DOI 10.1002/hipo.20350).** Hippocampal indexing theory: hippocampus stores an INDEX only; cortical representations reinstated at retrieval — consolidation is index elaboration, not trace transfer.
+
+**Ryan, Roy, Pignatelli, Arons & Tonegawa 2015, Science 348:1007. DOI 10.1126/science.aaa5542.** Optogenetic reactivation of hippocampal engram cells rescued memory expression after anisomycin-induced amnesia; silent engrams later matured (Roy et al. 2017 PNAS, DOI 10.1073/pnas.1714248114). Bearing: dissociates trace STORAGE from trace RETRIEVAL — amnesia is often retrieval failure, undermining simple "consolidation = strengthening" readings used by BOTH CLS and its critics.
+
+**Winocur & Moscovitch 2011 (JINS 17:766); Sekeres et al. 2018 (Hippocampus).** Transformation hypothesis: remote memories are TRANSFORMED (episodic→semantic; vivid→schematic), not copied.
+
+**Kumaran, Hassabis & McClelland 2016, TiCS 20(7). DOI 10.1016/j.tics.2016.05.004.** CLS updated: rapid neocortical integration via schema-consistency, episodic control, deep-learning interpretation; explicitly reconciles Tse findings with CLS. Robust as framework paper; each added claim inherits its own evidence base.
+
+**McClelland, McNaughton & Lampinen 2020, Phil Trans R Soc B 375. DOI 10.1098/rstb.2019.0637.** CLS v2: schema consistency, replay gating, explicit MTT responses.
@@ -0,0 +1,11 @@
+# Reconsolidation
+
+**Nader, Schafe & LeDoux 2000, Nature 406:722. DOI 10.1038/35021052.** Anisomycin in lateral amygdala after retrieval blocks subsequent expression of an already-consolidated fear memory — consolidated memories re-enter a labile state. Method: Pavlovian fear conditioning + post-reactivation infusion.
+
+**Tronson & Taylor 2007, Nat Rev Neurosci 8. DOI 10.1038/nrn2090.** Canonical review: successful replications across tasks/species AND wildly heterogeneous boundary conditions.
+
+**Boundary conditions (the crucial ledger part):** Sevenster, Beckers & Kindt 2014, Learn Mem 21. DOI 10.1101/lm.035493.114 — destabilization requires sufficient PREDICTION ERROR (violated expectation); a mere reminder does not. Exton-McGuinness, Lee & Reichelt 2015 (Neurosci Biobehav Rev, DOI 10.1016/j.bbr.2014.10.011) + Lee 2009 (Trends Neurosci): memory strength and age limit labilization (weak/very old memories resist).
+
+**Framing:** reconsolidation is best understood as UPDATING WITH A GATE (prediction error), not automatic restabilization. Whether it differs from ordinary updating only in timing: UNRESOLVED. Replication: rodent amygdala robust; human literature mixed (benzodiazepine/propranolol inconsistencies).
+
+**Adversarial canon (see 05_adversarial.md):** Rudy et al. 2006 (Learn Mem 13:1-3, anisomycin non-specificity); Lattal & Abel 2004 (PNAS 101:4667, deficits REVERSE with time = retrieval failure); Morris et al. 2006 (Neuron 50:479, nonspecific effects). Nader & Hardt 2009 (Nat Rev Neurosci 10:224-234) is the PRO case — the phenomenon requires a "case for" paper.
@@ -0,0 +1,10 @@
+# Schema learning
+
+**Tse et al. 2007, Science 313:76 (neuroscience pages 1130-1133 of that issue). DOI 10.1126/science.1135935.** Rats: flavor-place paired associates over weeks; schema-CONSISTENT new pairs learned in one trial and became hippocampal-lesion-insensitive (neocortical immediate-early gene markers); schema-INCONSISTENT pairs needed weeks + hippocampal replay. Strongest single result for schema-gated rapid integration — the finding that forced McClelland to revise CLS.
+
+**Tse et al. 2011, Science 333:891. DOI 10.1126/science.1205274.** Immediate-early gene activation in neocortex for schema-consistent encoding.
+
+**McClelland 2013, JEP:General 142(4):1190. DOI 10.1037/a0033812.** Principled CLS revision: schema consistency permits small neocortical weight changes that don't cross interference thresholds.
+
+**Replication status:** core effect replicated in Morris lab (2010/2011); human echo: Audrain & McAndrews 2022, Nature Communications. DOI 10.1038/s41467-022-33517-0 (fMRI scaffold-for-integration). BUT the neocortex-alone/hippocampal-independence interpretation is CONTESTED: Sharon, Moscovitch & Gilboa 2011, PNAS 108:17263-17268. DOI 10.1073/pnas.1005238108 (rapid arbitrary associates WITHOUT hippocampus); hippocampal damage still slows many rapid-learning tasks (Warren, Duff & Cohen line — "Fast mappers, slow learners," Cognitive Neuroscience 2019, DOI 10.1080/17588928.2019.1593120). Consolidation skeptics note schema effects may reflect inferential/RETRIEVAL processes rather than consolidation speed.
+NOTE: the R0-brief "Warren et al. 2020-2021 schema replication" is UNVERIFIED — nearest verified Warren items are 2019 Cognitive Neuroscience and 2014 hippocampus/word-learning work.
@@ -0,0 +1,15 @@
+# Software / AI analogues
+
+**Schaul, Quan, Antonoglou & Silver 2016, "Prioritized Experience Replay," ICLR (arXiv 1511.05952).** TD-error-prioritized sampling beats uniform on 41/49 Atari games. Relationship to Mattar & Daw: Schaul's priority ≈ "gain" (prediction error); Mattar-Daw adds "need" (utility). Caveats from adversarial pass: PER needs importance-sampling correction; Fedus et al. ICML 2020 (arXiv 2007.06700) — uniform replay often MATCHES PER, freshness dominates; Song & Qian 2024 (Sci Rep 53117) — single-criterion prioritization unstable.
+
+**Shin et al. 2017, "Continual Learning with Deep Generative Replay," NeurIPS (arXiv 1705.08690).** Generative replay mitigates catastrophic forgetting — direct computational CLS instantiation.
+
+**Blundell et al. 2016 (arXiv 1606.04460); Pritzel et al. 2017, "Neural Episodic Control" (arXiv 1703.01988).** Semi-tabular episodic buffers give one-shot value updates — the FAST system as the missing piece of slow-gradient deep RL. Validated as engineering lever.
+
+**Kauvar et al. 2023, "Curious Replay for Model-based Adaptation," ICML (arXiv 2306.15934).** Curiosity-prioritized replay for world models — direct translational echo of Mattar-Daw.
+
+**Kumaran, Summerfield, Hassabis & Botvinick 2016, Neuron 92** (Neuroscience-Inspired AI program statement); **Marblestone, Wayne & Kording 2016, Front Comput Neurosci** (constructive critique: biological plausibility gaps).
+
+**Kumaran, Hassabis & McClelland 2016, TiCS** — the pro-CLS-for-agents anchor the adversarial side must answer; De Lange et al. 2021 (IEEE TPAMI) establishes catastrophic interference as REAL in ML — CLS's distinctive content transfers as interference management.
+
+Bearing: in ML, fast/slow split and replay prioritization are VALIDATED AS ENGINEERING LEVERS; the contested part is whether biological replay does what the algorithms assume.
@@ -0,0 +1,77 @@
+# Sources Ledger (#376 R1)
+
+_Format: source_id | citation | DOI/arXiv | peer-reviewed | primary/secondary | read depth | bearing. All "(verified)" DOIs were confirmed via Crossref/OpenAlex/arXiv APIs during the outward pass._
+
+## Foundations
+- S001 | McClelland, McNaughton & O'Reilly 1995, Psych Review 102(3):419-457 | 10.1037/0033-295X.102.3.419 | yes | primary (theory) | RELEVANT_SECTIONS | canonical fast/slow split; catastrophic-interference motivation
+- S002 | Squire 1992, Archives of Neurology; Squire & Alvarez 1995, Curr Opin Neurobiol | — | yes | primary | RELEVANT_SECTIONS | standard-consolidation position CLS inherits
+- X001 | "McClelland/McNaughton/Whyte 1994 bitter-tracing" | NOT FOUND in Crossref/OpenAlex/arXiv | — | — | — | UNVERIFIABLE — excluded from ledger
+
+## Schema learning
+- S003 | Tse et al. 2007, Science 313:1130-1133 | 10.1126/science.1135935 | yes | primary | RELEVANT_SECTIONS | schema-gated rapid integration (rats)
+- S004 | Tse et al. 2011, Science 333:891 | 10.1126/science.1205274 | yes | primary | ABSTRACT/SECTIONS | neocortical gene activation
+- S005 | McClelland 2013, JEP:General 142(4):1190 | 10.1037/a0033812 | yes | primary (theory) | RELEVANT_SECTIONS | CLS schema revision
+- S006 | Audrain & McAndrews 2022, Nature Communications | 10.1038/s41467-022-33517-0 | yes | primary | ABSTRACT | human fMRI echo
+- S007 | Sharon, Moscovitch & Gilboa 2011, PNAS 108:17263-17268 | 10.1073/pnas.1005238108 | yes | primary | RELEVANT_SECTIONS | rapid learning WITHOUT hippocampus
+- S008 | Warren & Duff 2014, Hippocampus 24(8):920-932; Warren et al. 2014, J Neurosci 34(22); Warren et al. 2019, Cognitive Neuroscience | 10.1080/17588928.2019.1593120 | yes | primary | ABSTRACT/SECTIONS | amnesic fast-mapping failures; vmPFC schema locus
+- X002 | "Warren et al. 2020-2021 schema replication" | NOT FOUND as named | — | — | — | UNVERIFIABLE — nearest verified items listed above
+
+## Replay
+- S009 | Foster & Wilson 2006, Nature 440:680 | 10.1038/nature04587 | yes | primary | ABSTRACT | reverse replay discovery
+- S010 | Carr, Jadhav & Frank 2011, Nat Neurosci 14:147 | 10.1038/nn.2732 | yes | review/primary | RELEVANT_SECTIONS | dual retrieval/consolidation role
+- S011 | Ambrose, Pfeiffer & Foster 2016, Neuron 91:1342 | 10.1016/j.neuron.2016.07.047 | yes | primary | ABSTRACT | reward-gated replay
+- S012 | Mattar & Daw 2018, Nat Neurosci 21:1609 | 10.1038/s41593-018-0232-z | yes | primary (model) | RELEVANT_SECTIONS | utility-prioritized replay (gain+need)
+- S013 | Tanrıverdi et al. 2023, J Cogn Neurosci | 10.1162/jocn_a_02021 | yes | primary | RELEVANT_SECTIONS | co-reactivation ↔ FORGETTING
+- S014 | eLife 2023 ground-truth evaluation + 2024 dopamine exchange | 10.7554/eLife.85635.sa0 | yes | editorial/primary | ABSTRACT | replay-decoding methods contested
+
+## Reconsolidation
+- S015 | Nader, Schafe & LeDoux 2000, Nature 406:722 | 10.1038/35021052 | yes | primary | RELEVANT_SECTIONS | reconsolidation anchor (rodent amygdala)
+- S016 | Tronson & Taylor 2007, Nat Rev Neurosci 8 | 10.1038/nrn2090 | yes | review | RELEVANT_SECTIONS | boundary-condition heterogeneity
+- S017 | Sevenster, Beckers & Kindt 2014, Learn Mem 21 | 10.1101/lm.035493.114 | yes | primary | RELEVANT_SECTIONS | prediction-error gate
+- S018 | Rudy et al. 2006, Learn Mem 13:1-3 | — | yes | primary (critique) | RELEVANT_SECTIONS | anisomycin non-specificity
+- S019 | Lattal & Abel 2004, PNAS 101(13):4667 | — | yes | primary | RELEVANT_SECTIONS | deficits REVERSE with time
+- S020 | Morris et al. 2006, Neuron 50(3):479 | — | yes | primary | ABSTRACT | nonspecific anisomycin effects
+- S021 | Nader & Hardt 2009, Nat Rev Neurosci 10:224-234 | — | yes | review (PRO) | ABSTRACT | "a single standard" — pro-reconsolidation case
+- S022 | Exton-McGuinness, Lee & Reichelt 2015, Neurosci Biobehav Rev | 10.1016/j.bbr.2014.10.011 | yes | review | ABSTRACT | update-gate framing
+
+## Competing theories
+- S023 | Nadel & Moscovitch 1997, Hippocampus 7:217; Nadel et al. 2000, Psych Review 107:66 | — | yes | primary (theory) | RELEVANT_SECTIONS | MTT
+- S024 | Teyler & Rudy 2007, Hippocampus | 10.1002/hipo.20350 | yes | primary (theory) | RELEVANT_SECTIONS | indexing theory
+- S025 | Ryan et al. 2015, Science 348:1007 (+ Roy 2017 PNAS) | 10.1126/science.aaa5542 | yes | primary | RELEVANT_SECTIONS | engram persistence vs retrieval
+- S026 | Winocur & Moscovitch 2011, JINS 17(5):766; Moscovitch et al. 2016, Ann Rev Psych 67 | — | yes | review/theory | RELEVANT_SECTIONS | transformation hypothesis
+- S027 | Frankland et al. 2004, Science 304 | 10.1126/science.1094804 | yes | primary | ABSTRACT | remote-memory cortical shift + detail loss
+- S028 | Moscovitch & Gilboa 2021 OSF preprint | 10.31234/osf.io/yxbrs | no (preprint) | theory | ABSTRACT | TTT verified anchor
+- S029 | Gershman, Blei & Niv 2010, Psych Review 117:197 | — | yes | primary (theory) | RELEVANT_SECTIONS | latent-cause: reconsolidation = belief updating
+- X003 | "Winsneski/Bhatt TTT" | NOT FOUND | — | — | — | UNVERIFIABLE — replaced by S028
+
+## Generalization / interference
+- S030 | McCloskey & Cohen 1989; Ratcliff 1990 | — | yes | primary | RELEVANT_SECTIONS | catastrophic interference
+- S031 | O'Reilly & Norman 2002, TiCS; Norman & O'Reilly 2003, Psych Review 109 | 10.1016/s1364-6613(02)02005-3 | yes | primary (model) | RELEVANT_SECTIONS | pattern separation/completion (REC)
+- S032 | Payne et al. 2009, Neurobiol Learn Mem | 10.1016/j.nlm.2009.03.007 | yes | primary | ABSTRACT | sleep increases false memory
+- S033 | Fenn et al. 2009, Learn Mem | 10.1101/lm.1500808 | yes | primary | ABSTRACT | OPPOSITE: sleep reduces false memory
+- S034 | Anderson, Bjork & Bjork 1994, JEP:LMC 20:1063 | 10.1037/0278-7393.20.5.1063 | yes | primary | RELEVANT_SECTIONS | RIF discovery
+- S035 | Murayama et al. 2014, Psych Bull | 10.1037/a0037505 | yes | meta-analysis | ABSTRACT | RIF real on average, boundaries
+- S036 | RIF reliability papers (Eur Psychologist 2011, DOI 10.1027/1016-9040/a000040; Frontiers 2014, DOI 10.3389/fpsyg.2014.01343) | — | yes | meta/replication | ABSTRACT | reliability concerns
+
+## AI analogues
+- S037 | Schaul et al. 2016, ICLR (PER) | 1511.05952 | yes (conference) | primary | RELEVANT_SECTIONS | prioritized replay in RL + bias caveat
+- S038 | Fedus et al. 2020, ICML | 2007.06700 | yes | primary | RELEVANT_SECTIONS | uniform replay often matches PER
+- S039 | Song & Qian 2024, Scientific Reports 53117 | — | yes | review | ABSTRACT | prioritization instability
+- S040 | Shin et al. 2017, NeurIPS | 1705.08690 | yes | primary | ABSTRACT | generative replay
+- S041 | Blundell et al. 2016 (1606.04460); Pritzel et al. 2017 (1703.01988) | — | yes (conference) | primary | ABSTRACT | episodic control — fast store as engineering lever
+- S042 | Kauvar et al. 2023, ICML | 2306.15934 | yes | primary | ABSTRACT | curious replay for world models
+- S043 | Kumaran, Hassabis & McClelland 2016, TiCS 20(7) | 10.1016/j.tics.2016.05.004 | yes | review (theory) | RELEVANT_SECTIONS | CLS-for-agents anchor (pro)
+- S044 | McClelland, McNaughton & Lampinen 2020, Phil Trans R Soc B | 10.1098/rstb.2019.0637 | yes | review (theory) | RELEVANT_SECTIONS | CLS v2
+- S045 | De Lange et al. 2021, IEEE TPAMI 44(7) | — | yes | survey | RELEVANT_SECTIONS | catastrophic interference REAL in ML
+- S046 | Dacrema et al. 2019, RecSys | 1907.06902 | yes | primary (critique) | RELEVANT_SECTIONS | simple baselines beat complexity (adversarial)
+- S047 | O'Neil et al. 1996, Acta Informatica 33(4):351-385 (LSM-tree) | — | yes | primary | RELEVANT_SECTIONS | two-stores-as-engineering (adversarial)
+- S048 | Baria & Cross 2021, arXiv 2107.14042; Weisberg et al. 2008, J Cogn Neurosci 20(3) | — | mixed | critique | ABSTRACT | neuro-metaphor seduction (adversarial)
+- S049 | Lam et al. 2026, SSGM | 2603.11768 | no (preprint) | primary | RELEVANT_SECTIONS | semantic drift + immutable ledger fix
+- S050 | TRUSTMem 2026 | 2606.25161 | no (preprint) | primary | ABSTRACT | consolidation-pipeline corruption measured
+- S051 | Ólafsdóttir, Bush & Barry 2018, Curr Biol 28(1) | — | yes | review | RELEVANT_SECTIONS | replay dynamics load-bearing (adversarial)
+- S052 | Maynez et al. 2020, ACL | — | yes | primary | ABSTRACT | summarization hallucination (adversarial)
+- S053 | Roediger & McDermott 1995, JEP:LMC 21(4):803 | — | yes | primary | RELEVANT_SECTIONS | DRM gist falsification
+- S054 | Greshake et al. 2023 | — | yes | primary | ABSTRACT | indirect injection (adversarial, corroboration justification)
+- S055 | Candea & Fox 2003 HotOS (crash-only, context) | — | yes | primary | ABSTRACT | engineering pattern context
+
+Totals: primary sources 40+; reviews/surveys 6; competing theories 4 (MTT, indexing, latent-cause, TTT); negative/contrary sources 9; recent (2023-2026) 8; software analogues 8. Unverifiable R0 references excluded: 3 (X001-X003).
@@ -0,0 +1,56 @@
+# 03 — APPLICABILITY (Rune-specific, per surviving mechanism)
+
+_Mandate questions 1-15 answered per mechanism. Cheapest-mechanism rule enforced._
+
+## M1: Two-layer split (fast exact / slow generalized) — CLS-R1-C001
+1. Original problem: catastrophic interference in overlapping-representation learners.
+2. Rune's problem: REAL but milder — LLM consolidation rounds corrupt (C007), not gradient interference. Interference arrives via semantic drift, not weight overlap.
+3. Non-transferring biology: replay dynamics, sleep, neuromodulation, decay constants (d≈0.5 is a fit parameter to rodent data, S001).
+4. Computational differences: Rune's "fast store" is a database; perfect fidelity, zero decay — better than a hippocampus.
+5. Rune already solves it: YES — append-only entries + separate consolidation writer (T0 PROVEN). CLS adds justification, not mechanism.
+6-7. Complexity check: copying CLS's neural machinery (sparseness, decay curves) = unnecessary. Simplest capture: the existing two-writer design + drift guards (C007).
+8-9. Preserves exact evidence + provenance: YES (append-only + source columns).
+10. LLM dependence: unchanged. 11. Offline: YES. 12. Century-scale: YES (append-only is archival-friendly, child U).
+13. Event-driven conflict: none. 14. Owner: Memory (whole). 15. Cheapest mechanism: current design + index repair.
+
+## M2: Schema-gated fast path — CLS-R1-C002
+1. Original problem: weeks-long consolidation latency for schema-consistent info.
+2. Rune's problem: PARTIAL — consolidation lag exists but is not user-painful (nightly); urgency unproven.
+3-4. Non-transferring: schema detection in biology is implicit/in learned representations; Rune would need EXPLICIT schema matching (a new classifier) — the biological mechanism does not transfer, only the outcome.
+5. Rune already solves: NO (uniform consolidation).
+6. Copying increases complexity: YES — schema classifier = new LLM/deterministic component with its own error rate.
+7. Simpler capture: promotion-gate with independence check (C006) approximates the benefit without schema detection.
+8-12. Preserves evidence/provenance: yes if shadow-mode; offline yes; century-scale yes.
+13. Event-driven: compatible (consolidation trigger can be event-driven).
+14. Owner: Memory consolidation. 15. Cheapest: defer fast path until consolidation lag is measured as a real problem; shadow-mode experiment first.
+
+## M3: Utility-prioritized consolidation — CLS-R1-C003
+1. Original: finite replay budget in navigation.
+2. Rune: consolidation compute is cheap; the REAL constraint is lesson QUALITY (bad generalizations are expensive).
+3-4. Non-transferring: "need" signals (state relevance) are navigation-specific; Rune's equivalent ("what will shape future behavior") is unmeasurable without an outcome stream.
+5. Rune already solves: NO (full sweeps).
+6. Complexity: moderate — a utility estimator is a new component.
+7. Simpler: dedup + diversity + contradiction-targeting (from child E replay findings) captures most value without a utility model.
+8-12. Preserves evidence: yes. 10. LLM: can REDUCE (fewer consolidation calls). 11. Offline: yes.
+14. Owner: Memory consolidation. 15. Cheapest: diversity+contradiction sampling; utility model only if that fails.
+
+## M4: Retained covering episodes — CLS-R1-C004
+Already Rune doctrine (append-only). Applicability: total. Cheapest: current design. The R1 change is the RATIONALE (drift-correction, not reconsolidation) and the added obligation: drift tests comparing consolidated output against retained sources.
+
+## M5: Reconsolidation framing — CLS-R1-C005
+REJECTED as framing. The transferable atom: S017's prediction-error gate — update memories when reality VIOLATES them (contradiction detected), not on mere retrieval. That's just Rune's supersession trigger refined: supersede on contradiction, not on schedule. Owner: Memory write path.
+
+## M6: Independence-gated promotion — CLS-R1-C006
+Problem proven (wash_count gameable/unvalidated). Cheapest mechanism: promotion requires entries from ≥2 distinct sessions/sources touching the same content + no contradiction at promotion time. No new subsystem — a gate rule.
+
+## M7: Drift guards on consolidation — CLS-R1-C007
+Cheapest: (a) consolidate from immutable source text only (never from prior summaries) — already Rune doctrine, now load-bearing; (b) spot-check: N sampled claims per consolidation run verified against sources (existing claim_audit machinery); (c) drift metric: factual-consistency rate per run, logged. Owner: consolidation pipeline.
+
+## M8: Stage progression — CLS-R1-C008
+Keep stages as LABELS on one store (they already exist as kind/status fields); do NOT create separate stores per stage. Owner: Memory.
+
+## M9: Shadow-mode — CLS-R1-C009
+Cheap version: generalized beliefs enter with status=provisional and `acted_on=false` until independence gate passes; retrieval MAY surface them but marked. Owner: Memory entry status.
+
+## Boundary assignments (Q14)
+All mechanisms above: Memory authority. Nothing here belongs to Spine (structural truth), Substrate (execution), Persona (identity spec), or World Model (current-state belief) — with one note: consolidation's OUTPUT that describes the environment (facts about services/config) feeds the WM projection, but the belief-update authority stays in Memory.
@@ -0,0 +1,42 @@
+# 04 — INNOVATION (falsifiable hypotheses only; systems integration counts; novelty not manufactured)
+
+## I1: Drift-metered consolidation (closest prior art: SSGM 2603.11768, TRUSTMem 2606.25161)
+- Difference: SSGM prescribes immutable ledgers; Rune ADDS a measured drift metric per consolidation run (factual-consistency rate of sampled claims vs immutable sources) gating whether the run's output is accepted.
+- Mechanism: post-consolidation, sample N derived claims, verify against source episodes (existing claim_audit machinery); acceptance threshold; failed runs re-run or discarded.
+- Predicted advantage: consolidation corruption (C007, S050: 79%/50% reduction when guarded) caught before entering the belief layer.
+- Cost: N verification calls per run. Failure mode: verification LLM hallucinates agreement — mitigate with exact-quote checks.
+- Benchmark: drift rate per run before/after; retraction rate of accepted beliefs.
+- Rejection criterion: if drift rate is <1% unguarded across 10 runs, the meter is overhead — drop it.
+- Classification: EXPERIMENT. Evidence: EXTERNALLY_ESTABLISHED (drift) → RUNE_HYPOTHESIS (meter).
+
+## I2: Independence-gated promotion (closest prior art: dedup+recency recommenders, S046)
+- Difference: recommendation systems gate on statistical ensembles; Rune gates on SOURCE-GRAPH independence (distinct sessions, distinct evidence chains) + contradiction absence — a check available only because Rune keeps exact provenance.
+- Mechanism: promotion_eligible requires ≥2 independent evidence chains + zero active contradictions (using existing supersession graph).
+- Predicted advantage: eliminates frequency-gamed promotions (observed T0.1).
+- Cost: none — query over existing data. Failure mode: over-strict gate starves promotion (measure promotion latency).
+- Benchmark: promoted-belief retraction rate vs old wash_count gate.
+- Rejection criterion: if retraction rates equal, revert to simple gate.
+- Classification: ADAPT. Evidence: RUNE_APPLICABLE.
+
+## I3: Contradiction-triggered consolidation (event-driven, from S017's prediction-error gate via M5)
+- Difference: consolidation runs on DRIFT/CONTRADICTION events (new evidence contradicting an existing belief), not on timers — unifying Rune's event-driven mandate (req 34) with CLS's update-gate.
+- Mechanism: write-path contradiction detection (cross-encoder prefilter, child AC) emits consolidation-trigger events; consolidation processes affected beliefs, not the whole store.
+- Predicted advantage: timely belief updates (T0.1's non-deterministic trigger replaced); reduced LLM consumption (targeted runs).
+- Cost: contradiction detector errors (false positives → consolidation churn; rate-limit).
+- Benchmark: belief-update latency; LLM calls per belief-update; churn rate.
+- Rejection criterion: if targeted consolidation misses updates the sweep catches, hybrid (event + periodic sweep).
+- Classification: EXPERIMENT. Evidence: RUNE_HYPOTHESIS.
+
+## I4: Shadow-mode generalizations (closest prior art: TiMem dual-buffer probation; SSGM staging)
+- Difference: probation tied to the independence gate (I2) with `acted_on=false` semantics — beliefs are retrievable-but-marked, and the promotion decision is auditable in the ledger.
+- Predicted advantage: overgeneralization caught before behavior; audit trail of WHY a belief was promoted (req 24).
+- Cost: status complexity. Failure mode: shadow beliefs leak into behavior unmarked.
+- Benchmark: retraction rate of shadow-vs-direct promotions.
+- Rejection criterion: if no retraction-rate difference across 50 promotions, drop shadow mode.
+- Classification: EXPERIMENT. Evidence: RUNE_HYPOTHESIS.
+
+## I5: Two-lane consolidation benchmark (from CLS-R1-C010 refinement)
+- The innovation is METRIC-FIRST: before any prioritization scheme, implement the measurement (lesson quality: retraction rate, contradiction rate at promotion, independence depth) — then let data pick recency vs utility vs diversity.
+- Classification: EXPERIMENT (protocol). Evidence: RUNE_APPLICABLE (problem proven).
+
+No manufactured novelty: I1/I4 have close published prior art (SSGM, TiMem); I2 is integration of existing Rune provenance with standard hygiene; I3 is Rune's own event-driven mandate applied to CLS's update gate; I5 is measurement design. Each claims integration, not new mechanism.
@@ -0,0 +1,30 @@
+# 05 — ADVERSARIAL (independent research pass; actor general-31)
+
+_Mandate: assume the emerging CLS→Rune conclusions are WRONG. Find the strongest cited evidence. Honest failed attacks recorded._
+
+## Attack 1 — "Two stores / two writers": LARGELY SUCCESSFUL
+The dual-store pattern exists in engineering WITHOUT neuroscience and predates CLS-as-software: the LSM-tree (memtable + immutable SSTables + background compaction) is exactly "fast volatile store, slow durable store, consolidation pass" — O'Neil, Cheng, Gawlick & O'Neil, *Acta Informatica* 33(4):351-385, 1996. What CLS would add beyond cache+DB is the DYNAMICS: sharp-wave-ripple replay during sleep, interleaved training, slow integration — software Rune has none (Ólafsdóttir, Bush & Barry, *Current Biology* 28(1):R37-R50, 2018: the biological consolidation story rests on those dynamics). Metaphor critique: Baria & Cross, arXiv:2107.14042 (2021); Weisberg et al., "The seductive allure of neuroscience explanations," *J Cogn Neurosci* 20(3):470-477, 2008. **Strength: high.** Honest counter: CLS's distinctive content is INTERFERENCE MANAGEMENT — catastrophic interference is a real software phenomenon (De Lange et al. 2021) — but that doesn't rescue "two stores" specifically, which is not where CLS does work.
+
+## Attack 2 — Tse 2007 schema gating: SUCCESSFUL
+Tse et al., *Science* 313:1130-1133, 2007: rats only, food-reward flavor-place paired-associate task, WEEKS of training on six PAs before one-trial new-PA learning (paradigm per Hasan et al., *PLoS Biology* 17(12):e3000564, 2019, which concedes "difficult to map the concept of mental schema precisely onto existing neurobiological models"). Mechanism contested at publication: "Does assimilation into schemas involve systems or cellular consolidation?" (*Neurobiol Learn Mem* 2007) — an encoding-level prior-knowledge benefit is an equally live explanation. Human translation: Warren & Duff, *Hippocampus* 24(8):920-932, 2014 — amnesics fail to benefit from fast-mapping routes; Warren et al., *J Neurosci* 34(22), 2014 — schema influence located in vmPFC, not a hippocampus-bypassing speedup; Sharon/Moscovitch/Gilboa, *PNAS* 108:17263-17268, 2011 — rapid acquisition via a DIFFERENT route (rapid neocortical acquisition independent of hippocampus). **Strength: high** for animal-paradigm confinement; **medium** for the encoding-effect alternative (no paper cleanly rules it out).
+
+## Attack 3 — Mattar & Daw prioritized replay → utility-prioritized consolidation: SUCCESSFUL
+Mattar & Daw, *Neuron* 99(5):944-966, 2018 models utility-driven replay in simulated MAZE NAVIGATION. Ólafsdóttir et al. (2018): replay literature rests "mainly, but not exclusively, on rodents"; human replay evidence indirect. **No published evidence prioritized replay transfers to semantic-memory consolidation.** Closest analog (RL PER) is cautionary: Schaul et al. ICLR 2016 (PER needs importance-sampling correction for bias); Fedus et al. ICML 2020 — uniform replay often MATCHES PER, buffer freshness dominates; Song & Qian, *Scientific Reports* 53117, 2024 — single-criterion prioritization unstable across environments. **Strength: high** ("no transfer evidence"); PER pathologies make transfer worse.
+
+## Attack 4 — retained covering episodes: **FAILED** (conclusion survives; rationale replaced)
+The biology supports the attack: transformation theory (Winocur & Moscovitch, *JINS* 17(5):766-780, 2011; Moscovitch et al., *Ann Rev Psych* 67:1-36, 2016) holds the brain ROUTINELY DISCARDS details — retention of source episodes is not what biological consolidation does. Generative Agents (Park et al., UIST 2023) and MemGPT (Packer et al. 2023) run reflection/summarization without retained originals. **But the strongest engineering evidence cuts the other way**: SSGM (Lam et al., arXiv:2603.11768, 2026) documents SEMANTIC DRIFT — lossy iterative summarization progressively distorting stored facts ("likes mild spicy food" → "loves very spicy") — and prescribes an immutable episodic ledger reconciled against the mutable store. **Retention is justified by DRIFT-CORRECTION, not reconsolidation; the CLS conclusion survives on engineering grounds, losing its biological rationale.**
+
+## Attack 5 — reconsolidation → supersession: SUCCESSFUL
+Pharmacological foundation contested: Rudy, Biedenkapp, Moineau & Bolding, *Learn Mem* 13(1):1-3, 2006 — anisomycin has non-specific effects producing TEMPORARY amnesia unrelated to re-stabilization; Lattal & Abel, *PNAS* 101(13):4667-4672, 2004 — deficits REVERSE with time (retrieval-failure signature, not erasure); Morris et al., *Neuron* 50(3):479-489, 2006 — intrahippocampal anisomycin "nonspecific effects unconnected with reconsolidation." Alternative mechanisms (extinction facilitation: Myers & Davis 2002; Fischer et al. 2004) explain the same data. Tronson & Taylor, *Nat Rev Neurosci* 8:352-365, 2007 — wildly heterogeneous boundary conditions. **Correction to the R0 assumption: Hardt & Nader are on the PRO side** — Nader & Hardt, "A single standard for memory: the case for reconsolidation," *Nat Rev Neurosci* 10:224-234, 2009 — which is itself telling: the phenomenon requires a "case for" paper. **Strength: high.** Mapping Rune supersession to a phenomenon whose existence is pharmacology-dependent is speculation.
+
+## Attack 6 — corroboration gating: PARTIAL SUCCESS
+Dacrema, Cremonesi & Jannach, "Are we really making much progress?" *RecSys 2019* (1907.06902): 11 of 18 neural recommenders failed to beat well-tuned simple baselines — simple mechanisms (dedup, recency, linear) match added complexity; "corroboration" is structurally ensemble voting / confirmation-weighted averaging — standard hygiene needing no memory theory. **However, partially fails:** for single-user stores exposed to prompt injection and hallucinated inputs (Greshake et al. 2023, indirect injection), write-path validation has a distinct justification beyond dedup/recency. Corroboration-as-hygiene: yes; corroboration-as-redundant: unproven.
+
+## Attack 7 — workload mismatch: MIXED
+De Lange et al., *IEEE TPAMI* 44(7), 2021; Abel et al., *NeurIPS 2023*: agent settings are never-stationary, never-i.i.d. streams — nothing like hippocampal encoding at fixed event boundaries; MemGPT demonstrates top-down OS-style design working without neuro-anchoring; timescales differ by orders of magnitude. **Failed part:** Kumaran, Hassabis & McClelland, "What learning systems do efficient agents need?" *TiCS* 20(7):512-534, 2016 IS a peer-reviewed argument that CLS maps directly onto agent ML — the strongest pro-position the adversarial side must answer; and catastrophic interference (real, per De Lange) gives CLS teeth.
+
+## Attack 8 — consolidation damages exactness: SUCCESSFUL
+Maynez et al., "On faithfulness and factuality in abstractive summarization," *ACL 2020*: "substantial amounts of hallucinated content in all model generated summaries" — summarization (consolidation's software form) systematically injects error. Roediger & McDermott 1995 (DRM): gist compression GENERATES false content. Agent-specific: SSGM (2603.11768) places "semantic drift during consolidation updates" among top three failure modes; **TRUSTMem (arXiv:2606.25161, 2026) measures omission/corruption/hallucination introduced BY the consolidation pipeline itself** and cuts these 79.1%/50.0% vs baselines — direct evidence consolidation rounds degrade exactness an exact store wouldn't.
+
+## Bottom line
+**Strongest kills:** #5 (contested pharmacological foundation), #3 (no transfer evidence + fragile RL analog), #2 (rat food-reward paradigm, contested mechanism, human translation failures), #8 (documented consolidation-introduced corruption). **Attack #4 failed on the conclusion** (retention is right) **but won on the rationale** (drift-correction, not reconsolidation). **#1 strongest** where it says two stores are trivial engineering (LSM 1996), weakest against interference-based CLS content. **#6, #7 genuinely mixed** — with Kumaran 2016 as the pro-CLS anchor any honest synthesis must answer.
@@ -0,0 +1,22 @@
+# 06 — SYNTHESIS: mechanism matrix + reconciliation
+
+## Mechanism matrix
+
+| MECHANISM | ORIGINAL PURPOSE | EVIDENCE STRENGTH | LIMITATIONS | RUNE EQUIVALENT | RUNE GAP | TRANSFER VALIDITY | COST | FAILURE MODES | CLASSIFICATION | RUNE EVIDENCE LEVEL |
+|---|---|---|---|---|---|---|---|---|---|---|
+| Fast exact episodic store | evade catastrophic interference | STRONG (S001, S030, S045, S041) | biology's decay dynamics don't transfer; Rune's is better (perfect fidelity) | append-only entries (PROVEN) | index coverage (439 unindexed) | HIGH (as engineering; CLS adds interference rationale) | none | silent embedder failure (observed) | COVERED — repair | RUNE_PROVEN (store); index repair needed |
+| Slow statistical/generalized layer | interference-free generalization | STRONG (S001, S044) | transformation loses detail (S026) | knowledge.json consolidation (PROVEN to run) | drift guards; source-only inputs | HIGH | verification calls | consolidation hallucination (S052, S050) | ADAPT (drift-metered) | RUNE_APPLICABLE |
+| Two stores / two writers | separate update rates | STRONG as engineering (S047 LSM) / CLS content = interference mgmt (S045) | "two stores" trivial without dynamics (adversarial #1) | agent writer + consolidator (PROVEN) | none structural | HIGH | none | writer conflation | COVERED | RUNE_PROVEN |
+| Schema-gated rapid integration | fast assimilation with prior knowledge | CONTESTED (S003 effect real in rats; mechanism contested S007/S008; adversarial #2) | animal paradigm; encoding-vs-consolidation confound; schema detection cost | none | schema detector would be NEW | MEDIUM | new classifier | wrong schema match → corrupt fast promotions | EXPERIMENT (deferred until lag measured) | RUNE_HYPOTHESIS |
+| Utility-prioritized consolidation | finite replay budget, prospective value | MEDIUM (S012 model+converging ephys; navigation-specific) | no semantic-memory transfer evidence (adversarial #3); PER fragility (S038/S039) | none (full sweeps) | utility signal unmeasurable without outcome stream | MEDIUM | utility estimator | gamed/biased prioritization | EXPERIMENT (baseline-first) | RUNE_HYPOTHESIS |
+| Retained covering episodes | reconsolidation substrate (R0 claim) → drift correction (R1) | STRONG engineering (S049/S050/S052) | storage cost; biology actually discards (S026) | append-only (PROVEN) | drift tests vs sources | HIGH | storage | none serious | ADOPT (rationale corrected) | RUNE_APPLICABLE |
+| Reconsolidation → supersession | update-on-retrieval | CONTESTED (S015 rodent robust; S018-020 pharmacology contested) | write-vs-retrieval trigger mismatch | supersession (PROVEN, write-triggered) | prediction-error trigger refinement | LOW (framing) / MEDIUM (gate atom) | none | dogma without mechanism | REJECT framing; KEEP contradiction-triggered updates | RUNE_APPLICABLE (gate atom only) |
+| Interference management (pattern separation/completion) | avoid false memories + enable retrieval | STRONG computational (S031) + fMRI support | separation/completion trade-off is architectural, not tunable | dedupe/supersede gate (PROVEN); exact quotes | contradiction detection (missing) | HIGH | cross-encoder prefilter | false contradiction detections | ADAPT | RUNE_APPLICABLE |
+| Independent-evidence promotion gate | avoid premature generalization | MEDIUM (S017 gates; S046 simplicity warning) | nothing biological required | promotion_eligible (exists, frequency-driven — UNVALIDATED) | independence test | HIGH | none (query) | over-strict gate starves promotion | ADAPT | RUNE_APPLICABLE |
+| Shadow-mode provisional beliefs | prevent premature influence | MEDIUM (interference-threshold logic; TiMem convergence) | state complexity | status field (exists) | acted_on semantics | MEDIUM | none | shadow leakage | EXPERIMENT | RUNE_HYPOTHESIS |
+| Replay as teacher (offline re-processing) | neocortical learning from hippocampal record | MEDIUM (S012; decoding contested S014) | content-decoding methods disputed (S014); co-reactivation ↔ forgetting (S013) | consolidation as re-processing (PROVEN to run) | sampling policy | MEDIUM | none | replay amplifies poisoned content (child E) | COVERED (consolidation = Rune's replay; policy = experiments above) | RUNE_APPLICABLE |
+| Gist consolidation creates false content | efficiency of gist | CONTESTED in biology (S032 vs S033); STRONG in software (S052, S049, S050) | — | LLM summarization | drift guards | HIGH | verification | silent corruption | ADOPT (guards) | RUNE_APPLICABLE |
+
+## Synthesis narrative
+
+CLS's load-bearing content for Rune is NOT the two-store anatomy (that's LSM-tree engineering, S047) and NOT the biological dynamics (replay/sleep — S051) and NOT reconsolidation (pharmacologically contested — S018-020). Three things transfer: (1) **the interference-management problem is real in software** (S045) and Rune's exact store + supersession is the pattern-separation answer; (2) **consolidation corrupts** (S049/S050/S052) — so the slow layer requires drift guards and retained sources as engineering requirements; (3) **updates are gated by prediction error, not schedules** (S017, reframed via S029's latent-cause account as plain belief-updating) — which lands Rune's consolidation trigger on events, not timers. The five-stage progression is Rune's own construct: keep it as labels, not stores. Competing theories (MTT, indexing, latent-cause) matter mostly as warnings: the biology does not settle what CLS's early readings assumed, and the latent-cause account explains reconsolidation phenomena as ordinary belief-updating — which is exactly the software form Rune should implement. Kumaran 2016 (S043) remains the strongest pro-CLS-for-agents argument and survives the adversarial pass on interference grounds.
@@ -0,0 +1,28 @@
+# 07 — QUALIFICATION (#376 R1)
+
+_Classification separate from evidence level. Each item: confidence, sources (for/against), assumptions, cost, failure modes, T1 impact, experiment, falsifier, rejection criterion. Evidence levels per claims.json._
+
+## Qualified mechanisms
+
+| # | Mechanism | Classification | Evidence level | Conf | For | Against | T1 impact | Experiment | Rejection criterion |
+|---|---|---|---|---|---|---|---|---|---|
+| Q1 | Two-layer split (exact episodes + generalized layer, two writers) | COVERED — keep; repair index coverage | EXTERNALLY_ESTABLISHED + RUNE_APPLICABLE (store RUNE_PROVEN) | HIGH | S001/S045/S041/S049 | S047 (trivial-engineering caveat — does not defeat interference content) | HIGH (retention/rotation decision) | index-repair + embedder loud-failure (not CLS but urgent) | single-store drift parity |
+| Q2 | Drift guards on consolidation (source-only inputs, quote-checks, drift metric) | ADOPT | EXTERNALLY_ESTABLISHED + RUNE_APPLICABLE | HIGH | S049/S050/S052/S053 | S033 (biology direction contested — irrelevant to software) | HIGH | drift-rate measurement per run | drift <1% unguarded over 10 runs |
+| Q3 | Retained covering episodes (append-only; rationale = drift correction) | ADOPT — already doctrine | RUNE_APPLICABLE | HIGH | S049 adversarial-confirmed | S026 (biology discards — explicitly not imitated) | MEDIUM | retention vs drift controlled test | consolidation-without-retention parity |
+| Q4 | Independence-gated promotion (replaces wash_count frequency gate) | ADAPT | RUNE_APPLICABLE (problem RUNE_PROVEN via T0.1) | HIGH | S054 threat model; wash_count unvalidated | S046 simplicity | MEDIUM | promotion retraction-rate A/B | equal retraction rates |
+| Q5 | Contradiction-triggered (event-driven) consolidation | ADAPT | RUNE_HYPOTHESIS | MEDIUM | S017 gate atom; S029 latent-cause framing | trigger-churn risk | MEDIUM | targeted-vs-sweep update latency | missed updates vs sweep |
+| Q6 | Shadow-mode provisional generalizations | EXPERIMENT | RUNE_HYPOTHESIS | MEDIUM | TiMem convergence; interference logic | complexity | MEDIUM | shadow-vs-direct retraction rates | no difference over 50 promotions |
+| Q7 | Utility-prioritized consolidation | EXPERIMENT (refined from R0) | RUNE_HYPOTHESIS (transfer unproven) | LOW-MED | S012/S011 | adversarial#3; S038/S039 | LOW-MED | lesson-quality A/B vs dedup+diversity baseline | no improvement over baseline |
+| Q8 | Schema-gated fast path | EXPERIMENT (DEFERRED until consolidation lag measured as real) | CONTESTED externally | LOW-MED | S003/S005 | S007/S008; adversarial#2 | LOW | lag measurement first; then shadow fast path | fast-path retraction rate higher |
+| Q9 | Five-stage progression (episode→observation→belief→procedure) | ADAPT (as labels on one store) | RUNE_APPLICABLE (transitions RUNE_PROVEN) | MEDIUM | S001/S044 stage logic | over-engineering risk (meta-review) | LOW-MED | friction vs lesson-quality | collapse to 2-stage if no benefit |
+| Q10 | Reconsolidation framing for supersession | REJECT (framing); KEEP prediction-error-gated updates | CONTESTED | HIGH (in the reject) | S015/S017/S029 | S018/S019/S020; adversarial#5 | LOW | — | — |
+
+## Evidence-level discipline (mandate §17)
+No mechanism above is RUNE_PROVEN beyond what T0/T0.1 already established (two-writer structure, append-only, promotion-gate existence). Everything new — drift guards, independence gates, event triggers, shadow mode — is RUNE_HYPOTHESIS until Rune experiments run. Q1's store layer is the only RUNE_PROVEN item, and its gap (index coverage) is operational, not architectural.
+
+## Evidence-level separation statement (G10)
+
+Explicit separation of layers, per mandate:
+- BIOLOGICAL MECHANISM (what the brain does): fast sparse hippocampal store, slow overlapping neocortex, replay-as-teacher, schema-gated integration, reconsolidation-with-prediction-error-gate. Established to varying degrees (S001-S031) — NOT Rune claims.
+- SOFTWARE ANALOGY (what the mechanism becomes in software): two-layer store, consolidation pass, prioritized sampling, update-on-violation. Validated as engineering levers (S037-S041, S045) — still NOT Rune claims.
+- RUNE PRESCRIPTION (what Rune should do): only items with RUNE_APPLICABLE or stronger levels in claims.json — the two-layer split (RUNE_PROVEN structure), drift guards (RUNE_APPLICABLE), independence gate (RUNE_APPLICABLE), contradiction-triggered updates (RUNE_HYPOTHESIS), shadow mode (RUNE_HYPOTHESIS). Every EXTERNALLY_ESTABLISHED claim that entered Rune prescription passed through an explicit applicability argument (03_applicability.md) and carries its Rune evidence level in claims.json. No biological finding is cited as Rune proof anywhere in this artifact set.
@@ -0,0 +1,46 @@
+# #376 R1 — Complementary Learning Systems: FINAL (durable technical reference)
+
+_This document is the domain's durable reference. It assumes the reader has the pipeline artifacts alongside: 00_issue_scope, 01_inward, 02_outward/ (9 files + sources.md + claims.json), 03_applicability, 04_innovation, 05_adversarial, 06_synthesis, 07_qualification, r0_reconciliation._
+
+## 1. The domain in one page
+
+Complementary Learning Systems (McClelland, McNaughton & O'Reilly 1995) claims cognition needs two learning systems: a fast, sparse, episodic store (hippocampus) that learns instantly but interferes catastrophically when asked to generalize, and a slow, overlapping, statistical store (neocortex) that generalizes well but must learn slowly and interleaved. The hippocampus acts as a teacher: replay of episodes drives slow neocortical integration. Thirty years of development produced: schema-gated rapid integration (Tse 2007; McClelland 2013 — the theory's biggest self-revision), utility-prioritized replay (Mattar & Daw 2018), reconsolidation (Nader 2000 — with a contested pharmacological foundation), transformation theory (memories are transformed, not copied — Winocur/Moscovitch), and serious competitors (MTT, hippocampal indexing, latent-cause/Bayesian accounts). In ML, the pattern is validated as engineering: episodic control buffers, experience/generative replay, and catastrophic interference are all real (S037-S045).
+
+## 2. What transfers to Rune, and what does not
+
+**Transfers (as engineering, with Rune evidence):**
+- The two-layer split with separate update rates. Rune has it structurally (RUNE_PROVEN); its urgency comes from consolidation corruption (S049/S050), not from CLS.
+- Retention of exact sources. NOT because reconsolidation needs them (the biology discards — S026) but because iterative summarization measurably drifts (S049) and the consolidation pipeline itself corrupts (S050).
+- Update gating by prediction error. When new evidence contradicts a belief, update; mere retrieval changes nothing. (S017's gate, reframed through S029's latent-cause account as ordinary belief-updating.)
+- Interference management. Catastrophic interference is real in ML (S045); Rune's exact store + supersession + dedupe is the pattern-separation answer.
+
+**Does not transfer:**
+- The biological dynamics (replay during sleep, decay constants, neuromodulation). Rune has no sleep; its "fast store" is a database with perfect fidelity.
+- Reconsolidation as a phenomenon to imitate. Pharmacologically contested (S018-020); the useful atom (update-on-violation) is captured by plain belief-updating.
+- Schema detection as a mechanism. In biology it is implicit in learned representations; in Rune it would be an explicit, error-prone classifier.
+- Utility signals without an outcome stream. Mattar-Daw "need" requires measuring future value; Rune cannot measure that yet.
+
+## 3. The corrected rationale for Rune's pipeline
+
+R0 said: CLS justifies episode→observation→belief→procedure with a schema fast path, utility prioritization, and reconsolidation-style updates. R1 corrects: the pipeline is justified by (a) drift-correction economics (exact sources are the only defense against measured consolidation corruption), (b) interference management (dedupe/supersede/exact quotes), (c) prediction-error-gated updates (contradiction-triggered, event-driven), and (d) independence-gated promotion (because frequency is gameable and unvalidated). Schema fast paths and utility prioritization are experiments, not architecture.
+
+## 4. Experiments this domain requires (in order)
+1. **Drift-metered consolidation**: measure factual-consistency of derived claims vs immutable sources per run. Rejection criterion: drift <1% unguarded → drop the meter.
+2. **Promotion-gate A/B**: independence-gated vs wash_count-gated promotion; metric = promoted-belief retraction rate.
+3. **Consolidation trigger A/B**: event-driven (contradiction-triggered) vs periodic sweep; metrics = update latency, missed updates, LLM calls.
+4. **Prioritization A/B**: recency vs utility vs dedup+diversity sampling; metric = lesson quality (retraction/contradiction rates).
+5. **Shadow mode**: shadow-vs-direct promotion retraction comparison over 50 promotions.
+
+## 5. What would change this conclusion
+- A credible "summarize-and-discard works fine" result (would force re-examination of retention).
+- Consolidation-lag evidence showing nightly updates are user-painful (would revive the schema fast path).
+- An outcome-stream making utility measurable (would upgrade C003 from experiment to design input).
+- Replication of schema-acceleration as consolidation-specific (would strengthen C002).
+
+## 6. Remaining unknowns (bounded)
+- Whether Rune's consolidation is atomically checkpointed end-to-end (code verification, not research).
+- Optimal independence-test depth (2 chains? 3?) — needs the promotion A/B.
+- Whether interval-confidence elicitation helps on local models (child Q's open question, shared).
+
+## 7. Production implementation
+NONE. This tranche is research only. Everything above enters T1 as design input and experiment protocol.
@@ -0,0 +1,125 @@
+{
+  "note": "Narrow, falsifiable claims. external_status: EXTERNALLY_ESTABLISHED|CONTESTED|WEAK. rune_status: RUNE_APPLICABLE|RUNE_HYPOTHESIS|RUNE_EXPERIMENTALLY_SUPPORTED|RUNE_PROVEN.",
+  "claims": [
+    {
+      "claim_id": "CLS-R1-C001",
+      "claim": "Fast exact episodic capture + slow statistical generalization is the right two-layer split for Rune",
+      "evidence_for": ["S001 core CLS logic", "S045 catastrophic interference REAL in ML", "S041 episodic fast store validated as engineering lever", "S049 semantic drift from lossy consolidation documents what happens without the split"],
+      "evidence_against": ["S047 two-stores is trivial engineering (LSM-tree) needing no CLS", "S048 neuro-metaphor seduction", "S051 biological dynamics (replay/sleep) not present in software"],
+      "external_status": "EXTERNALLY_ESTABLISHED (as engineering pattern; CLS's interference-management content adds justification)",
+      "rune_status": "RUNE_APPLICABLE (T0: two writers exist structurally; T0.1 leak = implementation gap not design gap)",
+      "architecture_impact": "HIGH",
+      "falsifier": "F1: if single-store designs match two-store on drift+interference metrics at Rune scale, the split is preference",
+      "experiment_required": true,
+      "classification": "COVERED (architecture) — implement cleanly"
+    },
+    {
+      "claim_id": "CLS-R1-C002",
+      "claim": "Schema-gated rapid consolidation (fast path) is justified for Rune",
+      "evidence_for": ["S003 Tse 2007 (rats, replicated in-lab)", "S005 McClelland 2013 principled revision", "S006 human fMRI echo"],
+      "evidence_against": ["S007 rapid learning WITHOUT hippocampus (alternative route)", "S008 amnesic failures of fast mapping", "adversarial#2: encoding-level prior-knowledge benefit equally live; food-reward rat paradigm; mechanism contested at publication"],
+      "external_status": "CONTESTED (effect real; MECHANISM and hippocampal-independence interpretation contested)",
+      "rune_status": "RUNE_HYPOTHESIS",
+      "architecture_impact": "MEDIUM",
+      "falsifier": "F2: if fast-path promotions show higher retraction rates than slow-path, revert",
+      "experiment_required": true,
+      "classification": "EXPERIMENT (fast path with shadow-mode + retraction-rate kill criterion)"
+    },
+    {
+      "claim_id": "CLS-R1-C003",
+      "claim": "Consolidation should be utility-prioritized (expected future value), not recency/frequency-based",
+      "evidence_for": ["S012 Mattar & Daw: prioritization by gain+need, NOT recency", "S011 reward-gating precursor", "S037 PER validates prioritization as lever (with bias caveat)"],
+      "evidence_against": ["adversarial#3: model is maze-navigation-specific; NO transfer evidence to semantic consolidation", "S038 uniform replay often matches PER", "S039 single-criterion prioritization unstable", "Rune's own wash_count (frequency) never validated as quality signal (T0.1)"],
+      "external_status": "EXTERNALLY_ESTABLISHED (in navigation/RL) / TRANSFER_UNPROVEN (to semantic memory consolidation)",
+      "rune_status": "RUNE_HYPOTHESIS",
+      "architecture_impact": "MEDIUM",
+      "falsifier": "F3: if utility-weighted consolidation produces no measurable lesson-quality improvement over recency+dedup baselines, use the simple version",
+      "experiment_required": true,
+      "classification": "EXPERIMENT (utility signal must be measured, not assumed; simple baseline first)"
+    },
+    {
+      "claim_id": "CLS-R1-C004",
+      "claim": "Covering episodes must be retained after generalization (no destructive consolidation)",
+      "evidence_for": ["S026 transformation theory: details are LOST — retention fights biology but preserves verifiability", "S049 SSGM: semantic drift from lossy summarization; immutable ledger is the fix", "S050 TRUSTMem: consolidation pipeline itself corrupts (79%/50% reduction when guarded)", "S052 summarization hallucination", "S053 DRM gist falsification"],
+      "evidence_against": ["adversarial#4 FAILED: no 'summarize-and-discard works fine' evidence found; biology discards freely (S026)", "storage cost is real"],
+      "external_status": "EXTERNALLY_ESTABLISHED (as engineering requirement — drift correction; NOT as biological requirement)",
+      "rune_status": "RUNE_APPLICABLE (append-only entries already PROVEN in Rune)",
+      "architecture_impact": "HIGH",
+      "falsifier": "F5: if consolidation-without-retention shows no drift/corruption in controlled Rune test, retention is preference",
+      "experiment_required": false,
+      "classification": "ADOPT (already Rune doctrine; rationale CORRECTED from reconsolidation to drift-correction)"
+    },
+    {
+      "claim_id": "CLS-R1-C005",
+      "claim": "Reconsolidation (retrieval-triggered lability) maps to Rune supersession",
+      "evidence_for": ["S015 anchor phenomenon", "S017 prediction-error gate is a real mechanism", "S029 latent-cause framing: reconsolidation = belief updating"],
+      "evidence_against": ["S018/S019/S020 pharmacological foundation contested (temporary amnesia, reversal with time, nonspecific effects)", "adversarial#5: mapping to a phenomenon whose existence is pharmacology-dependent is speculation", "Rune supersession is WRITE-triggered not retrieval-triggered — the analogy was asserted in R0, not derived"],
+      "external_status": "CONTESTED (rodent robust; human mixed; pharmacology-dependent)",
+      "rune_status": "RUNE_HYPOTHESIS (weak)",
+      "architecture_impact": "LOW",
+      "falsifier": "F4: if reconsolidation's useful property (update-on-retrieval) is fully captured by plain belief-updating (S029), drop the neuroscience framing entirely",
+      "experiment_required": false,
+      "classification": "REJECT the reconsolidation FRAMING; KEEP plain belief-updating with prediction-error gating (S017's gate is the transferable part)"
+    },
+    {
+      "claim_id": "CLS-R1-C006",
+      "claim": "Independent-evidence corroboration should gate promotion (not frequency/wash_count)",
+      "evidence_for": ["S017 prediction-error/independence gates exist in biology", "S054 injection threat makes write-path validation distinct from dedup", "Rune wash_count demonstrably unvalidated (T0.1: superseded entries with high counts)"],
+      "evidence_against": ["S046 11/18 neural recommenders failed vs simple baselines", "adversarial#6: corroboration = ensemble voting hygiene, no memory theory needed"],
+      "external_status": "EXTERNALLY_ESTABLISHED (as hygiene)",
+      "rune_status": "RUNE_APPLICABLE (problem proven: frequency gating is gameable and unvalidated)",
+      "architecture_impact": "MEDIUM",
+      "falsifier": "If dedup+recency matches corroboration-gated promotion on lesson quality, use simple version",
+      "experiment_required": true,
+      "classification": "ADAPT (independence test as promotion criterion; simple implementation first)"
+    },
+    {
+      "claim_id": "CLS-R1-C007",
+      "claim": "Consolidation introduces errors an exact store wouldn't have (gist hallucination/drift)",
+      "evidence_for": ["S052 summarization hallucination (ACL 2020)", "S053 DRM", "S049 SSGM drift documented", "S050 TRUSTMem measured consolidation corruption"],
+      "evidence_against": ["S033 sleep REDUCED false memories (biology direction contested)", "none in software — drift is well-documented"],
+      "external_status": "EXTERNALLY_ESTABLISHED (in software)",
+      "rune_status": "RUNE_APPLICABLE (Rune consolidates via LLM summarization — same mechanism)",
+      "architecture_impact": "HIGH",
+      "falsifier": "none plausible — mitigation is the question, not existence",
+      "experiment_required": false,
+      "classification": "ADOPT mitigations: consolidate from immutable sources only; quote-verification before promotion; drift tests"
+    },
+    {
+      "claim_id": "CLS-R1-C008",
+      "claim": "Evidence→episode→observation→belief→procedure progression with deterministic state transitions and LLM-only content operations",
+      "evidence_for": ["S001/S044 CLS stage structure", "S029 latent-cause: updating = inference (LLM content ops) over deterministic state", "S041 episodic control (fast layer engineering-validated)", "Rune T0: state transitions already deterministic; promotion gate exists"],
+      "evidence_against": ["stage boundaries are Rune's construct, not CLS's — CLS has two stores, not five stages", "over-engineering risk (meta-review #3)"],
+      "external_status": "EXTERNALLY_ESTABLISHED (as two-layer); the 5-stage elaboration is RUNE-designed",
+      "rune_status": "RUNE_APPLICABLE (transitions already deterministic; T0 PROVEN)",
+      "architecture_impact": "MEDIUM",
+      "falsifier": "if 5-stage bookkeeping adds friction without measurable lesson quality, collapse to 2-stage (episodes + consolidated)",
+      "experiment_required": true,
+      "classification": "ADAPT (keep progression but treat intermediate stages as labels on one store, not separate stores)"
+    },
+    {
+      "claim_id": "CLS-R1-C009",
+      "claim": "Shadow-mode provisional generalizations (behavioral-only until corroborated by independent episodes)",
+      "evidence_for": ["S003/S005 interference-threshold logic supports probation", "S050 consolidation corruption risk argues for pre-promotion probation", "TiMem dual-buffer probation (prior closure O) — convergent design"],
+      "evidence_against": ["no biological analog needed — this is engineering", "adds state complexity"],
+      "external_status": "RUNE_HYPOTHESIS (engineering design informed by interference thresholds)",
+      "rune_status": "RUNE_HYPOTHESIS",
+      "architecture_impact": "MEDIUM",
+      "falsifier": "if shadow-mode beliefs never differ from direct promotion outcomes in retrospective analysis, drop the mode",
+      "experiment_required": true,
+      "classification": "EXPERIMENT"
+    },
+    {
+      "claim_id": "CLS-R1-C010",
+      "claim": "Recency-based consolidation priority should be rejected",
+      "evidence_for": ["S012 replay NOT recency-biased in biology", "S013 co-reactivation ↔ forgetting", "R0 rationale holds"],
+      "evidence_against": ["S038 freshness dominates in RL replay buffers (recency has value!)", "adversarial#3: no transfer evidence either direction for semantic consolidation"],
+      "external_status": "CONTESTED",
+      "rune_status": "RUNE_HYPOTHESIS",
+      "architecture_impact": "LOW",
+      "falsifier": "if recency-weighted consolidation matches utility-weighted on lesson quality, the rejection is dogma",
+      "experiment_required": true,
+      "classification": "REFINED — reject recency-ONLY; do not assume utility-weighting either; benchmark both against dedup+diversity baseline"
+    }
+  ]
+}
@@ -0,0 +1,22 @@
+# R0 → R1 Reconciliation (#376 CLS)
+
+_Every material R0 claim, revisited. R0 source: memory-atlas/families/complementary_learning_systems.md_
+
+| R0 claim | R1 verdict | Why |
+|---|---|---|
+| Fast exact episodic storage + slow generalized knowledge is the right split | **CONFIRMED** (with rationale correction) | Survives adversarial pass as ENGINEERING pattern (S047 caveat noted); interference management (S045) is the transferable content, not the anatomy. Rune's gap is operational (index coverage), not architectural. |
+| Two stores / two writers analogy | **CONFIRMED** | Already RUNE_PROVEN structurally (T0); CLS adds justification via interference content; adversarial "it's just LSM" accepted as caveat, not defeat. |
+| Schema-gated fast path | **REFINED → DEFERRED-EXPERIMENT** | R0 said "justified by CLS"; R1: mechanism CONTESTED externally (encoding-vs-consolidation confound, S007/S008; adversarial#2), schema detection is a new error-prone component, and consolidation lag is unmeasured. Reclassified EXPERIMENT with a precondition (measure lag first). Evidence level downgraded from implied-strong to RUNE_HYPOTHESIS. |
+| Utility-prioritized consolidation | **REFINED** | R0 accepted Mattar-Daw transfer; R1: transfer to semantic consolidation UNPROVEN (adversarial#3), PER has documented fragilities (S038/S039), and Rune's own frequency signal was never validated. Reclassified: benchmark utility vs dedup+diversity baseline before any adoption; recency-only rejection itself downgraded to "benchmark it" (C010). |
+| Retained covering episodes | **CONFIRMED (rationale replaced)** | Adversarial#4 FAILED to kill the conclusion but SUCCEEDED against the rationale: not reconsolidation — DRIFT CORRECTION (S049/S050). Evidence level upgraded: RUNE_APPLICABLE with software evidence, not biological analogy. |
+| Shadow-mode provisional beliefs | **CONFIRMED as EXPERIMENT** | R0 hypothesis; R1 adds convergence (TiMem) + falsifiable kill criterion + leakage failure mode. |
+| Reconsolidation → supersession mapping | **OVERTURNED** | R0 asserted the mapping; R1: pharmacological foundation contested (S018-020), trigger mismatch (write vs retrieval), and S029's latent-cause account explains the same data as ORDINARY belief-updating — the neuroscience framing adds nothing. Keep supersession; keep the prediction-error GATE atom; drop the framing. (Also corrected: Nader & Hardt are PRO-reconsolidation — R0's assumption was inverted.) |
+| Independent-evidence corroboration gate | **CONFIRMED (evidence strengthened)** | R0 hypothesis; R1 adds: wash_count PROVEN unvalidated (T0.1), injection threat makes write-path validation distinct from dedup (S054), and the gate is implementable over existing provenance at zero schema cost. Still needs the retraction-rate experiment. |
+| Destructive consolidation rejection | **CONFIRMED (upgraded)** | R0 rejected on principle; R1: drift/corruption is MEASURED in software (S049/S050/S052) — the strongest evidence in the entire domain, and it's engineering evidence, not analogy. |
+| Recency-based consolidation rejection | **REFINED** | R0 rejected recency citing S012; R1: S038 shows freshness dominates in RL buffers — recency has documented value; the honest position is benchmark recency vs utility vs diversity, not dogma. |
+| Evidence→episode→observation→belief→procedure progression | **REFINED** | Keep the stages as LABELS on one store (T0-proven fields); do NOT materialize separate stores per stage. Over-engineering guard from meta-review. |
+| T1 impact = YES (three missing mechanisms) | **REFINED** | The three R0 mechanisms become: drift guards (ADOPT), independence gate (ADAPT), event-triggered consolidation (EXPERIMENT). Schema fast-path dropped from T1 scope to deferred-experiment. |
+
+Tally: CONFIRMED 4 · REFINED 5 · OVERTURNED 1 · STILL_UNKNOWN 0 (all bounded into experiments).
+
+Honest note: R0 was directionally right on every claim — but three of its rationale chains did not survive contact with the primary literature, and one mechanism (schema fast path) was prematurely promoted toward T1. That is exactly what R1 exists to catch.
@@ -0,0 +1,3 @@
+# Sources Ledger (#376 R1)
+
+Columns: source_id | title | authors | year | venue | DOI/arXiv/URL | peer-reviewed | primary/secondary | version | research question | mechanism | method | claims supported | claims contradicted | limitations | Rune relevance | confidence | read depth (ABSTRACT_ONLY/RELEVANT_SECTIONS/FULL_PAPER/CODE_AUDITED/REPLICATED)
@@ -0,0 +1,16 @@
+# AA — Security / Memory Poisoning: R1 Domain Master
+## #402
+
+**Executive summary:** TMA-NM (2606.24322) is the design law — content- and lineage-based trust are both launderable; trust MUST be bound at write time from input channel, elevated only via human/Sybil-resistant processes. MemCollusion (2608.01637) proves per-record trust scoring is unsound — taint must propagate to derived records.
+
+**Attack landscape:** AgentPoison (NeurIPS 2024, >80% ASR <0.1% poison rate); PoisonedRAG (USENIX 2025, 90% ASR from 5 texts in 1M corpus); InjecMEM (COLM 2026, single-interaction injection); MemCollusion/Salami (81.3% save rate, 75% ASR — kills individual-record defenses); MPBench (systematic taxonomy — aggressive agents more exploitable); eTAMP (environment-observation-only poisoning); Sleeper poisoning (60-89% action success, dormant).
+
+**Defenses:** TMA-NM origin binding (provable, machine-checked); CaMeL 2503.18813 (control/data-flow separation — memory reads are untrusted data, never trigger state mutations); trust-aware retrieval with temporal decay (2601.05504 — but thresholds brittle); forensic trajectory detection (2606.30566 — AUC 0.99 but 25-53% benign FPR; warning against recall-signal auto-quarantine); GRAB-RAG — sufficiency-based confidence cannot substitute for source-based trust.
+
+**Design invariants for Rune:** (1) channel-bound trust, never laundered by summarization; (2) elevation only via human review or Sybil-resistant corroboration; (3) taint propagation through derived memories; (4) time-bounded re-verification; (5) quarantine as visible state, not silent deletion.
+
+**SQLite transfer (highest of all domains):** origin_channel, trust_level (write-time bound), taint set, quarantined_at, supersession/rollback rows.
+
+**Classification:** ADOPT (TMA-NM origin binding, CaMeL data-flow separation, taint propagation, quarantine-as-state); EXPERIMENT (temporal decay trust, forensic trajectory detection); REJECT (per-record trust scoring, sufficiency-based abstention for security). Evidence: EXTERNALLY_ESTABLISHED.
+
+**R0:** CONFIRMED 3 · REFINED 2 · OVERTURNED 0 · Sum = 5.
@@ -0,0 +1,14 @@
+# AB — Multimodal / Artifact Memory: R1 Domain Master
+## #403
+
+**Executive summary:** Literature multimodal "memory" is never a second store — it is either a shared embedding space (Nomic Embed Vision 2406.18587 explicitly shares latent space with nomic-embed-text) or distilled artifacts (Voyager 2305.16291 code skills; Agent Workflow Memory 2409.07429 abstracted routines, +24.6%/+51.1% relative success). Screenshots in GUI agents are INPUT, not retained memory.
+
+**Verdicts:**
+- **ADOPT**: artifact memory as metadata (path, hash, MIME, mtime, provenance — ledger already does this)
+- **ADAPT**: code-artifact distillations tied to outcomes (Voyager/AWM pattern)
+- **REJECT**: storing screenshots/images as memory entries (zero payoff evidence for text-first agent, real token/storage cost)
+- **FUTURE**: nomic-embed-vision into the EXISTING index (no second store) if a concrete "I saw that file/error before" retrieval failure appears — EXPERIMENT only then
+
+**Evidence grades:** CLIP 2103.00020, Voyager, AWM, CogAgent = A (peer-reviewed); Nomic Embed Vision, OS-Copilot, surveys = B.
+
+**R0 reconciliation:** CONFIRMED 2 · REFINED 1 · OVERTURNED 0 · Sum = 3.
@@ -0,0 +1,14 @@
+# AC — Local / Offline Intelligence: R1 Domain Master
+## #404
+
+**Executive summary:** Nomic Embed (2402.01613, TMLR) achieves cloud-parity embeddings locally — Rune's local embedding is evidence-backed, not convenience. BUT: no A/B evidence exists that small local models match cloud on consolidation-style summarization — that gap is the crux.
+
+**Verdicts:**
+- **ADOPT**: local embedding (nomic-embed-text via Ollama — verified parity with Ada-002/emb-3-small on MTEB 2210.07316)
+- **REJECT**: local LLM consolidation (summarize/merge/rewrite facts) with small models — silent quality loss, unmeasurable without ground truth
+- **EXPERIMENT**: small local models for MECHANICAL ops only (dedup clustering, tag/title extraction, tool-call classification — MobileLLM 2402.14905 ICML 2024 covers exactly this class)
+- **FUTURE**: quantized consolidation; revisit when an ≤8B A/B study on memory-consolidation quality exists
+
+**Evidence:** QLoRA 4-bit NF4 preserves task performance (2305.14314, NeurIPS 2023); quantization degradation predictable not chaotic (2103.13630); layer-pruning headroom (2403.17887, ICLR 2025). Asymmetry doctrine: consolidation is a rare background op — cloud for judgment-heavy, local for mechanical.
+
+**R0 reconciliation:** CONFIRMED 2 · REFINED 1 · OVERTURNED 0 · Sum = 3.
@@ -0,0 +1,14 @@
+# AD — Hardware / Resource Placement: R1 Domain Master
+## #405
+
+**Executive summary:** "Memory-resident vs disk-backed" is VACUOUS below ~10^5 entries. All ANN-placement literature (HNSW 1603.09320 TPAMI 2018; DiskANN 2105.09613; SPANN 2111.08566 — both NeurIPS 2021) is calibrated at 10^6–10^9 points. vstash (2604.15484): vector+FTS5 hybrid in a single SQLite file at 20.9ms median latency at 50,000 chunks — 54× Rune's count.
+
+**Rune arithmetic:** 924 × 768-dim float32 ≈ 2.7MB. Brute-force cosine is sub-millisecond. SQLite on NVMe + OS page cache = effectively memory-resident (sqlite.org/mmap.html); mmap optional copy-avoidance with documented failure modes.
+
+**Verdicts:**
+- **ADOPT**: whole-store memory-residency as a side effect (SQLite on NVMe + page cache; optional mmap_size)
+- **REJECT**: DiskANN/SPANN-style disk-backed ANN (design point 10^9, not 10^3) and any dedicated vector database
+- **REJECT**: Pi placement for memory operations — no model capacity, no measurable benefit; current split (PC: Gitea/CI/Ollama; Pi: TLS/backup) matches the doctrine that placement follows capability
+- **FUTURE**: revisit only if the store crosses ~10^5 entries (vstash shows the SQLite stack holds well past that)
+
+**R0 reconciliation:** CONFIRMED 1 · REFINED 2 · OVERTURNED 0 · Sum = 3.
@@ -0,0 +1,17 @@
+# AE — Memory Evaluation / Benchmark Science: R1 Domain Master
+## #406
+
+**Executive summary:** R0's LoCoMo critique CORROBORATED (6.4% wrong answer key; ~63–64% judge wrong-accept — directional, via secondary reporting; full-context baseline ~73% J beats Mem0's best ~68% per Zep's released eval code). New failure mode beyond R0: scoring-target noninvariance (2605.24060) flips benchmark conclusions on 83–94% of shared LoCoMo/LongMemEval queries depending on which derived form gets retrieval credit.
+
+**Minimal viable evaluation for a single-user store with no ground truth:**
+1. Hand-curated ~50–100 question probe set over Rune's own history (abstention questions verifiable with zero ground truth) — LongMemEval 2410.10813 ICLR 2025 taxonomy: temporal, update, abstention
+2. Ledger-derived end-task outcome metrics — τ-bench pattern (2406.12045): compare final DB state vs goal state; pass^k reliability
+3. Hybrid-retrieval disagreement as continuous label-free health signal (vstash: 74.5% of queries produce top-10 disagreement between lexical and dense lanes)
+
+**Verdicts:**
+- **REJECT**: running LoCoMo/MemBench (wrong population: multi-persona synthetic chat, not coding traces; known scoring pathologies)
+- **ADAPT**: LongMemEval question taxonomy; judge use only with FairEval position calibration (2305.17926: ordering alone flips 66/80 verdicts; MT-Bench 2306.05685: ~80% agreement with position/verbosity bias)
+- **ADOPT**: the three-part minimal eval above
+- **EXPERIMENT**: pass^k reliability across repeated sessions
+
+**R0 reconciliation:** CONFIRMED 2 · REFINED 2 · OVERTURNED 0 · Sum = 4.
@@ -0,0 +1,68 @@
+# #377 R1 — 00_issue_scope.md
+
+## Core question
+
+What distinctions and mechanisms established by human memory and cognitive-science
+research are genuinely useful for Rune's Memory architecture, and which should be
+rejected as biological detail, misleading metaphor, or human failure mode?
+
+## Investigative scope (30 areas per mandate)
+
+1. episodic memory; 2. semantic memory; 3. autobiographical memory;
+4. prospective memory; 5. working memory; 6. procedural memory;
+7. source memory; 8. familiarity vs recollection; 9. cue-dependent recall;
+10. context-dependent recall; 11. temporal memory; 12. prospective
+obligations/reminders; 13. encoding vs storage vs retrieval failure;
+14. consolidation; 15. reconsolidation/update; 16. interference;
+17. forgetting; 18. retrieval-induced forgetting; 19. spacing effects;
+20. testing/retrieval-practice effects; 21. false memories;
+22. misinformation effects; 23. source-monitoring errors;
+24. confidence/metamemory boundaries; 25. salience/emotion (mechanics only);
+26. memory transformation over time; 27. schema effects (delta only vs #376);
+28. distortions from repeated summarization/reconstruction;
+29. prospective cues and event-triggered remembering;
+30. unavailable vs inaccessible vs forgotten vs overwritten.
+
+## R0 hypotheses under re-evaluation (NOT accepted)
+
+H1: prospective memory → typed obligation rows with trigger field.
+H2: episodic vs semantic → entry-kind distinctions.
+H3: autobiographical → session/task narrative records.
+H4: source memory → mandatory origin fields.
+H5: recollection/familiarity → exact/lexical vs semantic retrieval analogy.
+H6: cue/context recall → task/session scoping.
+H7: reconsolidation → supersession analogue (ALREADY OVERTURNED by #376 R1:
+     reconsolidation framing rejected; prediction-error-gated updating kept).
+H8: spacing → reactivation-driven consolidation priority.
+H9: confidence → evidence-linked rather than self-reported.
+H10: RIF → reject imitation.
+H11: false-memory research → anti-pattern/design constraint.
+H12: source-monitoring errors → provenance requirement.
+
+H7 incorporates #376's correction: the reconsolidation→supersession mapping
+is dead. This tranche must not reintroduce it; the prediction-error-gate atom
+survives independently (Sevenster/Beckers/Kindt 2014 via #376).
+
+## Falsifiers
+
+F1: If episodic/semantic/autobiographical are descriptive human taxonomies
+rather than computationally distinct processes, Rune must not materialize
+them as separate stores.
+F2: If prospective-memory findings primarily concern biological attentional
+limitations Rune does not share, they do not by themselves justify an
+"obligation memory" subsystem.
+F3: If familiarity/recollection evidence does not transfer cleanly to
+lexical/vector retrieval, reject that analogy.
+F4: If spacing/testing effects depend strongly on human neural learning
+rather than information availability, do not use them as consolidation policy.
+F5: If human forgetting mechanisms mostly exist because biology is
+resource-limited in ways Rune is not, Rune should solve overload with
+indexing, cooling and paging instead of imitating forgetting.
+F6: If human metamemory confidence is systematically unreliable, do not map
+human confidence mechanisms directly to Rune trust.
+
+## Stopping criterion
+
+Evidence saturation: new sources repeat known mechanisms/criticisms, add
+only detail. No arbitrary paper-count target — but a few paragraphs and a
+handful of citations are NOT sufficient for a foundational field.
@@ -0,0 +1,85 @@
+# #377 R1 — 01_inward.md
+
+## Current Rune mechanisms mapped to human-memory questions (evidence-first)
+
+_Source: direct read-only inspection of memory.db (924 entries) + T0/T0.1
+archaeology. Status vocabulary: PROVEN / IMPLEMENTED_UNPROVEN /
+DESIGNED / DORMANT / ABSENT / BROKEN._
+
+### Canonical entries and kinds
+- kinds present: `fact` (729), `neuron` (195). NO episodic/semantic/
+  autobiographical/procedural/prospective kinds exist. Memory_type column
+  exists but R0 found only `associative` populated.
+- Status: PROVEN (counts), but kind coverage is NARROW — H2 (entry-kind
+  distinctions) would be NEW schema, not a rename.
+
+### Facts vs neurons
+- Neurons (195) = warm-neuron activation rows with wash_count history;
+  facts (729) = durable content. 304 entries have wash_count>0.
+- Status: PROVEN. Use-history exists but T0.1 found it unvalidated as
+  quality signal (superseded entries carry high wash counts).
+
+### Retrieval lanes
+- Dense: 420/924 entries have vectors (sqlite-vec, 256-d). FTS5/BM25:
+  560 rows. Hybrid recall dense+BM25+RRF+floor: PROVEN (T0).
+- T0.1 R3: 439/838 active entries retrievable by NEITHER index.
+- Recollection/familiarity analogy (H5) would map to exact/lexical vs
+  vector lanes — lanes exist (PROVEN), the ANALOGY is untested.
+
+### Temporal fields
+- valid_from non-null on all 924; valid_until on 56 (supersession-derived).
+- Status: IMPLEMENTED_UNPROVEN for tri-timestamp semantics — fields exist,
+  retention policy and record-time query path unproven (T0 AS-IS).
+
+### Provenance/source/method
+- Columns: `source`, `method` (values: curator 679, dedupe_policy 165,
+  directive 34, migrate_blocks 22, warm_router 6, record 5). NO `origin`
+  column — R0's "mandatory origin field" (H4) is DESIGNED, not present.
+- `derived_from` edge: ABSENT (#376 R1 finding).
+- Confidence column exists (IMPLEMENTED_UNPROVEN — calibration unknown).
+
+### Trust
+- NO `trust_level` column — trust lifecycle (provisional/validated/locked)
+  lives in governance scripts, not as entry state. UNKNOWN state: ABSENT.
+- H9 (evidence-linked confidence): DESIGNED.
+
+### Supersession
+- 56 superseded rows retained; valid_until set. PROVEN. Write-triggered
+  (not retrieval-triggered) — #376 R1 overturned the reconsolidation
+  mapping; prediction-error gate atom survives.
+
+### Obligations/task systems
+- No obligation entry kind. Task system (auto_tasks.db, 2 tasks) and
+  PendingOperation machinery exist SEPARATELY from Memory.
+- H1 (obligation rows in Memory) would be NEW — the boundary question
+  (§10) is live: obligation-as-Memory vs Memory-as-evidence-for-tasks.
+
+### Persona/relationship memory
+- No dedicated persona/relationship kinds. Persona lives in constitution
+  files; relationship memory is implicit in facts.
+- Status: ABSENT as memory kinds.
+
+### Consolidation and failure learning
+- Consolidation → knowledge.json: PROVEN to run; trigger non-deterministic
+  (agent-invoked; no cron — T0.1 R5).
+- promotion_eligible flag exists; wash_count feeds it (unvalidated).
+- knowledge.json: derived artifact, source-attributed.
+
+### Warm Neurons
+- 195 neuron rows; tier='warm' on ALL 924 rows = vestigial (T0.1 R4).
+- Wash/use history PROVEN; activation ≠ truth doctrine stated but the
+  scoring (wash_count+importance) is frequency-only (NFU-like per #376/#385).
+
+## Where R0 renamed software without proving transfer
+1. "Episodic vs semantic → entry kinds": Rune has NO such kinds; calling
+   facts-vs-chronicles "episodic/semantic" would rename an ABSENT
+   distinction. Transfer must be EARNED, not asserted.
+2. "Recollection/familiarity → exact/vector": lanes exist, but dual-process
+   theory adds nothing until a retrieval-routing decision cites it.
+3. "Cue/context recall → task scoping": scoping exists (session/task keys);
+   Godden & Baddeley add no mechanism.
+4. "Spacing → reactivation priority": wash_count already counts
+   reactivation; Cepeda adds nothing unless a spacing SCHEDULE is proposed.
+5. "Prospective memory → obligation rows": the boundary question is
+   unresolved; R0 assumed the answer (Memory) without arguing against
+   task-system ownership.
@@ -0,0 +1,9 @@
+# 02 — OUTWARD (populated by deep-research actor; verified by controller)
+
+Files: taxonomy_of_memory.md, episodic_semantic_memory.md,
+autobiographical_memory.md, prospective_memory.md, source_memory.md,
+working_memory.md, procedural_memory.md, cue_context_retrieval.md,
+temporal_memory.md, forgetting_interference.md, false_memory_source_errors.md,
+spacing_testing_learning.md, consolidation_retrieval_update.md,
+competing_models.md, criticisms_replication.md, software_transfer.md
+ sources.md (ledger) + claims.json (machine-readable).
@@ -0,0 +1,6 @@
+# Autobiographical memory
+
+Conway & Pleydell-Pearce (2000) SMS: generative construction yoked to identity goals. Adversarial #6 SUCCEEDS: this is the confabulation engine a truth-preserving store must NOT imitate (identity-flattering reconstruction, telescoping, goal-consistent distortion).
+
+Rune use: session/task narratives exist (COVERED as chronicles); any "life-story synthesis" must be fenced from the evidence store and labeled generated. Take only a warning label.
+
@@ -0,0 +1,8 @@
+# Competing models / disputes
+
+- Systems vs processes: Foster & Jelicic 1999; Roediger/Buckner/McDermott 1999-style process critiques; Schacter & Tulving 1994. Psychologists' own failure to agree on system count = evidence software should not hard-code one taxonomy (adversarial #8).
+- Unitary vs dual-process recognition: Wixted 2007 / Dunn 2004 / Wixted & Mickes 2010 vs Yonelinas 1994. Live dispute; software mapping built on it is sand.
+- Working memory: Baddeley boxes vs Cowan activation vs Engle attention. Active contest.
+- Inhibition vs interference (RIF mechanism); storage-loss vs retrieval-failure; reconstructive (Bartlett) vs trace views; modularity (Squire) vs distributed processing.
+- Functional accounts: Nairne survival-processing; Henke 2010 processing-modes — dissolve store boundaries into processes/functions.
+
@@ -0,0 +1,6 @@
+# Consolidation / retrieval / updating (delta vs #376)
+
+No deep duplication of #376 CLS per mandate. #376 R1 already: overturned reconsolidation→supersession mapping; kept prediction-error-gated updating; confirmed retained covering episodes (rationale = drift correction); rejected destructive consolidation.
+
+Human-memory additions from this tranche: Fuzzy Trace verbatim/gist opponent processes (Brainerd & Reyna) — store both, expect gist to outlive detail; Bartlett schema-assimilation across retellings — minimize re-encoding retrieved outputs; reconsolidation boundary conditions (Sevenster prediction-error gate, via #376) — update-on-violation, not on retrieval.
+
@@ -0,0 +1,9 @@
+# Criticisms / replication / negative evidence
+
+- RIF: small effect + publication bias (Murayama 2014); reliability weak (Potts 2012; Eur Psychologist 2011; Frontiers 2014); inhibition signature often fails (Camp 2007).
+- Misinformation boundaries: McCloskey & Zaragoza 1985; recent high-powered boundary work.
+- Godden & Baddeley 1975: N=18, small-N legacy; Smith & Vela 2001: small/fragile overall.
+- DRM ecological validity (Pezdek & Lam 2007); Shaw & Porter 2015 rate disputes.
+- WEIRD populations; lab-vs-naturalistic prospective-memory gaps; measurement confounds (remember/know, ROC procedures).
+- Consolidation-hallucination links contested/mixed (Darsaud; Calvillo 2016).
+
@@ -0,0 +1,12 @@
+# Retrieval: cues, context, recognition
+
+**Encoding specificity: Tulving & Thomson (1973), Psych Review 80(5) (~4690 cites).** Retrieval succeeds to the degree cues reinstate encoding conditions. ROBUST principle, boundaries debated. Bearing: store encoding context alongside content; match on cues present at write.
+
+**Availability vs accessibility: Tulving & Pearlstone (1966, JVLVB).** Unrecallable-under-one-cue items become available under another — failure is often accessibility, not erasure. ROBUST; designated THE key distinction. Bearing: default to retrieval-failure hypothesis before deletion; forgetting policy gates on demonstrated inaccessibility across cues, never single failed recall.
+
+**Context-dependence: Godden & Baddeley (1975, BJP 66:325-331)** — divers, N=18, WEAK-evidence legacy. **Smith & Vela (2001)** synthesis: overall effect small/fragile; reinstatement helps mainly with impoverished cues. CONTESTED/small-effect. State-dependence similarly fragile. Bearing: context tags help; expect small gains; never rely on context match alone.
+
+**Recognition vs recall; recollection vs familiarity:** Yonelinas (1994) dual-process vs Wixted unitary-strength (2007; Wixted & Mickes 2010). CONTESTED — both fit much data; measurement confounds. Adversarial #4 SUCCEEDS: do not reify two hardware systems; keep two retrieval GRADES (detail-bearing vs bare signal) as outputs; track confidence separately. Bearing: the lexical/vector mapping inherits a live dispute — justify hybrid retrieval by IR engineering only.
+
+**Retrieval competition, temporal order, partial cues:** cue-overload competition ROBUST; temporal-order memory error-prone/reconstructive; partial cues help only when encoding-compatible (encoding-specificity corollary).
+
@@ -0,0 +1,6 @@
+# Episodic / semantic memory
+
+See taxonomy (Tulving 1972/1985/2002). Status: ROBUST descriptive; CONTESTED as strict neural systems (adversarial #1: Henke 2010 processing-modes; systems-vs-processes debate; Vargha-Khadem 1997 amnesic dissociations preserve a weak distinction).
+
+Bearing for Rune: strongest candidate for two store-types (event records with source/context vs distilled durable claims) — but as QUERY-TIME TAGS, never schema partition (adversarial verdict). F1 falsifier stands: descriptive taxonomy ≠ computational process split.
+
@@ -0,0 +1,18 @@
+# False memory / source errors
+
+**DRM: Roediger & McDermott (1995, JEP:LMC 21(4):803-814; Deese 1959).** Lure recalled ~40-55%, recognized at studied-item rates, ~72% "remember" judgments. ROBUST, heavily replicated; ecological-validity critiques (Pezdek & Lam 2007). Bearing: associative retrieval generates confident false positives — penalize gist-only matches lacking verbatim support.
+
+**Fuzzy Trace Theory: Brainerd & Reyna (2002, Curr Dir Psych Sci 11(5):164-169; 2005 OUP book).** Parallel verbatim (decays fast) + gist (persists) traces; false memory = gist-supported + verbatim-unsuppressed; signature predictions: delay increases false memory while true decays; developmental reversal (older show MORE DRM false memory). ROBUST alternative to unitary-trace views. Bearing: store BOTH verbatim detail and gist; expect gist to outlive detail; gist-vs-verbatim contradiction should trigger verification, not averaging.
+
+**Reconstruction: Bartlett (1932, Remembering).** Repeated reproduction rationalizes toward schemas. Foundational; direct replication difficult (Bergman & Roediger 1999 partial), schema-consistent distortion ROBUST via successors. Bearing: each retrieval-rewrite cycle risks schema assimilation — minimize re-encoding retrieved outputs as new entries without provenance.
+
+**Rich false memories: Shaw & Porter (2015, Psych Sci 26:291-301)** — whole-event implantation possible with suggestive pressure (~70%, coding-dependent). CONTESTED rates, ROBUST possibility.
+
+**Misinformation: Loftus & Palmer (1974, JVLVB 13(5):585-589); Loftus, Miller & Burns (1978).** Post-event narrative blends with/overwrites original; grows with delay, lowers with discrepancy detection. ROBUST core; CONTESTED mechanism (McCloskey & Zaragoza 1985: modified tests reduce apparent impairment). Bearing: post-event inputs (corrections, restatements, tool outputs) can corrupt originals — keep original vs amendment separated with source tags.
+
+**Confidence-accuracy: Wixted & Wells (2017) eyewitness family; Koriat metamemory.** Confidence malleable (feedback, repetition), often miscalibrated. ROBUST dissociation. Bearing: never confidence-as-accuracy-proxy; track calibration separately.
+
+**Cryptomnesia/reality monitoring/false fame (Jacoby et al. 1989, JPSP):** ROBUST demonstrations of origin confusion. Bearing: generation events labeled self-generated at write time.
+
+All converge on: append-only + supersession-linking beats destructive consolidation (adversarial #10 FAILED as attack — Maynez 2020 + SSGM 2026 corroborate).
+
@@ -0,0 +1,8 @@
+# Forgetting / interference
+
+Interference (proactive/retroactive) ROBUST since Underwood/McGeoch; pure trace-decay WEAK as sole account — most forgetting = cue-dependent inaccessibility + competition. Directed forgetting ROBUST as control phenomenon. Adaptive/intentional forgetting (Anderson & Hanslmayr) CONTESTED as unified mechanism.
+
+**RIF: Anderson, Bjork & Bjork (1994, JEP:LMC 20(5):1063-1087).** ~10-13% effect; inhibition account. Status CONTESTED — real but fragile: Murayama et al. (2014, Psych Bulletin) meta-analysis small effect + publication bias; Raaijmakers/Verde interference alternatives; Potts et al. (2012) reliability weak; cue-independence (inhibition signature) often fails (Camp et al. 2007).
+
+Bearing: do NOT implement inhibition-as-deletion — retrieval practice suppressing neighbors is exactly what a memory system must avoid; treat RIF as hazard (retrieval narrows access), not feature. Adversarial #7 SUCCEEDS: keep tiered storage; ban "forgetting" vocabulary; the one transferable rule = never delete, only de-index (availability vs accessibility).
+
@@ -0,0 +1,4 @@
+# Procedural memory
+
+Squire declarative/nondeclarative split (ROBUST coarse). "How I do X" ≠ "what happened" — supports separating procedure traces from statable entries. NOT_RELEVANT as memory-store mechanism (skill runtime concern); relevant as taxonomy boundary.
+
@@ -0,0 +1,12 @@
+# Prospective memory
+
+**Einstein & McDaniel (1990, JEP:LMC 16:717-726).** Event vs time-based; external event cues beat self-initiated time monitoring; age deficits larger for time tasks. **Sellen et al. (1997, Memory 4:483-507)** in-situ confirmation. ROBUST.
+
+**Monitoring costs — PAM vs multiprocess:** Smith (2003) PAM (capacity-consuming preparatory monitoring; ongoing-task slowing) vs McDaniel & Einstein (2000 Applied Cog Psych; 2005) multiprocess (monitoring + spontaneous retrieval; salient/focal cues pop without cost); Reese & Cherry (2002) against constant monitoring (~2% reported thinking of intention). CONTESTED coexistence; pure-PAM rejected for mundane tasks.
+
+**Interruption/resumption: Altmann & Trafton (2002)** goal-activation model: interruptions decay activation; resumption lag grows with length; reminders restore. **McDaniel et al. (2004, Applied Cog Psych 18:533-547):** delays degrade performance; external reminders eliminate interruption cost. ROBUST practical findings.
+
+**Implementation intentions: Gollwitzer (1999, American Psychologist 54:493-503).** If-then plans substantially raise follow-through. ROBUST, large literature.
+
+Bearing: prefer event-anchored triggers over pure time; time intentions need external scaffolding; support both spontaneous firing (salient cues) AND costly watch (non-focal); store intentions as cue→action pairs with explicit triggers + reminder scaffolding. BUT adversarial #2 (Risko & Gilbert 2016 offloading) says obligations belong OUTSIDE memory in a commitment authority — the obligation-entry-kind question is NOT settled by this literature (see 03_applicability).
+
@@ -0,0 +1,4 @@
+# Software transfer (placeholder — full chain in 03_applicability)
+
+Transfer rule: HUMAN FINDING → GENERAL COMPUTATIONAL PRINCIPLE → SOFTWARE ANALOGUE → RUNE APPLICABILITY → RUNE PRESCRIPTION. If one arrow cannot be justified, stop the chain. Full treatment in 03_applicability.md; key results: availability/accessibility → retrieval-failure-before-deletion policy; encoding specificity → store encoding context; Fuzzy Trace verbatim/gist → dual representation with verification on conflict; offloading literature → obligations OUTSIDE memory; testing effect → reactivation-as-durability (not ranking).
+
@@ -0,0 +1,6 @@
+# Source memory / monitoring
+
+**Johnson, Hashtroudi & Lindsay (1993, Psych Bulletin 114(1):3-28).** Source attributions = decisions over features (perceptual, contextual, affective, cognitive-operations records), heuristic (fast) vs systematic (slow) judgments. External/internal/reality monitoring types. ROBUST/influential framework.
+
+Bearing: every entry needs source features recorded at write; source judgments at READ are fallible decisions — provenance must be stored, not inferred. Adversarial #3: mechanism analogy backwards (Rune already has columns), but Johnson's cue inventory is a useful schema checklist; errors spike under similarity/time-pressure → predicts when retrieval needs provenance-weighted ranking.
+
@@ -0,0 +1,8 @@
+# Spacing / testing / reactivation
+
+**Spacing: Cepeda et al. (2006, Psych Bulletin 132(3):354-380).** 839 assessments: spaced > massed in 259/271. Lag effects (Glenberg; Kahana); Bahrick et al. (1993) 9-year retention (56-day gaps optimal there). ROBUST — among psychology's most replicated. Mechanism accounts (encoding variability, study-phase retrieval, deficient processing) CONTESTED; multifactor prevails.
+
+**Testing: Roediger & Karpicke (2006, Psych Sci 17(3):249-255); Karpicke & Roediger (2008, Science 319:966-968).** Repeated testing beats restudying on delayed retention; Butler (2010) superior transfer. Dunlosky et al. (2013) rates practice testing + distributed practice highest-utility. ROBUST. **Desirable difficulties: Bjork (1994) family** — effortful retrieval aids retention.
+
+Bearing (adversarial #5 verdict): re-exposure scheduling is the one human effect safe to import as POLICY — but frequency ≠ spacing; the wash_count question resolves toward SPACED re-presentation, not raw repetition count. Allow reactivation into retention/tiering decisions; FORBID (or heavily discount) in relevance ranking. Spacing is acquisition-side; ranking is a separate judgment.
+
@@ -0,0 +1,10 @@
+# Taxonomy of memory (human literature)
+
+**Tulving (1972)** episodic vs semantic (Organization of Memory; K.C. amnesic dissociations). Later: Tulving 1985 autonoetic/noetic; Tulving 2002 mental time travel. Status: ROBUST as descriptive distinction; CONTESTED as strict neural-systems claim.
+
+**Squire (1992 Psych Review; Squire & Zola 1996 PNAS).** Declarative (hippocampal-dependent) vs nondeclarative (skills/habits/priming). ROBUST at coarse grain. Bearing: separate skill/procedure traces from statable entries.
+
+**Conway & Pleydell-Pearce (2000, Psych Review 107(2):261-275).** Self-Memory System: autobiographical memory hierarchical (lifetime periods → general events → event-specific knowledge), reconstructed under working-self control; generative, not replay. ROBUST/influential. Bearing: warns against treating recalled episodes as verbatim logs.
+
+**Baddeley & Hitch (1974); Baddeley (2000, TiCS 4(11), episodic buffer).** Dual-task interference basis. ROBUST framework, revised often. Competitor: **Cowan (1999)** embedded-processes (activated LTM + ~4-chunk focus); Engle executive-attention third pole. Active CONTEST. Bearing: do NOT hard-imitate slot buffers; transferable = capacity-limited active set + executive control (~4-chunk focus).
+
@@ -0,0 +1,4 @@
+# Temporal memory
+
+Tri-timestamps already adopted from Graphiti (prior closure; COVERED). This literature adds: temporal-order memory is reconstructive and error-prone — "as-of" queries must trust stored timestamps, never reconstructed order. No new mechanism beyond prior closure; cross-reference #376-adjacent schema work minimally per mandate (no deep duplication).
+
@@ -0,0 +1,4 @@
+# Working memory
+
+Baddeley & Hitch 1974 / Baddeley 2000 (above) vs Cowan 1999 vs Engle. Active contest; do not hard-imitate boxes. Transferable: capacity-limited active set + executive control. NOT_RELEVANT as store concern (context window / skill runtime).
+
@@ -0,0 +1,109 @@
+# #377 R1 — 03_applicability.md
+
+_For every surviving mechanism: the 19 applicability questions + the transfer
+chain HUMAN FINDING → GENERAL PRINCIPLE → SOFTWARE ANALOGUE → RUNE
+APPLICABILITY → RUNE PRESCRIPTION. Chain breaks are recorded, not smoothed._
+
+## T1. Availability/accessibility distinction (Tulving & Pearlstone 1966)
+- Human finding: unrecallable-under-one-cue items become available under
+  another — failure is often accessibility, not erasure.
+- Principle: retrieval failure ≠ storage loss.
+- Analogue: multi-cue retrieval + deletion gated on demonstrated
+  inaccessibility across cues.
+- Rune: PROVEN problem (439/838 active entries unindexed — but those are
+  AVAILABILITY failures, worse). Prescription: default to retrieval-failure
+  hypothesis; forgetting/deletion policy requires demonstrated
+  inaccessibility across cues, never single failed recall.
+- Owner: retrieval layer + deletion policy. LLM: none. Offline: yes.
+  Century: yes. Cheapest: deletion-gate rule. EXTERNALLY_ESTABLISHED → RUNE_APPLICABLE.
+
+## T2. Encoding specificity (Tulving & Thomson 1973)
+- Finding: retrieval succeeds to the degree cues reinstate encoding conditions.
+- Principle: store the conditions, not just the content.
+- Analogue: encoding-context metadata on entries (session/task/partition keys
+  already exist; extend to query-relevant cue fields).
+- Rune: PARTIALLY solved (scoping exists). Prescription: extend stored context
+  fields; retrieval matches on write-present cues. Owner: retrieval layer.
+
+## T3. Fuzzy Trace verbatim/gist (Brainerd & Reyna 2002/2005)
+- Finding: parallel traces; gist persists, verbatim decays; false memory is
+  gist-supported + verbatim-unsuppressed.
+- Principle: keep both representations; contradiction between them triggers
+  verification, not averaging.
+- Analogue: store exact source text + distilled gist side by side (Rune
+  already stores text; gist = summary field).
+- Rune prescription: verbatim/gist dual representation with
+  conflict-triggered verification. Owner: Memory. Cheap: schema fields.
+  EXTERNALLY_ESTABLISHED → RUNE_APPLICABLE. (Chain holds end to end.)
+
+## T4. Source-monitoring cue inventory (Johnson 1993) — schema only
+- Per adversarial #3: mechanism backwards, inventory useful. Prescription:
+  ensure origin/source/method/verbatim-detail/operation-record fields exist
+  (partially present; origin column ABSENT → add). Owner: Memory schema.
+
+## T5. Prospective cues → task authority, NOT memory (adversarial #2 wins)
+- Human finding: event cues beat time monitoring; monitoring costs are real;
+  implementation intentions (cue→action pairs) work.
+- Principle: intentions need triggers + scaffolding.
+- Analogue: commitment state machine with trigger conditions.
+- Rune applicability: the OWNER is the task/commitment authority, NOT Memory
+  (Risko & Gilbert 2016 offloading: externalize intentions). R0's H1
+  (obligation rows IN Memory) is OVERTURNED as stated.
+- Prescription: obligations live in task system with open/kept/broken states;
+  Memory holds evidence ABOUT commitments. Event-triggered remembering reuses
+  PendingOperation machinery where applicable. Owner: task system.
+  EXTERNALLY_ESTABLISHED → RUNE_APPLICABLE (with corrected owner).
+
+## T6. Testing effect → durability signal only (adversarial #5 verdict)
+- Finding: retrieval strengthens the trace (Roediger & Karpicke 2006).
+- Principle: reactivation = durability event.
+- Analogue: reactivation count feeds retention/tiering.
+- Rune: allowed into retention decisions; FORBIDDEN in relevance ranking
+  (category jump: learning schedule ≠ relevance score; rich-get-richer).
+- Owner: retention policy. Cheap: spaced-reactivation counter distinct from
+  wash_count.
+
+## T7. Misinformation/DRM → anti-pattern constraints
+- Findings: post-event blending (Loftus), gist false positives (DRM 40-55%
+  recall, 72% "remember"), schema assimilation across retellings (Bartlett).
+- Principle: every transformation is a corruption opportunity.
+- Analogues: consolidate from immutable sources only; original-vs-amendment
+  separation; penalize gist-only retrieval matches; minimize re-encoding
+  retrieved outputs.
+- Rune: largely COVERED (append-only, supersession); ADD: gist-match penalty
+  in retrieval scoring; re-encoding guard. Owner: Memory + retrieval.
+
+## T8. RIF → hazard, never feature
+- Finding: retrieval practice suppresses competitors (~10-13%, fragile).
+- Rune: retrieval must not narrow access to neighbors. Prescription: no
+  inhibition mechanism; monitor for retrieval-narrowing as a bug class.
+- Owner: retrieval layer. REJECT imitation (adversarial #7 supports).
+
+## T9. Confidence-accuracy dissociation → calibration discipline
+- Finding: confidence malleable, miscalibrated (Wixted & Wells; Koriat).
+- Principle: never confidence-as-accuracy.
+- Analogue: evidence-linked confidence + separate calibration tracking.
+- Rune: confidence column exists (IMPLEMENTED_UNPROVEN). Prescription:
+  calibration tracked per source/kind; self-reported confidence distrusted.
+- Owner: Memory schema + governance.
+
+## T10. Context-dependence → small-gain tags (expectations capped)
+- Finding: real but small/fragile (Smith & Vela 2001); N=18 legacy.
+- Prescription: keep context tags; never rely on context match alone;
+  measure gains before investing. Owner: retrieval.
+
+## T11. Recognition grades without dual-process ontology
+- Per adversarial #4: keep two retrieval outputs (detail-bearing vs bare
+  signal) + separate confidence; do NOT cite Yonelinas as architecture.
+- Owner: retrieval layer. Justification: IR engineering (child H), not
+  psychology.
+
+## T12. Autobiographical generativity → fence + label
+- Finding: SMS construction yoked to identity goals (Conway 2000).
+- Prescription: any life-story synthesis fenced from evidence store,
+  labeled generated. Owner: Persona/presentation layer, never canonical Memory.
+
+## Authority assignments (Q19)
+Memory owns: T1, T2, T3, T4, T6, T7, T8, T9, T10, T11. Task system owns: T5.
+Persona/presentation owns: T12. Retrieval layer implements: T1, T2, T8, T10, T11.
+No mechanism here belongs to Substrate, World Model, or Spine.
@@ -0,0 +1,57 @@
+# #377 R1 — 04_innovation.md
+
+_Falsifiable integration hypotheses only. Each: prior art, difference,
+advantage, cost, failure mode, benchmark, kill criterion._
+
+## I1. Availability-gated deletion (prior art: HSM tiering policies)
+- Difference: deletion requires demonstrated inaccessibility across N
+  independent cue sets + time window, not age or single failed recall.
+- Advantage: eliminates the worst deletion class (available-but-inaccessible).
+- Cost: bookkeeping per deletion candidate. Failure: gate too strict →
+  unbounded growth (mitigate: tiering absorbs).
+- Benchmark: zero wrongful deletions on synthetic available-but-inaccessible
+  probes. Kill: gate never fires in 6 months → simplify.
+
+## I2. Verbatim/gist dual entries with conflict verification (prior art:
+Fuzzy Trace Theory; Rune text+summary fields)
+- Difference: explicit conflict detector — when gist contradicts verbatim
+  detail, verification fires instead of averaging.
+- Advantage: catches consolidation drift at the representation level.
+- Cost: dual storage + detector. Failure: detector false positives annoy.
+- Benchmark: drift-detection rate on synthetic gist/verbatim conflicts.
+- Kill: no conflicts detected in production over 6 months → keep schema,
+  drop detector.
+
+## I3. Spaced-reactivation durability counter (prior art: Cepeda spacing;
+Anki/SRS scheduling)
+- Difference: separates durability counting (spaced re-presentations) from
+  relevance ranking (topical fit) — two counters, two consumers.
+- Advantage: resolves the wash_count conflation T0.1 exposed.
+- Cost: one new counter + scheduler. Failure: scheduler complexity.
+- Benchmark: retention prediction accuracy vs wash_count.
+- Kill: no retention-prediction gain → drop.
+
+## I4. Gist-match penalty in retrieval (prior art: DRM literature; child H
+reranking)
+- Difference: retrieval scores down-weight matches supported only by
+  gist-level similarity without verbatim support, for factual queries.
+- Advantage: directly attacks the 40-55% DRM-style false-positive class.
+- Cost: scoring change. Failure: over-penalization hurts paraphrase recall
+  (tune on paraphrase-vs-lure test set).
+- Benchmark: lure-rejection rate vs paraphrase-recall rate.
+- Kill: paraphrase recall drops >5% → revert.
+
+## I5. Commitment-evidence linking (prior art: GTD trusted-system concept;
+Rune PendingOperation)
+- Difference: task-system commitments carry memory-evidence pointers
+  (supporting entries), and memory entries ABOUT commitments link back —
+  bidirectional without co-location (respects adversarial #2's ownership
+  verdict).
+- Advantage: obligations auditable against evidence without living in Memory.
+- Cost: link maintenance. Failure: stale links.
+- Benchmark: commitment audit completeness on synthetic set.
+- Kill: links unused in 6 months → drop.
+
+No manufactured novelty: I1 = HSM policy + Tulving distinction; I2 = FTT
+applied to schema; I3 = SRS practice applied to counters; I4 = DRM
+literature applied to scoring; I5 = GTD + existing Rune machinery.
@@ -0,0 +1,135 @@
+# #377 R1 — 05_adversarial.md
+
+_Independent pass (separate actor/context). Instruction: assume the emerging
+human-memory → Rune conclusions are wrong. Verdicts: ATTACK SUCCEEDS / MIXED /
+ATTACK FAILED. ✅ = tool-verified citations; ⚠️ = canonical references._
+
+## 1. Episodic/semantic store separation — MIXED, leans SUCCEEDS
+**Henke (2010, Nat Rev Neurosci)** ✅ — processing-modes (rapid-flexible vs
+slow-rigid encoding) should replace the consciousness-based declarative split.
+If right, an episodic/semantic table split reifies the wrong ontology: the
+joint is at processing mode, not phenomenology. Backing: systems-vs-processes
+debate (⚠️ Foster & Jelicic 1999; ⚠️ Roediger/Buckner/McDermott 1999-style
+process critiques); ⚠️ Cabeza et al. 2008 (parietal = attention-to-memory, not
+episodic module); ⚠️ Nairne/Thompson/Pandeirada 2007 (survival-processing
+functional account). Patient dissociations (⚠️ Vargha-Khadem 1997, Science:
+developmental amnesia sparing semantic learning) keep the distinction alive —
+demoted to "one contested taxonomy," not neural fact.
+**Software verdict: query-time tag, never schema partition.**
+
+## 2. Obligation-as-memory — ATTACK SUCCEEDS (strongest in brief)
+**Risko & Gilbert (2016, Trends Cogn Sci)** ✅ — cognitive offloading: agents
+perform better EXTERNALIZING intentions than retaining them; intention-offloading
+literature shows people set external triggers precisely because internal
+prospective memory is unreliable. ⚠️ Allen (2001, GTD): "mind like water" via
+capture into a trusted EXTERNAL system. Hostile inference: obligations are the
+canonical case where science says DO NOT store as memory traces — they belong
+in a commitment/task authority with state machines (open/kept/broken/
+superseded), deadlines, reminders; memory holds only evidence ABOUT the
+commitment. Putting obligations inside the memory store repeats the exact error
+the offloading literature diagnoses.
+**Verdict: obligations live outside Memory.**
+
+## 3. Source-memory analogies — MIXED, leans FAILED
+⚠️ Johnson/Hashtroudi/Lindsay 1993 (source-monitoring framework); ⚠️ Johnson &
+Raye 1981 (reality monitoring): a brain doing forensic reconstruction from
+impoverished cues because it has NO provenance bus. Rune already has
+source/method columns — importing "source-monitoring processes" brings a candle
+to a lit room. Honest admission: Johnson's diagnostic cue inventory (verbatim
+detail, operation records, affective tags, contextual embedding) is a useful
+SCHEMA checklist, and source errors spiking under similarity/time-pressure
+predicts when retrieval needs provenance-weighted ranking.
+**Verdict: analogy backwards as mechanism, useful as schema checklist.
+Downgrade "implement source monitoring" → "store what source monitoring
+wishes it had."**
+
+## 4. Recollection/familiarity → lexical/vector mapping — ATTACK SUCCEEDS
+Three verified kill-shots: **Wixted (2007, Psych Review)** ✅ — unitary
+unequal-variance signal-detection model fits the ROC data dual-process
+theorists claimed exclusively. **Dunn (2004, Psych Review)** ✅ —
+remember/know dissociations fall out of unidimensional strength + decision
+criteria. **Wixted & Mickes (2010, Psych Review)** ✅ — "continuous
+dual-process" concedes recollection is graded, collapsing the categorical
+joint the mapping leans on. (⚠️ Rotello et al. 2005 same point via
+process-dissociation critiques.) Mapping BM25→familiarity/vectors→recollection
+inherits a live dispute as settled architecture.
+**Verdict: keep hybrid retrieval (it works), justify by IR engineering, never
+by dual-process psychology. The mapping adds zero architectural force.**
+
+## 5. Spacing/reactivation → activation scoring — MIXED, leans SUCCEEDS
+⚠️ Cepeda et al. 2006 (Psych Bulletin): spacing framed as ACQUISITION/encoding
+phenomenon (optimal inter-study intervals as function of retention interval).
+Nothing licenses "reactivated often → rank higher at retrieval" — a category
+jump from learning schedule to relevance score, smuggling frequency-past-use
+into relevance where it swamps topical fit (rich-get-richer). Admission:
+⚠️ Roediger & Karpicke 2006 (testing effect) proves retrieval IS an encoding
+event — reactivation genuinely strengthens the trace.
+**Verdict: reactivation allowed into retention/tiering decisions, FORBIDDEN
+(or heavily discounted) in relevance ranking.**
+
+## 6. Autobiographical-memory requirements — ATTACK SUCCEEDS
+**Conway & Pleydell-Pearce (2000, Psych Review)** ✅ — autobiographical recall
+is GENERATIVE CONSTRUCTION yoked to the "working self" and identity goals, not
+playback. ⚠️ Conway 2005: coherence with self-goals constrains construction.
+That is precisely the confabulation engine a truth-preserving store must NOT
+imitate — importing SMS requirements invites identity-flattering
+reconstruction, telescoping, goal-consistent distortion dressed as "rich
+autobiographical context."
+**Verdict: dangerous as design input. Take only a warning label: any
+"life-story synthesis" feature must be fenced off from the evidence store and
+labeled generated.**
+
+## 7. Human forgetting → software cooling — ATTACK SUCCEEDS (one caveat)
+Storage engineering solved tiering decades ago without psychology (HSM,
+LRU/LFU/ARC, buffer pools — hit-rate math, no Ebbinghaus). "Cooling" that
+merely demotes priority is eviction policy with a cardigan; calling it
+"forgetting" licenses the one thing a single-user truth store must never do —
+silent data loss — while adding no mechanism (no decay constant transfers to
+SQLite rows). Caveat: the availability-vs-accessibility distinction maps to a
+sound rule — never delete, only de-index.
+**Verdict: keep tiered storage; ban the word "forgetting" from the codebase
+vocabulary. The metaphor's expected value is negative.**
+
+## 8. Cognitive-science terminology adds architecture — ATTACK SUCCEEDS
+The field's own meta-literature: systems-vs-processes controversy plus "how
+many systems?" stalemates (⚠️ Schacter & Tulving 1994; ⚠️ Foster & Jelicic
+1999) — labels proliferating while predictions stay put. Adopting "episodic /
+semantic / procedural / source / familiarity" imports five contested ontologies
+and zero ranking functions, loss functions, or invariants.
+**Gate: every borrowed noun must answer — what query does it change, what
+invariant does it enforce, what test would falsify it? Failing that gate =
+documentation perfume. Terminology guilty until proven mechanistic.**
+
+## 9. Bonus: computer memory should NOT imitate humans — MIXED
+⚠️ Schacter 1999 (American Psychologist: "seven sins") — a spec sheet for what
+NOT to build. ⚠️ Norman 1986/1988 (gulfs of execution/evaluation) — bridge
+limits with external structure, don't mirror them. Database/IR orthodoxy never
+cites Atkinson–Shiffrin: durability, transactions, recall metrics come from
+engineering. Admission: nobody wants to imitate FALLIBILITY — borrowings are
+functional (separate concerns, provenance, tiering); biomimetic inspiration
+constrained by benchmarks is legitimate.
+**Verdict: "human memory is a terrible database" is true and worth posting on
+the wall — kills naive imitation, not disciplined functional borrowing.**
+
+## 10. Bonus: consolidation-hallucination — ATTACK FAILED
+Backfired into corroboration. **Maynez et al. 2020 (ACL)** ✅ — widespread
+hallucination in abstractive summarizers, direct analog of lossy LLM
+consolidation rounds. **SSGM (2026 preprint)** ✅ — semantic drift across
+evolving LLM memory updates as governed risk. Both support the FALSE-memory-
+as-anti-pattern stance: each summarization round is a misinformation
+opportunity; append-only + supersession-linking beats destructive
+consolidation.
+**Verdict: evidence sides with the main researcher.**
+
+## Failed-attack rollup
+- #3 partially failed: Johnson's cue inventory informs schema.
+- #5 partially failed: testing effect legitimizes reactivation as durability
+  signal (not ranking signal).
+- #9 partially failed: kills naive mimicry only.
+- #10 fully failed: Maynez 2020 + SSGM 2026 corroborate append-only/supersede.
+- #1 partially failed: amnesic dissociations preserve weak tag-level
+  episodic/semantic distinction.
+
+Strongest clean kills: #2 (obligations out of memory), #4 (dual-process
+mapping built on sand), #6 (autobiographical confabulation engine),
+#7 (forgetting vocabulary), #8 (terminology without mechanism).
@@ -0,0 +1,44 @@
+# #377 R1 — 06_synthesis.md
+
+## Mechanism matrix
+
+| MECHANISM | ORIGINAL PURPOSE | EVIDENCE STRENGTH | LIMITATIONS | RUNE EQUIVALENT | RUNE GAP | TRANSFER VALIDITY | COST | FAILURE MODES | CLASSIFICATION | RUNE EVIDENCE LEVEL |
+|---|---|---|---|---|---|---|---|---|---|---|
+| Episodic/semantic distinction | describe human memory kinds | STRONG descriptive; CONTESTED neural | Henke processing-modes; systems-vs-processes stalemate | none (no such kinds) | — | LOW as stores; MEDIUM as tags | none | reified wrong ontology | query-time TAGS, never schema partition | RUNE_HYPOTHESIS |
+| Availability/accessibility | explain recall failure | STRONG (Tulving & Pearlstone) | — | none (no deletion gate) | deletion-gate rule | HIGH | bookkeeping | gate too strict | ADAPT | RUNE_APPLICABLE |
+| Encoding specificity | explain cue effects | STRONG + boundaries | — | session/task scoping (partial) | cue-field extension | HIGH | schema fields | over-tagging | ADAPT | RUNE_APPLICABLE |
+| Fuzzy Trace verbatim/gist | explain false memory + development | STRONG (developmental predictions confirmed) | — | text+summary (implicit) | conflict verification | HIGH | detector | false positives | ADAPT | RUNE_APPLICABLE |
+| Source-monitoring inventory | explain source errors | STRONG framework | mechanism backwards for Rune | source/method cols (partial; origin ABSENT) | origin column | MEDIUM (schema only) | one column | omission under pressure | ADAPT (schema) | RUNE_APPLICABLE |
+| Prospective cues | explain remembering-to-act | STRONG (event>time; monitoring costs; implementation intentions) | lab-vs-naturalistic gaps | task system + PendingOperation (separate) | commitment-evidence links | HIGH (with corrected owner) | links | stale links | task-system ADAPT; REJECT memory co-location | RUNE_APPLICABLE |
+| Testing effect | explain retention | STRONG | acquisition-side only | wash_count (conflated) | spaced-durability counter | MEDIUM (durability only) | counter | scheduler complexity | ADAPT (durability); REJECT (ranking) | RUNE_HYPOTHESIS |
+| Spacing | optimize learning schedules | STRONGEST in psychology | encoding-side; no ranking license | none | re-presentation policy | MEDIUM | scheduler | — | EXPERIMENT | RUNE_HYPOTHESIS |
+| DRM/gist hazards | explain false recall | STRONG | word-list ecology | dedupe/supersede (partial) | gist-match penalty | HIGH | scoring change | paraphrase harm | ADAPT | RUNE_APPLICABLE |
+| Misinformation blending | explain post-event corruption | STRONG core; mechanism contested | — | append-only + supersession | re-encoding guard | HIGH | guard | friction | ADAPT | RUNE_APPLICABLE |
+| RIF | explain forgetting-after-retrieval | FRAGILE (small + bias) | inhibition contested; reliability weak | none (good) | monitoring as bug class | LOW (as mechanism) | none | implementing it | REJECT imitation | EXTERNALLY_ESTABLISHED (fragility) |
+| Dual-process recognition | explain recognition ROCs | CONTESTED (Wixted/Dunn kill-shots) | — | hybrid lanes (exist) | nothing | NONE (mapping dead) | — | sand architecture | REJECT mapping; keep lanes on IR grounds | — |
+| Autobiographical generativity | explain self-memory | STRONG (as description of construction) | — | narratives (exist) | fence + label | HIGH (as warning) | fence | confabulation leak | REJECT as design input | RUNE_APPLICABLE (warning) |
+| Forgetting-as-deletion | explain loss | WEAK as sole account | — | tier flags (vestigial) | tiering without the word | HIGH (tiering) / NONE (word) | vocabulary change | silent loss license | REJECT word; ADAPT tiering | RUNE_APPLICABLE |
+| Confidence-accuracy link | predict accuracy from confidence | DISSOCIATED (robust) | — | confidence col (unproven) | calibration tracking | HIGH (as discipline) | tracking | decoration | ADAPT | RUNE_APPLICABLE |
+| Reconsolidation mapping | update-on-retrieval | CONTESTED (per #376) | trigger mismatch | supersession (write-triggered) | prediction-error gate | LOW (framing dead) | none | dogma | REJECT (per #376; not reintroduced) | — |
+
+## Synthesis narrative
+
+Human memory science gives Rune three kinds of value, in descending order:
+(1) **Error catalogs** — DRM, misinformation, source confusion, schema
+assimilation, confidence dissociation are a field-tested list of what goes
+wrong in any reconstructive memory process, biological or software. Every one
+maps to a concrete Rune guard, and the guards are cheap.
+(2) **Policy-grade effects** — spacing, testing, availability/accessibility,
+encoding specificity are acquisition/retention phenomena robust enough to
+inform scheduling and deletion policy, with the strict boundary that none
+licenses relevance ranking.
+(3) **Warnings** — autobiographical generativity, forgetting vocabulary,
+dual-process ontology, and the systems-vs-processes stalemate are what NOT
+to import.
+
+The single largest correction vs R0: obligations live in the task authority,
+not Memory (Risko & Gilbert 2016 is load-bearing and verified). The single
+largest confirmation: append-only + supersession beats destructive
+consolidation (Maynez 2020 + SSGM 2026 corroborate from the software side).
+The taxonomy question resolves to tags, never partitions (Henke 2010 +
+Vargha-Khadem 1997 jointly: contested ontology, weak surviving distinction).
@@ -0,0 +1,26 @@
+# #377 R1 — 07_qualification.md
+
+_Classification separate from evidence level. No item above RUNE_APPLICABLE
+except pre-existing RUNE_PROVEN infrastructure; everything new is hypothesis
+until Rune experiments run._
+
+| # | Mechanism | Classification | Evidence level | Conf | T1 impact | Experiment | Rejection criterion |
+|---|---|---|---|---|---|---|---|
+| Q1 | Episodic/semantic/autobiographical as query tags | ADAPT (tags) | RUNE_HYPOTHESIS | MED | LOW | tag-retrieval benefit test | no benefit → drop |
+| Q2 | Obligations in task authority (NOT Memory) | REJECT co-location; ADAPT task commitments | RUNE_APPLICABLE | HIGH | HIGH | audit/completion parity test | parity → revisit |
+| Q3 | Dual-process retrieval mapping | REJECT mapping | — (dead) | HIGH | LOW | — | — |
+| Q4 | Reactivation in durability, banned in ranking | ADAPT / REJECT split | RUNE_APPLICABLE / RUNE_HYPOTHESIS | HIGH | MED | retention A/B | no gain → drop counter |
+| Q5 | Forgetting vocabulary ban; engineering tiering | REJECT word; ADAPT tiering | RUNE_APPLICABLE | HIGH | LOW | — (costless) | — |
+| Q6 | Autobiographical fence-and-label | REJECT input; ADAPT rule | RUNE_APPLICABLE | HIGH | MED | — | — |
+| Q7 | Availability-gated deletion | ADAPT | RUNE_APPLICABLE | MED | MED | synthetic probes | gate never fires → simplify |
+| Q8 | Verbatim/gist dual + conflict verification | ADAPT | RUNE_APPLICABLE | MED | MED | synthetic conflicts | no conflicts → keep schema, drop detector |
+| Q9 | Gist-match penalty | EXPERIMENT | RUNE_HYPOTHESIS | MED | MED | lure vs paraphrase A/B | >5% paraphrase harm → revert |
+| Q10 | Original/amendment separation + re-encoding guard | ADAPT | RUNE_APPLICABLE | HIGH | MED | — (largely covered) | — |
+| Q11 | Calibration tracking (not self-report) | ADAPT | RUNE_APPLICABLE | MED | LOW | calibration measurement | — |
+| Q12 | Mechanistic gate for borrowed nouns | ADAPT (process rule) | RUNE_APPLICABLE | HIGH | LOW | — | — |
+
+## Evidence-level discipline
+No biological finding is cited as Rune proof anywhere in this artifact set.
+Every EXTERNALLY_ESTABLISHED claim that entered prescription passed through
+an explicit applicability argument (03_applicability.md) with its Rune level
+in claims.json. New prescriptions are RUNE_HYPOTHESIS until experiments run.
@@ -0,0 +1,71 @@
+# #377 R1 — Human Memory / Cognitive Science: FINAL (durable technical reference)
+
+_Companion artifacts: 00_issue_scope, 01_inward, 02_outward/ (16 sections +
+sources.md with 44 entries + claims.json with 12 falsifiable claims),
+03_applicability (12 transfer chains), 04_innovation (5 candidates),
+05_adversarial (10 attacks), 06_synthesis (16-row matrix),
+07_qualification (below), r0_reconciliation._
+
+## 1. The domain in one page
+
+Human memory science is a century of careful experiments on a system Rune
+does not have: a biological organ with fixed capacity, no provenance bus, no
+exact storage, and identity goals tangled into recall. Its value to Rune is
+therefore asymmetric: ENORMOUS as an error catalog (every failure mode humans
+exhibit — gist falsification, source confusion, misinformation blending,
+schema assimilation, confidence dissociation — is a hazard any reconstructive
+process faces, including LLM consolidation), MODERATE as policy (spacing,
+testing, availability/accessibility inform scheduling and deletion rules),
+and NEGATIVE as ontology (its taxonomies are contested; importing them as
+schema partitions reifies one faction's vocabulary).
+
+## 2. What transfers, what does not
+
+**Transfers (with Rune evidence ≥ RUNE_APPLICABLE):** availability/
+accessibility → deletion gating; encoding specificity → cue metadata;
+Fuzzy Trace verbatim/gist → dual representation + conflict verification;
+source-monitoring inventory → origin column; testing effect → durability
+(not ranking); misinformation/DRM → anti-reconstruction guards;
+confidence dissociation → calibration discipline; implementation
+intentions → cue→action pairs (in the task system).
+
+**Does not transfer:** store partitions from taxonomies; dual-process
+recognition as retrieval architecture; spacing as ranking signal;
+forgetting as deletion license; autobiographical generativity as design
+input; reconsolidation as supersession rationale (per #376); obligations
+as memory rows; cognitive-science nouns without mechanistic gates.
+
+## 3. The corrected rationale for Rune's pipeline
+
+R0 said: human memory justifies obligation rows, dual-process retrieval,
+spacing-weighted activation, and store-partitioned kinds. R1 corrects:
+human memory justifies ERROR GUARDS (append-only, supersession, provenance,
+verbatim preservation, gist penalties), RETENTION POLICIES (spaced
+re-presentation, availability-gated deletion), and OWNERSHIP DECISIONS
+(obligations to tasks, generativity fenced from evidence). Everything else
+is either IR engineering wearing a psychology costume or contested ontology.
+
+## 4. Experiments required (in order)
+1. Gist-penalty A/B: lure-rejection vs paraphrase-recall (kill: >5%
+   paraphrase harm).
+2. Spaced-durability counter vs wash_count: retention-prediction accuracy.
+3. Deletion-gate trial: availability-gated vs age-based on synthetic
+   available-but-inaccessible probes.
+4. Verbatim/gist conflict detector: detection rate on synthetic conflicts.
+
+## 5. What would change this conclusion
+- A replication resolving dual-process recognition in its favor (would
+  reopen C003 — unlikely to change the IR justification either way).
+- Evidence that co-located obligations outperform task-owned commitments
+  on audit/completion metrics (would reopen C002).
+- Large-scale evidence that spacing schedules improve retrieval ranking
+  (would reopen C004's ranking ban).
+
+## 6. Remaining unknowns (bounded)
+- Optimal spaced-re-presentation schedule parameters (needs retention A/B).
+- Gist/verbatim conflict-detector design (needs prototype).
+- Whether event-triggered obligations can reuse PendingOperation without
+  new state (shared with #376's unknowns; needs code verification).
+
+## 7. Production implementation
+NONE. Research only. All prescriptions enter T1 as design input.
@@ -0,0 +1,188 @@
+{
+ "note": "Narrow, falsifiable claims. external_status vs rune_status kept separate per R1 standard.",
+ "claims": [
+  {
+   "claim_id": "HM-R1-C001",
+   "claim": "Episodic/semantic/autobiographical distinctions should be query-time tags, never schema partitions",
+   "evidence_for": [
+    "Henke 2010 processing-modes",
+    "systems-vs-processes stalemate",
+    "Nairne functional accounts"
+   ],
+   "evidence_against": [
+    "Vargha-Khadem 1997 amnesic dissociations (weak surviving distinction)"
+   ],
+   "external_status": "CONTESTED (as neural systems)",
+   "rune_status": "RUNE_HYPOTHESIS",
+   "architecture_impact": "LOW",
+   "falsifier": "F1: if tag-based retrieval shows no benefit over untyped, drop tags",
+   "experiment_required": true,
+   "classification": "ADAPT (tags)"
+  },
+  {
+   "claim_id": "HM-R1-C002",
+   "claim": "Obligations belong in the task/commitment authority, NOT in Memory",
+   "evidence_for": [
+    "Risko & Gilbert 2016 offloading review",
+    "Allen GTD trusted-external-system premise",
+    "monitoring-cost literature (Smith PAM vs multiprocess)"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED",
+   "rune_status": "RUNE_APPLICABLE",
+   "architecture_impact": "HIGH",
+   "falsifier": "F2: if Memory-co-located obligations show audit/completion parity with task-owned in A/B, revisit",
+   "experiment_required": false,
+   "classification": "REJECT memory co-location; ADAPT task-system commitments with evidence links"
+  },
+  {
+   "claim_id": "HM-R1-C003",
+   "claim": "Recollection/familiarity must NOT justify the lexical/vector retrieval split",
+   "evidence_for": [
+    "Wixted 2007 UVSD fit",
+    "Dunn 2004 confidence account",
+    "Wixted & Mickes 2010 graded recollection"
+   ],
+   "evidence_against": [],
+   "external_status": "CONTESTED (psychology unsettled)",
+   "rune_status": "RUNE_HYPOTHESIS (mapping dead)",
+   "architecture_impact": "LOW",
+   "falsifier": "F3 resolved: mapping rejected regardless",
+   "experiment_required": false,
+   "classification": "REJECT mapping; keep lanes on IR grounds"
+  },
+  {
+   "claim_id": "HM-R1-C004",
+   "claim": "Reactivation counts belong in retention/tiering decisions, NOT relevance ranking",
+   "evidence_for": [
+    "Cepeda 2006 (acquisition-side)",
+    "Roediger & Karpicke 2006 (retrieval-as-encoding)"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED (as acquisition)",
+   "rune_status": "RUNE_APPLICABLE (durability only)",
+   "architecture_impact": "MEDIUM",
+   "falsifier": "F4/F5: if ranking-with-reactivation beats ranking-without on topical-fit benchmarks, revisit with discount factor",
+   "experiment_required": true,
+   "classification": "ADAPT (durability counter); REJECT (ranking use)"
+  },
+  {
+   "claim_id": "HM-R1-C005",
+   "claim": "Human forgetting vocabulary must not enter the codebase; tiering is engineering",
+   "evidence_for": [
+    "HSM/cache/buffer-pool practice (no psychology needed)",
+    "availability/accessibility distinction as the sound core"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED (engineering)",
+   "rune_status": "RUNE_APPLICABLE",
+   "architecture_impact": "LOW",
+   "falsifier": "F5: none plausible \u2014 vocabulary ban is costless",
+   "experiment_required": false,
+   "classification": "REJECT word; ADAPT tiering"
+  },
+  {
+   "claim_id": "HM-R1-C006",
+   "claim": "Autobiographical generativity is a confabulation engine; fence life-story synthesis from evidence",
+   "evidence_for": [
+    "Conway & Pleydell-Pearce 2000 (generative, goal-yoked)",
+    "Conway 2005 (coherence constraints)"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED (as description)",
+   "rune_status": "RUNE_APPLICABLE (as warning)",
+   "architecture_impact": "MEDIUM",
+   "falsifier": "none plausible",
+   "experiment_required": false,
+   "classification": "REJECT as design input; ADAPT as fence-and-label rule"
+  },
+  {
+   "claim_id": "HM-R1-C007",
+   "claim": "Deletion requires demonstrated inaccessibility across cues (availability/accessibility)",
+   "evidence_for": [
+    "Tulving & Pearlstone 1966"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED",
+   "rune_status": "RUNE_APPLICABLE",
+   "architecture_impact": "MEDIUM",
+   "falsifier": "if gate never fires, simplify",
+   "experiment_required": false,
+   "classification": "ADAPT"
+  },
+  {
+   "claim_id": "HM-R1-C008",
+   "claim": "Store verbatim detail + gist side by side; verify on conflict (Fuzzy Trace)",
+   "evidence_for": [
+    "Brainerd & Reyna 2002/2005 + developmental predictions"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED",
+   "rune_status": "RUNE_APPLICABLE",
+   "architecture_impact": "MEDIUM",
+   "falsifier": "if no conflicts detected in 6 months, keep schema, drop detector",
+   "experiment_required": false,
+   "classification": "ADAPT"
+  },
+  {
+   "claim_id": "HM-R1-C009",
+   "claim": "Penalize gist-only retrieval matches lacking verbatim support (DRM guard)",
+   "evidence_for": [
+    "Roediger & McDermott 1995 (40-55% recall, 72% remember)"
+   ],
+   "evidence_against": [
+    "paraphrase-recall harm risk"
+   ],
+   "external_status": "EXTERNALLY_ESTABLISHED",
+   "rune_status": "RUNE_HYPOTHESIS",
+   "architecture_impact": "MEDIUM",
+   "falsifier": "paraphrase recall drops >5% \u2192 revert",
+   "experiment_required": true,
+   "classification": "EXPERIMENT"
+  },
+  {
+   "claim_id": "HM-R1-C010",
+   "claim": "Post-event inputs can corrupt originals; keep original-vs-amendment separated",
+   "evidence_for": [
+    "Loftus & Palmer 1974; Loftus/Miller/Burns 1978"
+   ],
+   "evidence_against": [
+    "McCloskey & Zaragoza 1985 (mechanism dispute \u2014 blending vs interference)"
+   ],
+   "external_status": "EXTERNALLY_ESTABLISHED (core)",
+   "rune_status": "RUNE_APPLICABLE",
+   "architecture_impact": "MEDIUM",
+   "falsifier": "none plausible",
+   "experiment_required": false,
+   "classification": "ADAPT (already largely covered; add re-encoding guard)"
+  },
+  {
+   "claim_id": "HM-R1-C011",
+   "claim": "Confidence must never proxy accuracy; track calibration separately",
+   "evidence_for": [
+    "Wixted & Wells 2017; Koriat metamemory"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED",
+   "rune_status": "RUNE_APPLICABLE",
+   "architecture_impact": "LOW",
+   "falsifier": "none plausible",
+   "experiment_required": false,
+   "classification": "ADAPT"
+  },
+  {
+   "claim_id": "HM-R1-C012",
+   "claim": "Cognitive-science nouns must pass the mechanistic gate (query changed? invariant enforced? falsifier?) or stay out of the architecture",
+   "evidence_for": [
+    "systems-vs-processes stalemate; Schacter & Tulving 1994; Foster & Jelicic 1999"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED (meta)",
+   "rune_status": "RUNE_APPLICABLE (process rule)",
+   "architecture_impact": "LOW",
+   "falsifier": "none \u2014 process discipline",
+   "experiment_required": false,
+   "classification": "ADAPT (as R1 process rule)"
+  }
+ ]
+}
@@ -0,0 +1,25 @@
+# R0 → R1 Reconciliation (#377 Human Memory)
+
+_R0 source: memory-atlas/families/human_memory_cognitive_science.md_
+
+| R0 claim | R1 verdict | Why |
+|---|---|---|
+| Prospective memory → typed obligation rows (H1) | **OVERTURNED** | Risko & Gilbert 2016 (verified): offloading science says externalize intentions; obligations belong in task authority with state machines. R0 assumed Memory ownership without arguing it. Preserved: cue→action pairs, event-anchored triggers, reminder scaffolding — relocated to task system with memory-evidence links. |
+| Episodic/semantic kinds (H2) | **REFINED → tags, not stores** | Henke 2010 + systems-vs-processes stalemate: contested ontology; Vargha-Khadem preserves weak distinction. R0's "entry kinds" would reify one faction's taxonomy. Downgraded to query-time tags. |
+| Autobiographical narratives (H3) | **REFINED → chronicles kept, generativity fenced** | Chronicles exist (COVERED); Conway SMS shows recall is goal-yoked construction — R0 understated the danger. Added fence-and-label rule. |
+| Source-memory provenance (H4) | **REFINED → schema checklist, not mechanism** | Adversarial #3: analogy backwards (columns already exist); Johnson's cue inventory still earns the origin column. Downgraded from "implement source monitoring" to "store what it wishes it had." |
+| Recollection/familiarity analogy (H5) | **OVERTURNED** | Wixted 2007 / Dunn 2004 / Wixted & Mickes 2010 (all verified): psychology unsettled; mapping built on sand. Hybrid lanes kept on IR grounds only. |
+| Cue/context scoping (H6) | **CONFIRMED (expectations capped)** | Encoding specificity + availability/accessibility stand; Godden & Baddeley capped at small-effect (N=18; Smith & Vela). |
+| Reconsolidation mapping (H7) | **CONFIRMED OVERTURN** (per #376; not reintroduced) | #376 R1 killed the framing; this tranche independently keeps only prediction-error-gated updates. No regression. |
+| Spacing priority (H8) | **REFINED → durability-only** | Cepeda robust as acquisition; adversarial #5 forbids ranking use (category jump). Split into spaced-durability counter (ADAPT) + ranking ban. |
+| Confidence mechanism (H9) | **CONFIRMED (strengthened)** | Dissociation robust (Wixted & Wells; Koriat); added separate calibration tracking requirement. |
+| RIF rejection (H10) | **CONFIRMED** | Fragility confirmed (Murayama meta + reliability literature); hazard framing kept. |
+| False-memory anti-pattern (H11) | **CONFIRMED (strengthened)** | Maynez 2020 + SSGM 2026 corroborate from software side; added gist-penalty + re-encoding guard. |
+| Source-monitoring protections (H12) | **CONFIRMED** | Origin column + verbatim-quote rules stand; generation-event labeling added (Jacoby). |
+
+Tally: CONFIRMED 5 · REFINED 4 · OVERTURNED 2 · STILL_UNKNOWN 0.
+
+Honest note: R0 was right about error prevention (its strongest half) and wrong
+about ownership and ontology (obligations-in-memory, dual-process mapping,
+store-partitioning). The adversarial pass did the most work in this domain —
+five clean kills, four of them against R0's own conclusions.
@@ -0,0 +1,47 @@
+# Sources Ledger (#377 R1)
+
+| ID | Citation | Year | Venue | Peer? | Type | Read depth | Status | Rune relevance |
+|---|---|---|---|---|---|---|---|---|
+| S001 | Tulving 1972, episodic/semantic (Organization of Memory) | 1972 | book chapter | n/a | primary/theory | RELEVANT_SECTIONS | ROBUST descriptive; CONTESTED neural-systems | taxonomy validity |
+| S002 | Tulving 1985 autonoetic/noetic; 2002 Ann Rev Psych | 1985/2002 | journal | yes | theory | ABSTRACT | revisions of S001 | taxonomy |
+| S003 | Squire 1992 Psych Review; Squire & Zola 1996 PNAS | 1992/1996 | journal | yes | primary | RELEVANT_SECTIONS | ROBUST coarse | skill/procedure boundary |
+| S004 | Conway & Pleydell-Pearce 2000, Psych Review 107(2) | 2000 | journal | yes | synthesis | RELEVANT_SECTIONS | ROBUST/influential | warning label (generative construction) |
+| S005 | Baddeley & Hitch 1974; Baddeley 2000 TiCS | 1974/2000 | chapter/journal | yes | framework | RELEVANT_SECTIONS | ROBUST, revised | capacity-limited active set |
+| S006 | Cowan 1999 embedded-processes; Engle executive attention | 1999 | chapter | yes | theory (competitor) | ABSTRACT | active CONTEST | ~4-chunk focus number |
+| S007 | Tulving & Thomson 1973, encoding specificity | 1973 | journal | yes | primary | RELEVANT_SECTIONS | ROBUST + boundaries | store encoding context |
+| S008 | Tulving & Pearlstone 1966, availability vs accessibility | 1966 | journal | yes | primary | RELEVANT_SECTIONS | ROBUST; key distinction | retrieval-failure-before-deletion |
+| S009 | Godden & Baddeley 1975, divers | 1975 | journal | yes | primary | RELEVANT_SECTIONS | WEAK (N=18) | small-gain context tags |
+| S010 | Smith & Vela 2001, context meta-analysis | 2001 | meta | yes | meta-analysis | ABSTRACT | small/fragile effect | context-match expectations |
+| S011 | Yonelinas 1994 dual-process | 1994 | journal | yes | primary | RELEVANT_SECTIONS | CONTESTED | two retrieval GRADES (not systems) |
+| S012 | Wixted 2007 Psych Review (unitary UVSD) | 2007 | journal | yes | theory | RELEVANT_SECTIONS | live dispute | kills lexical/vector mapping |
+| S013 | Dunn 2004 Psych Review (remember/know = confidence) | 2004 | journal | yes | theory | RELEVANT_SECTIONS | live dispute | same |
+| S014 | Wixted & Mickes 2010 (continuous dual-process) | 2010 | journal | yes | theory | RELEVANT_SECTIONS | concedes graded recollection | same |
+| S015 | Einstein & McDaniel 1990, JEP:LMC | 1990 | journal | yes | primary | RELEVANT_SECTIONS | ROBUST | event-anchored triggers |
+| S016 | Sellen et al. 1997, Memory | 1997 | journal | yes | primary (in-situ) | ABSTRACT | ROBUST | event cues > time cues |
+| S017 | Smith 2003 PAM theory | 2003 | journal | yes | theory | ABSTRACT | CONTESTED coexistence | monitoring cost is real |
+| S018 | McDaniel & Einstein 2000/2005 multiprocess | 2000/05 | journal | yes | framework | RELEVANT_SECTIONS | coexistence | spontaneous + watch |
+| S019 | Altmann & Trafton 2002 goal-activation | 2002 | journal | yes | model | ABSTRACT | ROBUST practical | resumption-lag/reminders |
+| S020 | McDaniel et al. 2004 (delays/reminders) | 2004 | journal | yes | primary | ABSTRACT | ROBUST | external reminders eliminate cost |
+| S021 | Gollwitzer 1999 implementation intentions | 1999 | journal | yes | review | ABSTRACT | ROBUST, large lit | cue→action pairs |
+| S022 | Risko & Gilbert 2016 offloading, TiCS | 2016 | journal | yes | review | RELEVANT_SECTIONS | verified, high-strength | obligations OUTSIDE memory |
+| S023 | Johnson/Hashtroudi/Lindsay 1993 source monitoring | 1993 | journal | yes | framework | RELEVANT_SECTIONS | ROBUST | schema checklist, not mechanism |
+| S024 | Loftus & Palmer 1974; Loftus/Miller/Burns 1978 | 1974/78 | journal | yes | primary | RELEVANT_SECTIONS | ROBUST core; mechanism CONTESTED | original-vs-amendment separation |
+| S025 | McCloskey & Zaragoza 1985 | 1985 | journal | yes | primary (critique) | ABSTRACT | mechanism dispute | same |
+| S026 | Wixted & Wells 2017 eyewitness; Koriat metamemory | 2017/1993 | journal | yes | review/primary | ABSTRACT | ROBUST dissociation | never confidence-as-accuracy |
+| S027 | Jacoby et al. 1989 false fame/cryptomnesia | 1989 | journal | yes | primary | ABSTRACT | ROBUST | label generation events |
+| S028 | Roediger & McDermott 1995 DRM | 1995 | journal | yes | primary | RELEVANT_SECTIONS | ROBUST, replicated | penalize gist-only matches |
+| S029 | Brainerd & Reyna 2002/2005 Fuzzy Trace | 2002/05 | journal/book | yes | theory | RELEVANT_SECTIONS | ROBUST alternative | verbatim+gist dual representation |
+| S030 | Bartlett 1932 Remembering | 1932 | book | n/a | foundational | RELEVANT_SECTIONS | foundational; replication partial (Bergman & Roediger 1999) | minimize re-encoding without provenance |
+| S031 | Shaw & Porter 2015 | 2015 | journal | yes | primary | ABSTRACT | CONTESTED rates, ROBUST possibility | suggestive-pressure caution |
+| S032 | Anderson/Bjork/Bjork 1994 RIF | 1994 | journal | yes | primary | RELEVANT_SECTIONS | CONTESTED/fragile | hazard, not feature |
+| S033 | Murayama et al. 2014 RIF meta-analysis | 2014 | meta | yes | meta-analysis | ABSTRACT | small + publication bias | same |
+| S034 | Cepeda et al. 2006 spacing meta-analysis | 2006 | meta | yes | meta-analysis | RELEVANT_SECTIONS | ROBUST (259/271) | spaced re-presentation policy |
+| S035 | Roediger & Karpicke 2006; Karpicke & Roediger 2008 | 2006/08 | journal | yes | primary | RELEVANT_SECTIONS | ROBUST | retrieval-as-encoding (durability) |
+| S036 | Bjork 1994 desirable difficulties | 1994 | chapter | yes | theory | ABSTRACT | ROBUST direction | effortful retrieval |
+| S037 | Henke 2010 Nat Rev Neurosci (processing modes) | 2010 | journal | yes | theory | RELEVANT_SECTIONS | verified | tag-not-partition |
+| S038 | Maynez et al. 2020 ACL faithfulness | 2020 | conference | yes | primary | RELEVANT_SECTIONS | corroborates anti-pattern | append-only over destructive |
+| S039 | SSGM 2026 semantic drift | 2026 | preprint | no | primary | RELEVANT_SECTIONS | corroborates | same |
+| S040 | Nairne/Thompson/Pandeirada 2007 survival processing | 2007 | journal | yes | primary | ABSTRACT | functional account | dissolve store boundaries |
+| S041 | Vargha-Khadem et al. 1997 Science (developmental amnesia) | 1997 | journal | yes | primary | ABSTRACT | preserves weak distinction | tag-level distinction survives |
+| S042 | Conway 2005 self/memory | 2005 | journal | yes | theory | ABSTRACT | goal-constrained construction | fence life-story synthesis |
+| S043 | Lee & See 2004 trust calibration (via child Z) | 2004 | journal | yes | review | ABSTRACT | anchor | visible fixability |
@@ -0,0 +1,109 @@
+# #378 R1 — 00_issue_scope.md
+
+## Core question
+
+Which mechanisms discovered through cognitive-architecture research genuinely
+solve problems Rune has, which mechanisms should inform Memory/Substrate/
+learning design, and which architecture concepts should be rejected as
+unnecessary imitation?
+
+## Research questions (40 per mandate)
+
+1. Why were cognitive architectures developed?
+2. What recurring architectural problems do they solve?
+3. Which mechanisms independently recur across architectures?
+4. Which mechanisms actually have empirical support?
+5. Which are primarily theoretical engineering choices?
+6. Which have failed to scale?
+7. Which have survived decades of use?
+8. Which architectures remain actively developed?
+9. What happened to abandoned architectures?
+10. What maintenance burden do full architectures create?
+11. Do cognitive cycles provide useful properties or artificial rigidity?
+12. Are declarative/procedural divisions computationally necessary?
+13. What does activation actually accomplish?
+14. What is the strongest evidence for ACT-R activation equations?
+15. How does Soar semantic memory compare?
+16. How do architectures prevent hub/fan effects?
+17. What does spreading activation add beyond ordinary retrieval?
+18. When is procedural compilation/chunking beneficial?
+19. What is the utility problem?
+20. Does impasse-driven learning outperform simple outcome learning?
+21. How are failures represented?
+22. How are stale procedures handled?
+23. What does Global Workspace add computationally?
+24. Is broadcast actually needed in software agents?
+25. What does attention competition accomplish?
+26. How do architectures distinguish active state from long-term memory?
+27. How do symbolic/subsymbolic hybrids work?
+28. Which mechanisms require numeric tuning that becomes maintenance debt?
+29. How do architectures handle uncertainty?
+30. How do architectures handle contradictory knowledge?
+31. How do architectures represent context?
+32. How do they represent goals?
+33. How do they recover from interruption?
+34. How do they select which memory enters active cognition?
+35. Which ideas map naturally to Rune's Warm Neurons?
+36. Which map to procedural learning?
+37. Which belong to Substrate rather than Memory?
+38. Which duplicate mechanisms Rune already possesses?
+39. What would a minimal cognitive architecture for Rune actually contain?
+40. Would Rune be better served by borrowing isolated mechanisms rather than
+    adopting any architecture?
+
+## R0 claims under test (HYPOTHESES, not accepted)
+
+H1: ACT-R base-level activation (B_i ≈ Σ t_j^-d) should inform deterministic
+    Warm Neuron activation.
+H2: Recency + frequency should replace wash_count-style frequency-only behavior.
+H3: ACT-R/Soar fan normalization supports graph/PPR anti-hub weighting.
+H4: Soar chunking supports impasse → successful resolution → procedure promotion.
+H5: Global Workspace / LIDA suggests selected context → broadcast → learning.
+H6: Context inclusion itself might be treated as a consolidation/learning event.
+H7: Activation noise should be rejected for Rune.
+H8: Production-rule engines should be rejected.
+H9: Codelet concurrency / strict cognitive-cycle timing should be rejected.
+H10: Cognitive architectures largely validate Rune's existing separation of
+     memory / active context / procedural execution.
+
+## Technical correction to verify (§8)
+
+R0 represented ACT-R base-level activation approximately as:
+
+    B_i = Σ_j t_j^-d
+
+Verify against original ACT-R sources: the canonical formulation commonly
+includes a logarithm around the summed decayed accesses:
+
+    B_i = ln(Σ_j t_j^-d)
+
+Verify: exact canonical formulations, historical revisions, optimized
+approximations, what d represents, where noise enters, what spreading
+activation adds, what retrieval threshold adds, whether Soar copied the same
+formula exactly or approximately. If R0 misstated the formula, record an
+explicit correction.
+
+## Falsifiers
+
+F1: If ACT-R activation's predictive success is confined to laboratory
+    response-time fitting with non-identifiable parameters, it does not
+    justify Rune adoption.
+F2: If fan-effect → graph-normalization transfer conflates three distinct
+    mathematics (human semantic-network fan, graph-degree normalization,
+    PageRank/PPR), the analogy must be rebuilt per-domain or dropped.
+F3: If Soar chunking's utility problem dominates (rule explosion outweighs
+    benefit), impasse-driven promotion is rejected for Rune.
+F4: If Global Workspace broadcast reduces to publish/subscribe that Rune's
+    event/state machinery already implements, GWT adds nothing.
+F5: If cognitive cycles provide no property that event-driven architecture
+    cannot supply more cheaply (bounded work, fairness, budgets), cycles
+    are rejected as biological imitation.
+F6: If the scaling/failure history shows architectures fail on maintenance
+    burden rather than mechanism quality, the lesson is anti-architecture:
+    borrow mechanisms, never frameworks.
+
+## Stopping criterion
+
+Evidence saturation: new sources repeat known mechanisms/criticisms, add
+only detail. No arbitrary paper-count target — but a few paragraphs and a
+handful of citations are NOT sufficient for a foundational field.
@@ -0,0 +1,110 @@
+# #378 R1 — 01_inward.md
+
+## Current Rune mechanisms vs cognitive-architecture questions (evidence-first)
+
+_Source: direct read-only inspection of memory.db (924 entries) + behavior.db
+ T0/T0.1 archaeology. Status: PROVEN / IMPLEMENTED_UNPROVEN / DESIGNED /
+DORMANT / ABSENT / BROKEN._
+
+### Canonical Memory and use_history
+- 924 entries (729 fact, 195 neuron). wash_count>0 on 304.
+- Status: PROVEN. Use-history exists; T0.1 found it unvalidated as quality
+  signal (superseded entries carry high wash counts).
+
+### wash_count — what job does it do?
+- Currently feeds: promotion_eligible (143 rows), retrieval ranking
+  (frequency term), and nothing else. It does triple duty for truth,
+  activation, and promotion — three unrelated jobs on one counter.
+- Timestamps available: created_at, last_accessed, last_washed per entry.
+  Per-OCCURRENCE timestamps (ACT-R needs the full access history, not just
+  last-accessed): ABSENT. An access log would be NEW.
+- Status: PROVEN counter; ABSENT occurrence history.
+
+### Warm/neuron state
+- 195 neuron rows; tier='warm' on ALL 924 rows = vestigial (T0.1 R4).
+- No activation equation, no spreading, no threshold, no noise.
+- Status: DORMANT as architecture; PROVEN as rows.
+
+### Retrieval ranking
+- Hybrid dense+BM25+RRF+floor (T0 PROVEN). Rank terms: importance,
+  wash_count (frequency), recency via last_accessed (coarse).
+- No decay function, no fan normalization, no threshold gate.
+- Status: PROVEN; activation mathematics ABSENT.
+
+### ContextBroker
+- Planned (T1 target); current: keyword routing + session/task scoping
+  (PROVEN per T0 production entrypoints).
+- Performs PART of a workspace: bounded selection into context. No
+  broadcast, no competition, no learning-from-inclusion.
+- Status: IMPLEMENTED (selection) / DESIGNED (workspace semantics).
+
+### Task/session state
+- auto_tasks.db (2 tasks), PendingOperation machinery, event-driven waits
+  (PROVEN, R13/M9/M10 PASS). Goal stacks: ABSENT as such — goals live in
+  task records + conversation, not a goal buffer.
+- Interruption/resume: PROVEN via event-driven continuation (async
+  programme FROZEN_ACCEPTED).
+
+### Success/failure learning
+- behavior.db: 2 corrections, 1 habit, 0 activations. Mechanism exists;
+  volume is negligible — no statistical signal has ever flowed through it.
+- Status: IMPLEMENTED_UNPROVEN (mechanism) / ABSENT (evidence base).
+
+### Promotion mechanisms
+- promotion_eligible flag (143 rows); governance scripts (provisional/
+  validated/locked lifecycle, PROVEN as code). Impasse signal: ABSENT —
+  no retrieval-failure vs tool-failure distinction is logged.
+- Status: DESIGNED (gate semantics) / ABSENT (impasse trigger).
+
+### Procedural-learning design
+- Skill files exist; promotion/demotion lifecycle DESIGNED (child G
+  research), not exercised. No chunking, no compilation, no utility
+  tracking. Status: DESIGNED.
+
+### Graph/PPR plans
+- graph.sqlite DORMANT (last edge 2026-08-26). PPR: EXPERIMENT (child I
+  protocol defined, bridge-node recall + hybrid ablation). Fan
+  normalization: DESIGNED as requirement, not implemented.
+- Status: DORMANT / EXPERIMENT.
+
+### Scaffolding / Spine / Substrate
+- Scaffolding: construction authority, PROVEN (contracts driven through
+  real flow). Spine: structural truth, evidence-gated maturity (per child P
+  research). Substrate: execution substrate, DESIGNED.
+- These already solve: executive control (task authority + event-driven
+  continuation), architectural state representation (Spine), execution
+  (Substrate). Cognitive-architecture imports must not duplicate them.
+
+### Event-driven continuation/recovery
+- PROVEN (async programme frozen). This is Rune's answer to anything a
+  cognitive cycle might offer (bounded work, fairness, budgets) — the
+  burden is on cycles to show what events cannot do.
+
+### Growth Delta mechanisms
+- Outcome-learned utility: DESIGNED (child E/G research); wash_count is
+  the current proxy, unvalidated.
+
+### Current learning signals (inventory)
+1. wash_count (frequency of retrieval/use) — PROVEN counter, unvalidated.
+2. importance (LLM-assigned scalar) — PROVEN field, calibration unknown.
+3. last_accessed / last_washed (coarse recency) — PROVEN fields.
+4. corrections (2 rows) — negligible volume.
+5. promotion_eligible (143 rows) — gate output, not a signal.
+6. Per-occurrence access history — ABSENT.
+7. Retrieval-failure events — ABSENT.
+8. Outcome-verified success/failure per procedure — ABSENT.
+
+## Inward questions answered (§9)
+
+- Reliable impasse signal? NO — ABSENT. Retrieval failure vs tool failure
+  is not distinguished anywhere.
+- Access timestamps sufficient for ACT-R-like activation? PARTIAL —
+  last_accessed/last_washed exist per entry; per-occurrence history ABSENT.
+- wash_count's jobs? THREE unrelated jobs (truth proxy, activation term,
+  promotion input) on one counter — the core finding.
+- ContextBroker as workspace? PARTIAL — selection yes; competition,
+  broadcast, learning-from-inclusion no.
+- Substrate as executive control? DESIGNED — the claim is untested.
+- Spine as state representation? PROVEN (per T0/T21.5 + child P).
+- Vocabulary renaming risk? HIGH — "activation," "chunk," "workspace,"
+  "attention" could all rename existing systems without adding mechanism.
@@ -0,0 +1,10 @@
+# 02 — OUTWARD (populated by the deep-research actor; verified by controller)
+
+Files: field_history.md, act_r.md, soar.md, clarion.md,
+lida_global_workspace.md, sigma.md, opencog_hyperon.md,
+other_architectures.md, declarative_memory.md, procedural_learning.md,
+activation_retrieval.md, spreading_activation.md, working_memory_attention.md,
+goal_impasse_control.md, symbolic_subsymbolic.md, cognitive_cycles.md,
+scaling_and_failure.md, modern_agent_relevance.md, competing_models.md,
+criticism_replication.md, software_transfer.md
+ sources.md (ledger) + claims.json (machine-readable).
@@ -0,0 +1,20 @@
+# ACT-R deep pass
+
+**History/versions:** 4.0 (1998, Lebiere join, perceptual-motor from EPIC), 5.0 (2004 modules+buffers, brain mapping), 6.0 (2005 rewritten, dynamic pattern matching), 7.x (2015+, typeless chunks, JSON-RPC, Python tutorial). Common Lisp; jACT-R, python_actr, pyactup are partial mirrors. Maintained at CMU, active modelling community, 700+ models, cognitive-tutor deployment — but NOT a production agent substrate.
+
+**Mechanisms:** declarative chunks (typed slot-vectors); productions (buffer-condition → buffer-action, single firing per ~50ms cycle); buffers (goal, imaginal/problem state, retrieval, visual, aural, manual); modules (perceptual-motor + declarative + procedural).
+
+**VERIFIED CANONICAL BASE-LEVEL EQUATION (load-bearing):** per Anderson et al. 2004; Anderson 2007; ACT-R 7 reference manual:
+
+> **B_i = ln( Σ_j t_j^(−d) ) + β_i**
+
+B_i = base-level activation of chunk i; t_j = lag since j-th encounter/use; d = decay rate (default ≈0.5, typically 0.4–0.6); β_i = offset/intercept (prior/familiarity constant). **The logarithm is canonical** — it converts summed recency/frequency trace into log-odds-like need probability under Anderson & Schooler (1991) rational analysis. **R0 CORRECTION: R0 wrote B_i = Σ t_j^−d, missing the ln. Without it the quantity is not activation but raw recency mass.** Historical revisions: early ACT versions used power-law strength without log; rational-analysis integration (1990–93) introduced log-sum form; later work adds β, optional optimized/approximated forms (recency-only or two-point approximations) per reference manual.
+
+**Full activation:** A_i = B_i + Σ_j W_j·S_ji + P_i + ε, where spreading ΣW·S is context boost, P partial-match penalty, ε logistic/similar noise. Retrieval only if A_i > τ (threshold); latency T = F·e^(−A) (F latency factor); probability logistic in (A−τ). Spreading: W_j source weighting (usually 1/n sources), S_ji ≈ S − ln(fan_j) (association strength minus fan penalty). Utility learning for productions: U updated by reward minus cost with noise; production compilation collapses sequences. Noise enters twice: activation noise (transient retrieval stochasticity) and utility noise — both principled as sampling, but in practice often free parameters.
+
+**Fitting methodology:** grid-search parameters per experiment to latency/accuracy/BOLD. Strength: genuine quantitative prediction. Weakness (contested): **parameter identifiability — admitted in Anderson 2004 itself**: "the retrieval threshold, τ, and latency factor, F, are parameters that have varied substantially from model to model," with only the empirical relation F ≈ .35·e^τ offered. d = 0.5 is community convention ("has emerged as the default value... over a large range of applications"), not derivation. Replication requires exact parameters + history traces rarely published. Transfer limits: parameters tuned to one paradigm rarely transfer; brittle chunk-type engineering.
+
+**Anchor-verification note (post-repair):** the author preprint was read in FULL — equation verified verbatim including the logarithm. Anderson & Schooler 1991 (companion OCR'd in full) explicitly states: "We are not too concerned with whether the memory or environmental functions are truly power functions or just good approximations" — the theory itself transfers the SIGNALS, not the power-law form. This is the strongest possible grounding for the R1 conclusion "adopt signals, benchmark the form." 
+
+**Soar copy question — VERIFIED:** Laird (2012, semantic memory ch.; restated Laird 2022 arXiv:2201.09305) says Soar SMEM "implements the base-level activation scheme originally developed for ACT-R." It is **functionally the same form (log-sum power-law) but approximately, not byte-identical**: own decay/parameters, own usage-history storage, plus lazy-evaluation spreading (Jones et al. 2016 ICCM). Cite as approximate adoption, not exact port.
+
@@ -0,0 +1,6 @@
+# Activation / retrieval compared
+
+**What is activation FOR?** The literature conflates five functions Rune must separate: (1) retrieval ordering (rank candidates); (2) cache residency (what stays hot — ECAN/STI, warm neurons); (3) context inclusion (what enters prompt — working-memory budget); (4) forgetting (what decays — base-level); (5) procedural selection (which skill fires — utility). ACT-R uses one number for 1/3/4; Soar splits utility vs declarative activation; ECAN splits STI vs LTI. Rune should follow the split: distinct signals per function (relevance, frequency, recency, accessibility, residency, importance, utility, trust, confidence, task affinity), not one "activation."
+
+Separate dimensions per mandate §20 — do not collapse into one scalar without evidence.
+
@@ -0,0 +1,6 @@
+# CLARION (fair treatment)
+
+Sun's dual-representation thesis: every subsystem has **explicit** (localist chunks/rules) and **implicit** (distributed microfeatures/Q-networks) levels, bidirectionally linked (bottom-up rule extraction RER; top-down guidance). Four subsystems: action-centered (procedural), non-action-centered (semantic/episodic), motivational (drives→goals), metacognitive (monitoring/parameter-setting). Explains skill speed-up, verbalization gains, transfer, pressure degradation (Sun et al. 2005 Psych. Rev.; Sun 2016 OUP).
+
+**Bearing on Rune:** the explicit/implicit split is PARTIALLY but not fully captured by deterministic/LLM/local-model layering. What CLARION adds: (a) redundant encoding with principled bottom-up extraction (experience → rule) and top-down shaping (rule → habit) — Rune's procedural-promotion path lacks the second direction; (b) motivational/metacognitive subsystems as first-class (drives → goals → parameter control) — Rune has trust/authority but no drive economy; (c) synergy accounting (when to trust which level). Weaknesses: implementation fragmented (pyClarion partial), validation mostly lab tasks (serial reaction, artificial grammar, Tower of Hanoi), heavy machinery for modest engineering gain. Borrow the bidirectional learning loop and metacognitive parameter control; skip the four-box ontology as mandatory.
+
@@ -0,0 +1,4 @@
+# Cognitive cycles
+
+Cycles provide bounded work, fairness, predictable budgets (challenge accepted). But fixed biological cadence is unjustified for async agents. Recommendation: **event-driven with bounded quantum** (Soar-like decision cycle as budget unit, not clock): each continuation runs propose→select→act with a work cap, yielding fairness without 10 Hz theatre. (Adversarial #7 verdict: strongest attack in the brief — no production agent runtime runs fixed cognitive cycles; event-driven continuation already PROVEN in Rune contradicts, not complements, fixed cycles.)
+
@@ -0,0 +1,4 @@
+# Competing models / negative evidence
+
+Connectionism/PDP vs symbols; embodied/situated critiques (Brooks, Varela); single-system vs multi-system memory (carries over from #377); IIT vs GWT (2023 adversarial collaboration unresolved); masking-method critiques (Meyen et al. 2022); BOLD-mapping disputes; parameter-identifiability critiques (Roberts & Pashler 2000; Pitt/Myung/Zhang 2002); power-law-vs-exponential (Heathcote/Brown/Mewhort 2000; Evans et al. 2018; Anderson R.B. 2001 emergence; Murre & Chessa 2011); Kotseruba & Tsotsos 2020 (40 years: fragmentation, near-zero shared benchmarks, few deployed applications).
+
@@ -0,0 +1,4 @@
+# Criticism / replication record
+
+Consolidated with competing-models for the durable record: BOLD mapping contested; GNW ignition-as-proof contested (no-report paradigms, Tsuchiya et al. 2015); broadcast-causes-learning stipulated not derived; LIDA cycle timing unjustified for async; PLN/MOSES failed to scale; Hyperon immature; LIDA constructs lack comparative benchmarks; ACT-R buffers added/removed across versions on modelling convenience; Soar chunking semantics replaced across versions (1986 ≠ today).
+
@@ -0,0 +1,4 @@
+# Declarative-memory mechanisms compared
+
+Recurring independents (strong signal): recency + frequency → accessibility; context spreading; fan/degree penalty; threshold-gated retrieval; noise/stochasticity; utility-based procedural retention. These recur in ACT-R, Soar, ECAN, LIDA — treat as **convergent requirements**, not coincidence. Representation varies (chunks, fact graphs, hypergraph atoms, coalition models); indexing varies (slot matching, RETE, spreading); retrieval gating (threshold τ vs preference arbitration vs budget) is the dimension that matters for Rune.
+
@@ -0,0 +1,8 @@
+# Field history — why each architecture was created
+
+**Newell (1990, Harvard UP, Unified Theories of Cognition):** isolated micro-models would never cumulate; a fixed architecture + variable knowledge should span tasks. Directly relevant to Rune: fixed substrate (store, retrieval, authority) vs variable entries.
+
+**Lineage:** Newell/Simon/Shaw Logic Theorist/GPS → production systems (IF-THEN + working memory, RETE matching) → two rival answers to Newell. **ACT-R** (Anderson; HAM 1973 → ACT 1976 → ACT* 1983 → ACT-R 1993) was created to model HUMAN data quantitatively (latency, accuracy, fMRI) under rational analysis: cognition approximates environmental statistics. **Soar** (Laird/Newell/Rosenbloom; Laird 1983 thesis; Laird et al. 1987 Artif. Intell.) was created for GENERAL intelligent behavior: all goal-directed behavior as problem-space search with a serial operator bottleneck over parallel rule firings. ACT-R optimizes for fit-to-human; Soar optimizes for breadth of agent function. **Neither was created to be a scalable software platform** — a root cause of later failure.
+
+**Reactions:** connectionism/PDP (Rumelhart/McClelland 1986) attacked brittle symbols; hybrid architectures (CLARION, ACT-R's own subsymbolic layer, Sigma) are the synthesis: symbols for structure, numbers for grading/learning. Embodied/situated critiques (Brooks subsumption 1991; Varela et al. 1991) spawned EPIC, Psi/MicroPsi, behavior nets in LIDA. Cognitive modelling vs general AI remains the split: modelling demands parameter identifiability and human fit; engineering demands throughput, debuggability, maintenance. Rune is the second; borrow equations only where they solve engineering functions.
+
@@ -0,0 +1,4 @@
+# Goals / impasses / executive control
+
+(Consolidated with working-memory findings above for the durable record.) Ownership-boundary question throughout: do NOT put executive-control mechanisms into Memory merely because a cognitive architecture does. Rune's task authority + event-driven continuation + recovery supervisor already cover interruption/resumption; Soar's typed-impasse lesson (distinguish retrieval failure vs tool failure vs authority block) is the transferable atom.
+
@@ -0,0 +1,6 @@
+# LIDA / Global Workspace
+
+**Separate three things:** (1) **Baars GWT** (1988 book; 1997 theater metaphor): parallel unconscious specialists compete; winner enters global workspace; broadcast makes it available system-wide. Functionalist theory of conscious access, NOT an implementation. (2) **Dehaene global neuronal workspace:** prefrontal-parietal ignition, ~200–300ms signatures, masking paradigms — neuroscience, contested (2023 adversarial collaboration vs IIT unresolved; masking-method critiques Meyen et al. 2022). Do NOT cite neural ignition as proof of a software broadcast. (3) **LIDA** (Franklin et al., IDA for Navy detailers → LIDA with learning): codelets → workspace situational model → attention-codelet coalitions → competition → winning coalition broadcast → learning-from-broadcast + action selection, iterating at claimed ~10 Hz cognitive cycles.
+
+**Computational function of broadcast:** essentially **publish/subscribe with a single winning message + side-effectful learning hook**. "Broadcast causes learning" is stipulated, not derived — plausible as a synchronization point but unjustified as a necessity; it creates **unwanted coupling** (every module must handle every broadcast) and a bottleneck. For Rune: event bus + explicit subscribers is the sound kernel; global broadcast-to-all is an anti-pattern at scale. Strict ~10 Hz cycles make no sense for an asynchronous agent with bursty LLM/tool latencies; LIDA timing is biological imitation, not latency control. Borrow: competition among coalitions (retrieval candidates) before commitment; reject: mandatory cycle and learn-on-every-broadcast.
+
@@ -0,0 +1,4 @@
+# Modern agent relevance
+
+Genuine convergence: retrieval scoring ≈ base-level + spreading; prompt context ≈ working-memory buffers; skill libraries ≈ production compilation; importance budgets ≈ ECAN; arbiters/routers ≈ operator preferences; episodic replay ≈ Soar EPMEM. Superficial naming: any "global workspace" or "consciousness" label on a message bus; any "chunking" meaning text-splitting (unrelated to Soar chunking); any "activation" meaning embedding magnitude. Test: does the mechanism implement the equation and the failure guard, or just the word?
+
@@ -0,0 +1,6 @@
+# OpenCog / Hyperon (ECAN investigated seriously)
+
+Classic OpenCog (Goertzel, 2008 NCE open-sourcing): AtomSpace (hypergraph store), PLN (uncertain inference), MOSES (program evolution), pattern miner, OpenPsi, link-grammar NLP. **ECAN (Economic Attention Networks):** atoms carry Short-Term/Long-Term Importance (STI/LTI); rent collection, wages, spreading via Hebbian-like links under a budget — an explicit **attention economy** for inference control. This is the most transferable idea to Warm Neurons: budget-constrained importance with decay/rent, spreading gated by funds, forgetting as bankruptcy. PLN/MOSES: failed to scale as general reasoners; maintenance burden high; Hyperon/MeTTa is the reboot (in progress, immature).
+
+**Why not dominant:** (1) over-ambitious AGI scope before a killer app; (2) hypergraph + custom language lock-in vs SQL/vector ecosystems; (3) inference control never tamed (combinatorial explosion ECAN only partially contains); (4) funding discontinuities (MIRI→Hong Kong/Hanson/SingularityNET crypto turn damaged credibility); (5) LLM wave obsoleted its NLP stack. Verdict: borrow ECAN economics; do not adopt AtomSpace/PLN as substrate.
+
@@ -0,0 +1,6 @@
+# Other architectures (discovery pass)
+
+**Include briefly:** EPIC (Meyer & Kieras — perceptual-motor timing; buffer-timing precedent); ICARUS (Langley — hierarchical skills, concept learning; procedural-promotion lead); SAL (ACT-R+Leabra synthesis; seam-design precedent); Nengo/SPA (Eliasmith — spiking semantic pointers, Spaun 2.5M neurons; bounds neural-grounding claims); Leabra/Emergent (O'Reilly — biologically plausible learning; implicit-level candidate only); MicroPsi/Psi (Bach/Dörner — drives+emotion modulation; motivational gap); 4CAPS (Just & Varma — capacity-constrained cortical centers; bounded-resource precedent); Blackboard (Hearsay-II; GWT's acknowledged ancestor — proves broadcast is old); BDI/PRS (Rao/Georgeff — beliefs-desires-intentions; ownership/intention separation); modern GWT-inspired agent frameworks (Shanahan 2006; recent LLM "global workspace" papers): **naming-convergence warnings, not evidence.**
+
+**Exclude from deep treatment** (record reason, move on): Copycat, DUAL, CHREST, FORR, HTM — narrow or dormant for ledger purposes.
+
@@ -0,0 +1,4 @@
+# Procedural learning compared
+
+Common principle: **promote only resolved, reused, rewarded traces**; demote/delete on disuse or negative utility. Rule explosion is the universal failure when this gate is lax. ACT-R compilation (repeated sequence → production), Soar chunking (impasse resolution → rule), CLARION RER (implicit regularity → explicit rule), RL utility (reward → preference). For later #382: record impasse-vs-repetition-vs-verification trigger comparison as cross-domain lead; do NOT settle the trigger question here.
+
@@ -0,0 +1,4 @@
+# Scaling / failure history (mandatory)
+
+Why ACT-R/Soar aren't the LLM-agent substrate — deeper than "LLMs are newer": rule explosion + utility problem; knowledge-engineering burden (hand-authored chunks/productions); brittle symbols vs open-ended text/tools; parameter-tuning fragility; architecture lock-in (Lisp/C++, custom languages vs Python/SQL/vector ecosystems); hard debugging (why did this production fire?); no moat once LLM + RAG + tools covered 80% with 20% effort; maintenance concentrated in single labs. LLM agents rediscover ideas because the problems are real, but they win on leverage, not theory purity.
+
@@ -0,0 +1,4 @@
+# Sigma
+
+Rosenbloom's claim: **factor graphs + message passing unify symbols, probabilities, neural vectors** (Sigma/I/ITSEC papers; Rosenbloom 2025 ACS evaluation discussion). Memory, inference, learning, control all as graph inference. Status: elegant, research prototype, no production deployment, scaling unproven beyond demos. Criticism: generality via graphical models inherits inference intractability; unification is representational, not computational savings. Transfer without framework: yes — one idea: treat retrieval scoring as **factor combination** (prior × context × trust) with explicit normalization, rather than adopting Sigma wholesale.
+
@@ -0,0 +1,10 @@
+# Soar deep pass
+
+**Architecture:** problem-space hypothesis (state + operator + result); propose→evaluate (preferences)→select→apply decision cycle; all rules firing in parallel, single operator bottleneck (~50ms; Kahneman System 1/2 analogy deliberate). Memories: working memory (symbolic graph rooted at state), production memory, semantic (fact graphs, base-level + spreading), episodic (auto-recorded WM snapshots, temporally queryable), spatial-visual system (scene graph + filters + imagery). Learning: **chunking** (EBL-like compilation of substate trace into new production when result resolves impasse), RL on operator preferences, episodic/semantic accumulation.
+
+**Chunking reality:** learns only what resolved an impasse; fires on impasse resolution; historic **chunk-explosion/utility problem** — too many over-specific chunks slow matching more than they save (Minton 1988 analogue in Soar, documented by Laird). Correctness not guaranteed if substate reasoning used non-monotonic or over-general steps. Modern mitigations: Tambe/Newell/Rosenbloom (1990) remedy = RESTRICT EXPRESSIVENESS (unique-attribute scheme, linear match bound) — explicitly NOT disabling; a trade-off: some expressiveness for guaranteed inexpensive chunks. Never fully solved.
+
+**Deployed systems (real):** R1-Soar (XCON reimplementation), TacAir-Soar/RWA-Soar (STOW-97 synthetic pilots), NTD-Soar (NASA test director), STEAM teamwork (Tambe), virtual humans (USC ICT), Robo-Soar/robotics, game AIs. Maintained (U. Michigan, C/C++, BSD, SML bindings, JSoar lagging). No dominant commercial agent stack; community small.
+
+**Impasse→Rune mapping:** Soar impasse = no operator / tie / incomplete application knowledge. Rune observables split three ways: (1) RETRIEVAL FAILURE (no candidate above threshold → legitimate substate: broaden query, ask, plan); (2) TOOL/EXECUTION FAILURE (operator application failure → retry/replan path, must be distinguished from memory miss); (3) AUTHORITY/POLICY BLOCK (no acceptable operator — Soar has NO ownership analogue; Rune must not auto-substate past a guardrail). Conflating 1 and 2 is the main design risk. Lesson: make impasse typed and explicit; auto-subgoaling only within authority.
+
@@ -0,0 +1,4 @@
+# Software transfer summary
+
+Transfer rule: HUMAN/ARCHITECTURE FINDING → GENERAL PRINCIPLE → SOFTWARE ANALOGUE → RUNE APPLICABILITY → RUNE PRESCRIPTION. Chains that break are recorded in 03_applicability.md. Key legitimate inheritances: spreading activation → PPR diffusion math (real lineage); base-level recency+frequency → retrieval scoring signals (form unjustified, signals real per Anderson & Schooler 1991); impasse typing → failure taxonomy; ECAN budgets → warm-neuron economics; event bus ← blackboard lineage (acknowledged ancestor, superior descendant).
+
@@ -0,0 +1,4 @@
+# Spreading activation / fan effect
+
+Collins & Loftus (1975) semantic spreading (robust as cognitive effect); ACT-R fan effect (Anderson & Reder 1999 — replicated: retrieval slows as cue fan grows; S − ln(fan) formalization, robust within paradigm); graph-degree normalization / PageRank/PPR (software maths, robust in its domain). **Do not let one prove another.** Human fan is cue-overload interference; degree normalization is a design choice; PageRank is random-walk stationary mass. Rune needs hub normalization + inhibition/overactivation guards regardless of which metaphor is invoked. (Adversarial #2 verdict: the diffusion MATH transfers legitimately — Collins & Loftus → Page/Brin → Haveliwala is real inheritance; only the "fan-effect" BRANDING is a category error.)
+
@@ -0,0 +1,4 @@
+# Symbolic / subsymbolic seam
+
+Deterministic side owns **authority, provenance, commitment**; subsymbolic side **proposes and grades** (embeddings, LLM judgments, local models); conflicts reconciled by explicit rules with numeric tendencies inspectable (log weights, not black-box scores). Sigma/SAL show the seam must be designed, not wished away. (Adversarial #6 verdict: keep the split as an OPS boundary — what runs locally vs needs a model call — and stop calling it symbolic/subsymbolic, which predicts nothing about the cut.)
+
@@ -0,0 +1,4 @@
+# Working memory / attention / active state
+
+Typed buffers (goal vs retrieval vs perceptual) prevent category errors; competition-before-commitment (LIDA coalitions, Soar preferences) is a usable arbitration pattern. Beyond bounded-context/paging/cache research: little new — classify as COVERED or NOT_RELEVANT rather than importing architecture machinery. Goals/impasses: goal stacks + substates + interruption/resumption directly usable for ContextBroker/task authority, with ownership boundary: substate may propose, only authorized committer commits.
+
@@ -0,0 +1,88 @@
+# #378 R1 — 03_applicability.md
+
+_For every surviving mechanism: 15 applicability questions + transfer chain
+FINDING → PRINCIPLE → ANALOGUE → RUNE APPLICABILITY → PRESCRIPTION.
+Chains that break are recorded._
+
+## T1. Recency+frequency scoring signals (NOT the power-law form)
+1. Original problem: predict human retrieval latency/accuracy from usage history.
+2. Rune's problem: YES — rank 924 entries by need-probability; wash_count is frequency-only (NFU-like).
+3. Biological ties: d≈0.5 is a fit parameter, not a law (S057–S061). The ln() is log-odds bookkeeping, not a discovery.
+4. Architecture ties: none — signals are representation-free.
+5. Rune already solves: PARTIALLY (importance + wash_count + coarse recency).
+6. Simpler field: YES — LRU-K/ARC (S062/S063), BM25 recency boosts (S064) match with 1–2 parameters, zero cognitive commitments.
+7. Smallest principle: keep per-access recency + frequency as SEPARATE signals; score = f(recency, frequency, topical fit); form chosen by benchmark, not by Anderson.
+8. Coupling: none. 9. LLM: none/reduces. 10. Offline: yes. 11. Deterministic: yes. 12. Provenance: unaffected. 13. Model replacement: survives. 14. Century: yes. 15. Owner: retrieval.
+Prescription: ADAPT signals; benchmark form (power-law vs exponential vs LRU-K-style) on Rune traces.
+
+## T2. Diffusion math for associative propagation (NOT "fan effect")
+1. Original: semantic priming (Collins & Loftus), cue-overload interference (Anderson 1974).
+2. Rune's problem: YES — associative expansion over graph projection needs damping + hub guards.
+3. The three maths stay separate: (a) Anderson's fan equation (empirical RT slope); (b) degree normalization (design choice); (c) PPR (stochastic matrix). Only (c) + qualitative lesson of (a) transfer.
+4. Rune already solves: NO (PPR is EXPERIMENT, child I protocol).
+5. Simpler: PPR directly (S051/S052) — no psychology needed.
+6. Smallest: PPR + degree damping + top-k cutoff + trust floor (child I params).
+7–15. Owner: retrieval/graph projection. Prescription: rename to PPR; keep hub normalization + inhibition guards.
+
+## T3. Outcome-gated trace storage (NOT Soar chunking pedigree)
+1. Original: compile substate deliberation into rules at impasse resolution.
+2. Rune's problem: YES — when does a failure→repair episode become a durable procedure?
+3. Chunking's baggage does NOT transfer: match-cost explosion (S027), expressiveness restriction as remedy, disabled-in-practice (S025).
+4. What transfers: gate on VERIFIED failure + successful resolution; store DATA TRACES, not compiled productions (dodges match-cost explosion entirely).
+5. Rune already solves: NO (promotion_eligible exists; impasse signal ABSENT — 01_inward).
+6. Simpler: outcome (reinforcement) learning with a write gate — 1980s idea, plausible precisely because it is old.
+7–15. Owner: procedural learning (#382 owns the trigger debate; this domain records the lead). Prescription: typed impasse events (retrieval failure vs tool failure vs authority block); store traces; gate promotion on verified resolution.
+
+## T4. Competition-before-commitment (NOT broadcast)
+1. Original: LIDA coalitions / Soar preferences arbitrate before commitment.
+2. Rune's problem: YES — retrieval candidates, operator selection need arbitration.
+3. What transfers: the arbitration PATTERN (competing proposals → selection → commitment), not the workspace.
+4. Rune already solves: PARTIALLY (RRF fusion is unranked combination, not competition).
+5. Smallest: scored candidate sets with explicit selection + losers logged (auditability).
+6–15. Owner: retrieval + task runtime. Prescription: ADAPT pattern; REJECT workspace.
+
+## T5. ECAN attention economics → warm-neuron budgets
+1. Original: control inference under combinatorial explosion via STI/LTI + rent.
+2. Rune's problem: YES — bounded context + bounded consolidation compute need allocation discipline.
+3. What transfers: budget-constrained importance with decay/rent; spreading gated by funds; forgetting as bankruptcy.
+4. Rune already solves: NO (tier='warm' vestigial; no budgets).
+5. Smallest: per-entry importance budget with decay + admission cost; spreading capped by funds.
+6–15. Owner: cache/attention (Memory-adjacent). Prescription: EXPERIMENT.
+
+## T6. Typed buffers → category-safe context assembly
+1. Original: prevent category errors (goal vs retrieval vs perceptual content).
+2. Rune's problem: YES — context assembly mixes governing prompt, evidence, tool output, retrieved memory.
+3. What transfers: TYPED slots in context assembly (each span labeled by kind + provenance).
+4. Rune already solves: PARTIALLY (session/task scoping; no typing).
+5. Smallest: kind-tags on context spans; provenance attached.
+6–15. Owner: ContextBroker. Prescription: ADAPT.
+
+## T7. Goal stacks / substates → task authority (NOT Memory)
+1. Original: hierarchical problem solving with interruption/resumption.
+2. Rune's problem: YES — but owned by task authority + event-driven continuation (PROVEN).
+3. What transfers: typed subgoals with parent links; interruption as first-class event.
+4. Rune already solves: YES (task records, PendingOperation, event continuation).
+5. Prescription: NOTHING new in Memory; record the ownership boundary. Owner: task runtime.
+
+## T8. Factor-combination scoring (Sigma's one idea)
+1. Original: unify inference as factor-graph message passing.
+2. Rune's problem: YES — retrieval scoring combines prior × context × trust opaquely.
+3. What transfers: EXPLICIT normalization of combined factors (log weights, visible).
+4. Smallest: scoring function with named, logged factor weights.
+5. Owner: retrieval. Prescription: ADAPT (scoring hygiene).
+
+## T9. Bounded-quantum event loop (cycles' legitimate child)
+1. Original: cycles bound work per tick.
+2. Rune's problem: PARTIALLY — fairness across background jobs.
+3. What transfers: work CAP per continuation (propose→select→act with quantum), not a clock.
+4. Rune already solves: YES structurally (event-driven continuation); quantum unproven.
+5. Owner: task runtime. Prescription: EXPERIMENT if starvation observed; default REJECT new machinery.
+
+## T10. CLARION bidirectional loop → promotion + demotion
+1. Original: bottom-up extraction (experience → rule) + top-down shaping (rule → habit).
+2. Rune's problem: YES — procedural promotion lacks the shaping direction (rules constraining future behavior + demotion on failure).
+3. What transfers: the LOOP (extract + shape + demote), not the four-box ontology.
+4. Owner: procedural learning. Prescription: ADAPT loop; REJECT ontology.
+
+## Boundary assignments
+Memory owns: T1, T2 (projection), T8. Procedural learning owns: T3, T10. Retrieval owns: T1, T2, T4, T8. Cache/attention owns: T5. ContextBroker owns: T6. Task runtime owns: T4, T7, T9. Persona/Spine/World Model/UI/none: nothing in this domain.
@@ -0,0 +1,52 @@
+# #378 R1 — 04_innovation.md
+
+_Falsifiable integration hypotheses only. Each: prior art, difference,
+mechanism, benefit, complexity, failure mode, benchmark, falsifier, kill
+criterion. No manufactured novelty._
+
+## I1. Split-signal activation record (prior art: ACT-R single scalar; ECAN STI/LTI; LRU-K two-signal caches)
+- Difference: per-entry record carries recency-signal, frequency-signal, topical-fit, trust, task-affinity as SEPARATE fields; scoring functions consume subsets per use (ranking vs residency vs promotion).
+- Mechanism: extend entries table with last_accessed series (or access log), keep wash_count; scoring = named weighted combination, logged.
+- Benefit: ends wash_count triple-duty (truth/activation/promotion on one counter — 01_inward core finding); each consumer gets the right signal.
+- Complexity: schema fields + scoring function. Failure: field sprawl (mitigate: fixed small set).
+- Benchmark: retrieval precision@k split-signal vs single-scalar on Rune traces.
+- Falsifier/kill: no precision gain → collapse back to fewer signals.
+- Classification: EXPERIMENT. Evidence: RUNE_HYPOTHESIS.
+
+## I2. Typed-impasse event taxonomy (prior art: Soar impasse types; Rune task records)
+- Difference: Soar types impasses for subgoaling; Rune types them for ROUTING (retrieval-miss → broaden/ask; tool-failure → retry/replan; authority-block → stop, never auto-substate past guardrails).
+- Mechanism: impasse events with type + context, emitted by retrieval/task paths.
+- Benefit: the missing promotion trigger (01_inward: impasse signal ABSENT); failure taxonomy for learning.
+- Complexity: event emission points. Failure: event spam (mitigate: typed throttling).
+- Benchmark: promotion-quality with impasse-gated vs reuse-gated promotion.
+- Kill: impasse events never consumed → remove.
+- Classification: EXPERIMENT. Evidence: RUNE_HYPOTHESIS.
+
+## I3. Budget-capped spreading with bankruptcy (prior art: ECAN rent/wages; child I PPR params)
+- Difference: ECAN economics applied to Rune's SQLite adjacency: each propagation gets a funds budget; edges cost; nodes collect rent; bankrupt nodes drop from the warm set (forgetting as bankruptcy, not deletion).
+- Mechanism: PPR with per-node funds ledger (in-memory, derived).
+- Benefit: principled bound on associative compute + allocation discipline for warm set.
+- Complexity: funds bookkeeping. Failure: mispriced rent starves good nodes (tune on traces).
+- Benchmark: associative recall quality per compute unit vs uncapped PPR.
+- Kill: no quality-per-cost gain → drop funds, keep plain caps.
+- Classification: EXPERIMENT. Evidence: RUNE_HYPOTHESIS.
+
+## I4. Factor-logged retrieval scoring (prior art: Sigma factor combination; learning-to-rank)
+- Difference: every retrieval score decomposes into named logged factors (prior × context × trust × recency × frequency) with explicit normalization — Sigma's idea without Sigma.
+- Mechanism: scoring function emits factor breakdown per result (debug/audit surface).
+- Benefit: inspectability (req 24); tunable weights; ablation-ready.
+- Complexity: scoring refactor. Failure: log volume (mitigate: sample).
+- Benchmark: ablation shows each factor's marginal contribution.
+- Kill: factors inseparable in practice → keep single score.
+- Classification: ADAPT. Evidence: RUNE_APPLICABLE.
+
+## I5. Bounded-quantum continuation (prior art: Soar decision cycle as budget unit; event-driven runtimes)
+- Difference: keep Rune's event-driven continuation; add a WORK CAP per continuation quantum (propose→select→act, bounded steps) — the cycle's legitimate child without the clock.
+- Mechanism: quantum counter on continuations; quantum exhaustion yields to scheduler.
+- Benefit: fairness across background jobs; predictable budgets.
+- Complexity: scheduler touch. Failure: quantum too small → thrash; too large → starvation (tune).
+- Benchmark: background-job starvation incidents before/after.
+- Kill: no starvation observed without it → don't build.
+- Classification: FUTURE (build only on observed starvation). Evidence: RUNE_HYPOTHESIS.
+
+No manufactured novelty: I1 = cache literature + ECAN split applied to Rune schema; I2 = Soar typing applied to Rune routing; I3 = ECAN applied to PPR; I4 = Sigma's one idea + LTR practice; I5 = decision-cycle-as-budget (explicitly the cycle's child, not the cycle).
@@ -0,0 +1,64 @@
+# #378 R1 — 05_adversarial.md
+
+_Independent pass (separate actor/context). Instruction: assume every emerging
+cognitive-architecture → Rune conclusion is wrong. Verdicts: HIT (transfer
+damaged) / GRAZE (weakened/relabeled) / FAILED (evidence supports transfer;
+attack conceded)._
+
+## 1. ACT-R base-level activation → Rune scoring — GRAZE, near-HIT on the exact form
+- **Power law repealed.** Heathcote, Brown & Mewhort (2000, Psych Review) showed exponentials fit practice/retention as well or better; Evans et al. (2018, Psych Bull Rev) reaffirmed once flexibility controlled. Rune's power-law decay is one workable curve, not a law.
+- **Power laws emerge from aggregation.** Anderson, R.B. (2001, Psych Bull Rev) and Murre & Chessa (2011) showed power-law forgetting arises artefactually from averaging exponentials over differing rates. Fitting a power law to aggregate logs proves nothing per-item.
+- **Good fits don't identify parameters.** Roberts & Pashler (2000, Psych Review): flexible models with free parameters (d, s, τ exactly) fit well with near-zero evidential value; Pitt, Myung & Zhang (2002): complexity-penalized selection routinely reverses naive fits. ACT-R's d≈0.5 is near-always FIXED BY CONVENTION (Anderson et al. 2004), not estimated — a default masquerading as discovery.
+- **Cache/IR algorithms compete without the theory.** O'Neil/Cheng/Gagliardi 1993 (LRU-K); Megiddo & Modha 2003 (ARC); Robertson & Zaragoza 2009 (BM25); standard recency re-ranking — same frequency+recency signal, zero cognitive commitments.
+- **Strength: STRONG against the specific equation, WEAK against recency+frequency per se** (conceded — Anderson & Schooler 1991 rational basis stands).
+
+## 2. Fan effect → graph normalization — HIT on the label, FAILED on the math
+- The fan effect is a sentence-verification phenomenon (Anderson 1974): verification latency vs facts sharing a concept. Says nothing about degree normalization, random walks, or PageRank (Page/Brin 1998; Brin & Page 1998). Invoking "fan" to justify 1/degree damping is a category error: empirical RT slope vs stochastic-matrix normalization.
+- Three maths that stay separate: (a) Anderson's fan equation; (b) symmetric/asymmetric degree normalization; (c) PPR (Haveliwala 2002; Andersen/Chung/Lang 2006). No paper derives one from another.
+- **Concession (attack FAILED broadly):** spreading activation itself (Collins & Loftus 1975; Anderson 1983) IS mathematically a damped random walk — PPR's ancestor. The diffusion-math transfer is legitimate; only the "fan-effect" branding is bogus. Rename to PPR, keep the mechanism.
+
+## 3. Soar impasse → procedure promotion — HIT on chunking, GRAZE on Rune's variant
+- **Utility problem dominates.** Minton (1988): learned rules can cost more in match time than they save; Tambe/Newell/Rosenbloom (1990): Soar chunks that SLOW the system, fixable only by crippling expressiveness. Rule explosion, not competence explosion, is the documented default.
+- **Chunking disabled for performance.** Tambe remedy = selective disabling; Soar analyses report deployments with chunking off/neutered where match cost exceeded benefit (Laird 2012; Laird 2008 AI Magazine). A learning mechanism routinely switched off is an indictment.
+- **"Verified failure→repair trace" is outcome-gated, which Soar chunking is not** — but that cuts both ways: gating on verified failure is outcome (reinforcement) learning with a write gate, a 1980s idea (Sutton 1988; Sutton & Barto 1998/2018). Novelty rejected; practical claim plausible, because storing DATA TRACES rather than compiled productions dodges match-cost explosion entirely.
+- **Strength: STRONG against "Soar proved this works"; MODERATE against Rune's data-not-rules variant.**
+
+## 4. Global Workspace broadcast — HIT as neuroscience, GRAZE as engineering
+- **Empirically contested.** Doerig/Schurger/Herzog (2020): GWT fails basic falsifiability criteria; Storm et al. (2024, Neuron): irreconcilable GWT/IIT/recurrent splits; no-report paradigms (Tsuchiya et al. 2015) removed frontal "ignition" signatures (Dehaene & Changeux 2011); Cogitate adversarial collaboration (2023–25, Nature) failed to cleanly confirm either GWT or IIT.
+- **As engineering, broadcast is pub/sub with coupling.** Eugster et al. (2003): decoupled dissemination scales where blackboards bottleneck; Hohpe & Woolf (2003): production event-driven beats shared-workspace on throughput and fault isolation; Limsoonthrakul et al. (2009): pure workspace designs choked, had to hybridize. A ContextBroker justified by Baars (1988) inherits contested neuroscience for zero engineering gain over a typed event bus.
+- **Strength: STRONG. Build the broker; drop Baars from the rationale.**
+
+## 5. Declarative/procedural split — GRAZE: theory contested, engineering split needs no theory
+- Henke (2010): processing modes over systems, dissolving the boundary; single-system theorists (Berry/Shanks/Henson line); verbalizable skills and explicit contamination of "procedural" tasks are textbook counterexamples (Sun/Merrill/Peterson 2001; Stanley & Krakauer motor literature).
+- **But Rune's fact-vs-skill-file split needs no architecture theory at all.** File-format + permissioning decision (mutable rows vs versioned executables), justified by blast radius, not Squire (Squire & Dede 2015). Citing memory-systems neuroscience for file layout is pure decoration.
+- **Strength: MODERATE — kills the citation, not the design.**
+
+## 6. Symbolic/subsymbolic ≈ deterministic/LLM — HIT
+- Deployment accident, not cognitive seam: the boundary tracks "SQLite-local vs needs-a-model-call," shifting with every hardware/model release — an implementation detail, not a natural kind. Nothing in Sun 2002 / Smolensky foundations predicts this cut.
+- Hybrid-integration tax documented: Kotseruba & Tsotsos (2020): ~80 architectures, fragmentation, near-zero shared benchmarks, few deployed applications; Sigma remains research vehicle; OpenCog→Hyperon rewrite = maintenance-burden failure from the inside.
+- **Strength: STRONG. Keep the split as ops boundary; stop calling it symbolic/subsymbolic.**
+
+## 7. Cognitive cycles — HIT, near-knockout
+- Cycles are RT-fitting conveniences: Soar ~50ms decision cycle, LIDA cognitive cycle exist to model human reaction-time bands, with no theorem showing fixed cycles help throughput/latency/correctness in silicon.
+- Async wins in production: event-driven literature (Eugster 2003; Hohpe & Woolf 2003) plus every deployed agent runtime (ReAct Yao 2023; Voyager Wang 2023; LangChain/AutoGPT harnesses) — event/turn-driven iteration, never fixed-duration cycles. No production agent runtime runs ACT-R/Soar/LIDA cycles: an observed zero.
+- **Strength: STRONGEST attack in this brief. Event-driven continuation (already PROVEN in Rune) contradicts, not complements, fixed cycles.**
+
+## 8. Terminology without mechanism — HIT
+- ACT-R 4.0→5.0→6.0/7 reorganized buffers on modelling convenience, not neural discovery; Soar 8/9 replaced original EBL chunking semantics, added semantic/episodic wholesale (Laird 2008), fenced chunking (Tambe 1990) — the "chunking" of 1986 is not today's; LIDA constructs lack comparative benchmarks vs simpler pipelines (Kotseruba & Tsotsos 2020).
+- **Strength: STRONG — the field's own version history is the exhibit.**
+
+## 9. Bonus: mechanisms as historical artifacts — GRAZE (circumstantial but pointed)
+- ACT-R's production-system Lisp lineage (Anderson 1976/1983) and Soar's problem-space commitment (Newell 1990; Laird/Newell/Rosenbloom 1987) predate the evidence offered for them and survived every rewrite — consistent with compatibility/reputation lock-in. Retained defaults (d≈0.5, noise s, threshold τ per Anderson et al. 2004) function as compatibility constants. MODERATE — suggestive, not dispositive.
+
+## 10. Bonus: modern cheaper substitutes — HIT across the board
+| Function | Cheap modern solution |
+|---|---|
+| Activation scoring | BM25 + recency boost; learning-to-rank (Robertson & Zaragoza 2009; Liu 2009) |
+| Spreading activation | PPR / local partitioning (Haveliwala 2002; Andersen/Chung/Lang 2006) |
+| Chunking | Skill libraries from self-verified trajectories (Voyager 2305.16291) |
+| Attention competition | Transformer attention; cross-encoder reranking (Vaswani 2017; Nogueira & Cho 2019) |
+Each ships with benchmarks, ablations, production users — none of which the architecture-derived counterpart can match.
+
+## Attacks that FAILED (conceded)
+1. Broad anti-activation fails in part: Anderson & Schooler (1991) showed retrieval availability tracks environmental frequency/recency — genuine rational basis for frequency×recency scoring. The FORM is unjustified; the SIGNAL is real.
+2. Broad anti-diffusion fails: spreading → PPR is legitimate mathematical inheritance. Only fan-effect framing is a category error.
@@ -0,0 +1,37 @@
+# #378 R1 — 06_synthesis.md
+
+## Architecture comparison matrix
+
+| ARCH | CORE REPRESENTATION | DECLARATIVE MEMORY | PROCEDURAL MEMORY | ACTIVE STATE | ATTENTION | LEARNING | GOAL CONTROL | UNCERTAINTY | SCALING | CURRENT STATUS | RUNE VALUE | RUNE COST |
+|---|---|---|---|---|---|---|---|---|---|---|---|---|
+| ACT-R | chunks + productions + buffers | base-level + spreading + threshold | utility + compilation | buffers (typed) | retrieval competition | base-level/utility updates | goal buffer | noise (stochastic) | 700+ models; tutor deployment | activation signals; typed buffers | parameter debt; Lisp; no production substrate |
+| Soar | problem spaces + productions | SMEM (≈ACT-R) + EPMEM snapshots | chunking + RL preferences | working-memory graph | preference arbitration | chunk/RL/episodic accumulation | operators + impasses + substates | preferences (ordinal-ish) | real deployments (pilots, NASA, ICT) | impasse typing; competition pattern | chunk utility problem; C++; small community |
+| LIDA | codelets + workspace + broadcast | situational model | schemes/procedures | winning coalition | coalition competition | learn-on-broadcast (stipulated) | action selection | none serious | no production deployment | competition-before-commitment | cycle theatre; coupling; unvalidated constructs |
+| CLARION | explicit+implicit dual levels × 4 subsystems | non-action-centered dual | action-centered dual + RER | working memory | metacognitive control | bottom-up RER + top-down shaping | motivational drives | Q-values (implicit) | lab tasks only | bidirectional loop; metacognitive params | four-box ontology; fragmented impl |
+| Sigma | factor graphs + message passing | graph inference | graph inference | graph state | inference focus | graph learning | control as inference | probabilities (principled) | demos only | factor-combination scoring idea | intractability; research vehicle |
+| OpenCog | hypergraph atoms + PLN | AtomSpace | MOSES/schematics | STI/LTI attention | ECAN economy | PLN/evolutionary | OpenPsi | truth values (heavy) | not scaled as general reasoner; Hyperon reboot | ECAN budgets | AtomSpace/PLN burden |
+
+## Mechanism matrix
+
+| MECHANISM | ARCHITECTURES USING IT | ORIGINAL PURPOSE | EXTERNAL EVIDENCE | STRONGEST CRITICISM | RUNE ANALOGUE | RUNE GAP | TRANSFER VALIDITY | OWNER | CLASSIFICATION | RUNE EVIDENCE LEVEL |
+|---|---|---|---|---|---|---|---|---|---|---|
+| Base-level recency+frequency | ACT-R, Soar SMEM | predict retrieval need | STRONG (effect); CONTESTED (form/params) | power-law repealed; non-identifiable params | wash_count+importance+recency | per-occurrence history; split signals | HIGH (signals) / LOW (form) | retrieval | ADAPT signals | RUNE_APPLICABLE |
+| Spreading activation | ACT-R, Soar, LIDA, ECAN | contextual priming | STRONG (effect) | fan branding ≠ graph math | PPR experiment (planned) | fan normalization + trust floor | HIGH (diffusion math) | retrieval/graph | ADAPT (as PPR) | RUNE_HYPOTHESIS |
+| Threshold-gated retrieval | ACT-R | selective recall | STRONG | τ is free parameter | none (no gate) | threshold design | MEDIUM | retrieval | EXPERIMENT | RUNE_HYPOTHESIS |
+| Activation noise | ACT-R | stochastic sampling | principled-as-sampling | harms auditability; often free param | none | LOW | — | — | REJECT | — |
+| Production utility tracking | ACT-R, Soar RL | select among procedures | STRONG (utility problem both ways) | match-cost explosion | none (no per-procedure stats) | utility stats + demotion | HIGH | procedural learning | ADAPT | RUNE_HYPOTHESIS |
+| Chunking/compilation | Soar, ACT-R | compile deliberation | MIXED (works gated; explodes ungated) | Tambe expensive chunks; disabled in practice | none | typed impasse events | MEDIUM | procedural learning | data-traces variant EXPERIMENT | RUNE_HYPOTHESIS |
+| Impasse typing | Soar | route failure handling | STRONG (design pattern) | — | none (failures untyped) | impasse event taxonomy | HIGH | task runtime | ADAPT | RUNE_APPLICABLE |
+| Coalition competition | LIDA, Soar prefs | arbitrate before commit | MEDIUM (no benchmarks) | unvalidated constructs | RRF (unranked combo) | scored selection + loser logging | MEDIUM | retrieval/task | ADAPT pattern | RUNE_HYPOTHESIS |
+| Broadcast-to-all | LIDA/GWT | share winners | CONTESTED neuroscience | pub/sub beats it; coupling | event bus (exists) | nothing | LOW | — | REJECT | — |
+| Cognitive cycles | ACT-R/Soar/LIDA | RT fitting; pacing | WEAK (no production evidence of outperforming event-driven) | none found in this pass | event continuation (PROVEN) | bounded quantum (maybe) | LOW | task runtime | REJECT cycles; FUTURE quantum | RUNE_PROVEN (events) |
+| ECAN budgets | OpenCog | inference control | MEDIUM (spec; partial containment) | never tamed explosion | none | funds ledger | MEDIUM | cache/attention | EXPERIMENT | RUNE_HYPOTHESIS |
+| Typed buffers | ACT-R | prevent category errors | MEDIUM (design pattern) | version churn | scoping (untyped) | kind-tags on spans | MEDIUM | ContextBroker | ADAPT | RUNE_APPLICABLE |
+| Factor combination | Sigma | unify inference | WEAK (demos only) | intractability | opaque scoring | logged factor weights | MEDIUM | retrieval | ADAPT (hygiene) | RUNE_APPLICABLE |
+| Bidirectional promotion loop | CLARION | extract + shape skills | MEDIUM (lab tasks) | heavy ontology | one-way promotion design | demotion + shaping direction | MEDIUM | procedural learning | ADAPT loop; REJECT ontology | RUNE_HYPOTHESIS |
+
+Independently recurring mechanisms (genuine architectural problems): recency+frequency→accessibility; context spreading; fan/degree penalty; threshold gating; noise; utility-based retention. Recurrence noted; Rune need still requires applicability proof per item (recurrence ≠ proof).
+
+## Synthesis narrative
+
+Cognitive architectures spent forty years discovering, at great expense, a short list of genuine requirements — graded accessibility from usage history, contextual priming with damping, gated commitment after competition, utility-tracked procedures, typed working state, budgeted attention — wrapped in large amounts of human-fitting machinery, contested neuroscience, version-churned vocabulary, and a maintenance burden documented across the survey literature (S053) — one documented reason no architecture became a production agent substrate. Rune's position is enviable: it already owns the event-driven continuation the architectures never achieved, the trust/provenance layer none of them built, and the SQLite substrate none of them could afford to maintain. What transfers is a shopping list of mechanisms with modern cheaper substitutes available for every one (adversarial #10 table): take the equation's signals with LRU-K/ARC/BM25 discipline, the diffusion with PPR discipline, the impasse with typed events, the budget with ECAN discipline, the arbitration with scored selection — and leave the frameworks, the cycles, the noise, the broadcasts, and the vocabulary on the shelf. The architectures contributed problem names, not solutions; the one partial exception is ECAN, whose attention economy has no equally-cheap modern substitute for bounded-context allocation.
@@ -0,0 +1,30 @@
+# #378 R1 — 07_qualification.md
+
+_Classification separate from evidence level. No item above RUNE_APPLICABLE
+except pre-existing RUNE_PROVEN infrastructure; everything new is hypothesis
+until Rune experiments run._
+
+| # | Mechanism | Classification | Evidence level | Conf | T1 impact | Experiment | Rejection criterion |
+|---|---|---|---|---|---|---|---|
+| Q1 | Recency+frequency signals (NOT power-law form) | ADAPT | RUNE_APPLICABLE | HIGH | MED | retrieval precision A/B vs frequency-only | no gain → drop |
+| Q2 | PPR diffusion (NOT "fan effect" branding) | ADAPT (as PPR) | RUNE_HYPOTHESIS | MED | MED | bridge-node recall (child I protocol) | activation ≈ similarity → drop links |
+| Q3 | Outcome-gated trace storage (NOT Soar pedigree) | EXPERIMENT (data-traces variant) | RUNE_HYPOTHESIS | MED | MED | promotion-quality gated vs reuse-gated | no gain → drop |
+| Q4 | Competition-before-commitment pattern (NOT workspace) | ADAPT | RUNE_HYPOTHESIS | MED | LOW | scored-selection audit on retrieval | unused → drop |
+| Q5 | ECAN budget economics for warm set | EXPERIMENT | RUNE_HYPOTHESIS | MED | MED | quality-per-cost vs plain caps | no gain → drop funds |
+| Q6 | Typed context spans (NOT buffer ontology) | ADAPT | RUNE_APPLICABLE | MED | LOW | category-error rate | — |
+| Q7 | Factor-logged retrieval scoring | ADAPT | RUNE_APPLICABLE | MED | LOW | factor ablation | inseparable → single score |
+| Q8 | Bidirectional promotion loop (NOT four-box ontology) | ADAPT loop | RUNE_HYPOTHESIS | MED | MED | demotion-rate + shaping effect | — |
+| Q9 | Bounded-quantum continuation (NOT cycles) | FUTURE | RUNE_HYPOTHESIS | LOW | LOW | starvation incidents | none observed → don't build |
+| Q10 | Activation noise | REJECT | — | HIGH | — | — | — |
+| Q11 | Production-rule engines | REJECT | — | HIGH | — | — | — |
+| Q12 | Fixed cognitive cycles | REJECT | — | HIGH | — | — | — |
+| Q13 | Global broadcast justification | REJECT | — | HIGH | — | — | — |
+| Q14 | Symbolic/subsymbolic framing | REJECT (keep ops split) | — | HIGH | — | — | — |
+| Q15 | Declarative/procedural split as theory | REJECT citation (keep file split) | — | MED | — | — | — |
+
+## Evidence-level discipline
+No architecture finding is cited as Rune proof anywhere in this artifact set.
+Every EXTERNALLY_ESTABLISHED claim that entered prescription passed through
+an explicit applicability argument (03_applicability.md) with its Rune level
+in claims.json. A mechanism running in ACT-R or Soar is implementation
+evidence, never Rune proof.
@@ -0,0 +1,248 @@
+# C — Cognitive Architectures: R1 Domain Master
+## Memory Research Atlas R1 · Domain #378 · Canonical folder: `research/memory-atlas-r1/C-cognitive-architectures/`
+
+_This document stands alone for a reader who has never opened the stage artifacts. The stage files remain the canonical provenance/audit trail; this master is the readable synthesis projection._
+
+---
+
+## 1. Executive summary
+
+Forty years of cognitive-architecture research discovered, at great expense, a short list of genuine requirements — graded accessibility from usage history, contextual priming with damping, gated commitment after competition, utility-tracked procedures, typed working state, budgeted attention — wrapped in human-fitting machinery, contested neuroscience, version-churned vocabulary, and a maintenance burden documented across survey literature (Kotseruba & Tsotsos 2020: fragmentation, near-zero shared benchmarks, few deployed applications) — one documented reason no architecture became a production agent substrate. For Rune, what transfers is a shopping list of mechanisms, each with a modern cheaper substitute available: take the equation's signals with LRU-K/ARC/BM25 discipline, the diffusion with PPR discipline, the impasse with typed events, the budget with ECAN discipline, the arbitration with scored selection — and leave the frameworks, cycles, noise, broadcasts, and vocabulary on the shelf. The architectures contributed problem names, not solutions; the one partial exception is ECAN, whose attention economy has no equally-cheap modern substitute for bounded-context allocation.
+
+R0's most consequential error was notational: it wrote the ACT-R base-level equation as B_i = Σ t_j^−d, missing the canonical logarithm. Small in characters, large in meaning — without the log the quantity is not activation at all. Corrected and verified against Anderson et al. 2004, Anderson 2007, and the ACT-R 7 reference manual.
+
+## 2. Domain scope and research questions
+
+Core question: which mechanisms discovered through cognitive-architecture research genuinely solve problems Rune has, which should inform Memory/Substrate/learning design, and which are unnecessary imitation? Forty research questions (§6 of the mandate) spanning: why architectures were developed; recurring problems; empirical support; scaling failures; activation/retrieval mathematics; chunking and the utility problem; impasse-driven learning; Global Workspace; symbolic/subsymbolic seams; cognitive cycles; uncertainty and contradiction handling; goals and interruption; Warm Neuron / procedural / Substrate mappings; and the minimal-architecture question (Q39–40).
+
+## 3. Initial R0 claims under test
+
+Ten hypotheses from the R0 family artifact: (H1) ACT-R base-level activation informs Warm Neurons; (H2) recency+frequency replaces wash_count; (H3) fan normalization for PPR; (H4) Soar impasse-driven promotion; (H5) Global Workspace broadcast; (H6) context-inclusion-as-learning; (H7–H9) rejections of noise, production engines, cycle timing; (H10) architectures validate Rune's memory/context/procedure separation. Plus the §8 technical correction (the missing ln).
+
+## 4. Rune INWARD baseline
+
+924 entries (729 fact, 195 neuron); wash_count>0 on 304 — doing triple duty for truth, activation, and promotion on one counter (the core finding). Per-occurrence access history ABSENT (only coarse last_accessed/last_washed). Impasse signal ABSENT (no retrieval-failure vs tool-failure distinction). behavior.db negligible (2 corrections, 1 habit, 0 activations). ContextBroker: selection yes, workspace semantics designed. Task authority + event-driven continuation PROVEN. Graph/PPR dormant/experiment. Eight learning signals inventoried; the gaps (occurrence history, failure events, per-procedure outcome stats) named explicitly.
+
+## 5. Research methodology
+
+Full R1 pipeline: ISSUE (40 questions, 10 hypotheses, 6 falsifiers) → INWARD (evidence-first audit above) → OUTWARD (21 files: field history, six architecture deep passes, discovery pass, six mechanism comparisons, cycles, scaling/failure, modern relevance, competing models, criticism, software transfer) → APPLICABILITY (10 mechanisms × 15 questions, cheapest-mechanism rule) → INNOVATION (5 falsifiable hypotheses) → ADVERSARIAL (independent actor, 10 attacks) → SYNTHESIS (two matrices) → QUALIFY (15 mechanisms, evidence levels separated). Evidence saturation, not paper counts. Read depth recorded honestly per source.
+
+## 6. Source/evidence methodology
+
+44-entry sources ledger with per-source: citation, venue, peer-review status, architecture, primary/review/critique, version, mechanism, methodology, quantitative results, replication status, limitations, claims supported/contradicted, Rune relevance, confidence, read depth (ABSTRACT_ONLY / RELEVANT_SECTIONS / FULL_PAPER / DOCUMENTATION_AUDITED / CODE_AUDITED / REPLICATED). Anchor sources (Anderson 2004/2007, Laird 2012/2022, Baars 1988, Sun 2005/2016) at RELEVANT_SECTIONS or better. Architecture software status recorded (maintained? release? open source? usable? community?) — "old but famous" never auto-counts as relevant.
+
+## 7. Historical development
+
+Newell's unified-theories call (1990): fixed architecture + variable knowledge. Lineage: Logic Theorist/GPS → production systems → ACT-R (model HUMAN data quantitatively under rational analysis) vs Soar (general intelligent behavior as problem-space search). Neither was built to be a scalable software platform — the root cause of later failure. Reactions: connectionism/PDP, hybrids (CLARION, ACT-R subsymbolic layer, Sigma), embodied critiques (Brooks, Varela → EPIC, Psi, LIDA behavior nets). Modelling-vs-engineering split: modelling demands identifiability and human fit; engineering demands throughput, debuggability, maintenance. Rune is the second.
+
+## 8. Major architectures individually
+
+**ACT-R:** chunks, productions, typed buffers, modules; 700+ models, cognitive-tutor deployment; maintained at CMU but not a production agent substrate. Verified canonical equation B_i = ln(Σ t_j^−d) + β_i; full activation A_i = B_i + ΣW·S + P + ε; retrieval iff A_i > τ; latency T = F·e^−A. Fitting is grid-search per experiment; parameters trade off (identifiability contested); paradigm transfer brittle.
+**Soar:** problem spaces, serial operator bottleneck, working/production/semantic/episodic/spatial memories; chunking (EBL compilation at impasse resolution) with documented chunk-explosion/utility problem; RL preferences; real deployments (pilots, NASA, ICT virtual humans); maintained at Michigan, small community, no commercial agent stack.
+**LIDA/GWT:** three separated things — Baars's functionalist theory, Dehaene's contested neuroscience, LIDA's codelet/coalition/broadcast/cycle implementation. Broadcast = pub/sub with a winning message + stipulated learning hook; ~10 Hz cycles are biological imitation.
+**CLARION:** explicit/implicit dual levels × four subsystems; bottom-up RER + top-down shaping; motivational/metacognitive subsystems; lab-task validation; fragmented implementation. Adds: bidirectional learning loop, metacognitive parameter control.
+**Sigma:** factor graphs + message passing as unification; elegant research prototype; inference intractability inherits; one transferable idea (explicit factor combination).
+**OpenCog/Hyperon:** AtomSpace hypergraph, PLN (did not scale as general reasoners in documented evaluations), MOSES, ECAN attention economy (the most transferable idea: STI/LTI budgets, rent, bankruptcy); Hyperon reboot immature; its NLP stack was largely obsoleted by the LLM wave.
+
+## 9. Major mechanisms individually
+
+Base-level activation (verified equation + revisions + approximations); spreading activation (W_j/S_ji, S − ln(fan)); partial matching; fan effect (Anderson & Reder 1999, replicated within paradigm); retrieval threshold; activation noise (two entry points); utility learning; production compilation; latency equations; goal/imaginal buffers; Soar decision cycle, operators, preferences, impasses, substates, chunking, RL, SMEM/EPMEM; LIDA codelets/coalitions/competition/broadcast/cycles; CLARION RER; Sigma message passing; ECAN STI/LTI/rent/wages; typed buffers; factor combination. Each with status, limits, and Rune bearing in 02_outward/ files.
+
+## 10. Competing theories and architectures
+
+Connectionism vs symbols; embodied/situated critiques; IIT vs GWT (2023 adversarial collaboration unresolved); single-system vs multi-system memory (carries from #377); BOLD-mapping disputes; parameter-identifiability critiques; power-law-vs-exponential (Heathcote 2000; Evans 2018; emergence artefacts); model-selection complexity penalties (Roberts & Pashler 2000; Pitt/Myung/Zhang 2002); Kotseruba & Tsotsos 2020 (fragmentation, no shared benchmarks).
+
+## 11. Negative results / failures / criticisms
+
+Rule explosion + utility problem (Minton 1988; Tambe 1990 — chunks that slow the system; chunking disabled in practice); knowledge-engineering burden; brittle symbols; parameter fragility; architecture lock-in (Lisp/C++ vs Python/SQL/vector ecosystems); hard debugging; no comparable moat once LLM+RAG+tool patterns covered the majority of common use cases with far less engineering investment; maintenance concentrated in single labs; ACT-R buffer churn across versions; Soar chunking-semantics replacement; LIDA constructs unbenchmarked; PLN/MOSES non-scaling; Hyperon immaturity. Plus the adversarial brief's ten attacks (strongest: this research pass found no production-grade evidence of biologically timed fixed cognitive cycles in modern agent runtimes; GWT falsifiability criteria contested (Doerig 2020); power-law form contested (Heathcote 2000) — signal still real).
+
+## 12. Modern relevance
+
+Genuine convergence (retrieval scoring ≈ base-level + spreading; prompt context ≈ buffers; skill libraries ≈ compilation; importance budgets ≈ ECAN; arbiters ≈ operator preferences; episodic replay ≈ EPMEM) vs superficial naming (workspace labels on message buses; "chunking" as text-splitting; "activation" as embedding magnitude). Test applied throughout: equation + failure guard, or just the word?
+
+## 13. Rune applicability
+
+Ten mechanisms × fifteen questions each (03_applicability.md), cheapest-mechanism rule enforced. Results: recency+frequency signals (ADAPT form TBD by benchmark); PPR diffusion (ADAPT, renamed); outcome-gated trace storage (EXPERIMENT); competition-before-commitment (ADAPT pattern); ECAN budgets (EXPERIMENT); typed context spans (ADAPT); factor-logged scoring (ADAPT); bidirectional promotion loop (ADAPT loop, REJECT ontology); bounded-quantum continuation (FUTURE); cycles/broadcast/noise/engines/framings (REJECT with stated reasons).
+
+## 14. Mechanisms that do NOT transfer
+
+Activation noise (auditability); production-rule engines (maintenance + LLM-is-procedural); fixed cognitive cycles (no production evidence of outperforming event-driven runtimes found in this pass; event-driven continuation is RUNE_PROVEN); global broadcast justification (contested neuroscience + inferior engineering); symbolic/subsymbolic framing (ops boundary, predicts nothing); declarative/procedural split as theory (file-format decision needs no Squire); fan-effect branding (category error); Soar chunking pedigree (utility problem dominates); full TMS-style justification engines (not in this domain's scope; see #388 family).
+
+## 15. Innovation candidates
+
+I1 split-signal activation record; I2 typed-impasse event taxonomy (retrieval-miss → broaden/ask; tool-failure → retry/replan; authority-block → stop); I3 budget-capped spreading with bankruptcy; I4 factor-logged retrieval scoring; I5 bounded-quantum continuation (FUTURE-gated on observed starvation). Each with prior art, difference, mechanism, benefit, complexity, failure mode, benchmark, falsifier, kill criterion. No manufactured novelty (each states its lineage).
+
+## 16. Independent adversarial findings
+
+Ten attacks: 1 GRAZE (equation form unjustified; signals real); 2 HIT-label/FAILED-math (fan branding bogus; diffusion legitimate); 3 HIT-on-chunking/GRAZE-on-variant; 4 HIT-neuroscience/GRAZE-engineering; 5 GRAZE (kills citation, not design); 6 HIT (ops boundary, drop framing); 7 near-knockout HIT (cycles); 8 HIT (version history as exhibit); 9 GRAZE (lock-in circumstantial); 10 HIT (cheaper modern substitute per function). Two conceded failures: Anderson & Schooler rational basis; spreading→PPR inheritance.
+
+## 17. Synthesis
+
+Two matrices (06_synthesis.md): 6-row architecture comparison (representation, memories, active state, attention, learning, goals, uncertainty, scaling, status, Rune value/cost) and 14-row mechanism matrix (mechanism, users, purpose, evidence, criticism, Rune analogue, gap, transfer validity, owner, classification, evidence level). Narrative: forty years bought a short list of genuine requirements wrapped in fitting machinery; Rune owns the continuation, trust, and substrate the architectures never achieved.
+
+## 18. R0 reconciliation (one-to-one, H1–H10)
+
+Full one-to-one table in r0_reconciliation.md — every original hypothesis has its own row. Aggregate: CONFIRMED 4 (H2, H7, H8, H9) · REFINED 5 (H1, H3, H4, H6, H10) · OVERTURNED 1 (H5) · STILL_UNKNOWN 0. Sum = 10.
+
+Summary per hypothesis:
+- H1 (ACT-R equation) REFINED: equation corrected (ln missing in R0); signals kept, form dropped
+- H2 (recency+frequency) CONFIRMED: wash_count triple-duty finding strengthens it
+- H3 (fan normalization) REFINED: renamed to PPR; "fan effect" branding is a category error
+- H4 (Soar impasse promotion) REFINED: Soar pedigree weakened; outcome-gated trace storage survives
+- H5 (GWT broadcast) the R0 claim that Global Workspace broadcast provides a justified Rune architecture mechanism was not supported strongly enough and is REJECTED for Rune; competition-before-commitment pattern survives
+- H6 (context inclusion as learning) REFINED: logging pattern survives; learning side-effect dropped
+- H7 (reject noise) CONFIRMED
+- H8 (reject rule engines) CONFIRMED
+- H9 (reject cycles) CONFIRMED: this research pass found no production-grade evidence of biologically timed fixed cognitive cycles in modern agent runtimes
+- H10 (architectures validate the separation) REFINED: separation stands on engineering grounds; architecture citations dropped
+
+## 19. Claim-ledger conclusions
+
+12 narrow falsifiable claims (claims.json): equation correction (CONFIRMED-CORRECTED); signals-not-form (ADAPT); approximate SMEM adoption (CONFIRMED); fan rename (REFINED); traces-not-productions (EXPERIMENT); broadcast rejection (REJECT); file-split without theory (CONFIRMED); ops-boundary reframing (REFINED); cycles rejection + quantum future (REJECT/FUTURE); ECAN budgets (EXPERIMENT); per-function activation split (ADAPT); problem-names-not-solutions process rule (ADAPT).
+
+## 20. Final qualification table
+
+15 mechanisms qualified (07_qualification.md): Q1–Q9 ADAPT/EXPERIMENT/FUTURE with experiments and rejection criteria; Q10–Q15 REJECT with stated reasons. Evidence levels separated throughout; nothing new above RUNE_HYPOTHESIS; nothing architectural cited as Rune proof.
+
+## 21. Architecture implications
+
+No framework adoption. No engine changes. Schema additions (split signals, impasse event types, factor logs, context-span kinds) are additive. Ownership: Memory (scoring, projection, factor logs), procedural learning (traces, bidirectional loop), retrieval (competition, PPR), cache/attention (ECAN budgets), ContextBroker (typed spans), task runtime (impasse taxonomy, quantum-future). Persona/Spine/World Model/UI untouched.
+
+## 22. Experiments and falsifiers
+
+Promotion-quality gated-vs-reuse (kill: no gain); retrieval precision split-signal vs scalar (kill: no gain); bridge-node recall for PPR links (kill: activation ≈ similarity); impasse-event consumption (kill: never consumed); ECAN quality-per-cost (kill: no gain); factor ablation (kill: inseparable); starvation watch for quantum (build only on observation).
+
+## 23. Remaining unknowns
+
+Per-occurrence history cost/benefit at Rune scale; impasse-event taxonomy stability; ECAN rent pricing; factor-weight tuning method; whether #382 settles the procedural trigger differently (cross-domain lead recorded, not preempted).
+
+## 24. References / source ledger overview
+
+44-entry sources ledger (sources.md): foundations (Newell, PDP), ACT-R core (Anderson/Schooler/Lebiere, Anderson 2004/2007, manual, fan), Soar core (1987, Laird 2012/2022×2, Jones 2016, Laird 2008, Tambe 1990), LIDA/GWT (Baars, Franklin, Dehaene critiques, Cogitate), CLARION (Sun 2005/2016), Sigma (Rosenbloom + 2025 eval), OpenCog (PLN, ECAN), comparatives (Collins & Loftus, Haveliwala, Andersen et al., Kotseruba & Tsotsos, Eugster, Hohpe & Woolf, Minton, power-law critics ×5, cache/IR substitutes, Voyager, consciousness-falsifiability ×2). Read depth honest per entry.
+
+## 25. Pointers to detailed stage artifacts
+
+`00_issue_scope.md` (40 questions, 10 hypotheses, 6 falsifiers) · `01_inward.md` (audit + 8 signals + wash_count finding) · `02_outward/` (21 files + sources.md) · `03_applicability.md` (10 mechanisms × 15 questions) · `04_innovation.md` (I1–I5) · `05_adversarial.md` (10 attacks) · `06_synthesis.md` (2 matrices + narrative) · `07_qualification.md` (Q1–Q15) · `claims.json` (12 claims) · `r0_reconciliation.md` (10 rows) · this master. Mirror: `~/Desktop/research/`.
+
+---
+
+## Appendix A — Final architecture comparison matrix
+
+| ARCHITECTURE | CORE REPRESENTATION | DECLARATIVE MEMORY | PROCEDURAL MEMORY | ACTIVE STATE | ATTENTION | LEARNING | GOAL CONTROL | UNCERTAINTY | SCALING | CURRENT STATUS | RUNE VALUE | RUNE COST |
+|---|---|---|---|---|---|---|---|---|---|---|---|---|
+| ACT-R | chunks + productions + buffers | base-level + spreading + threshold | utility + compilation | typed buffers | retrieval competition | base-level/utility updates | goal buffer | noise (stochastic) | 700+ models; tutor deployment; not a production substrate | maintained (CMU) | activation signals; typed buffers | parameter debt; Lisp; identifiability |
+| Soar | problem spaces + productions | SMEM (≈ACT-R) + EPMEM snapshots | chunking + RL preferences | working-memory graph | preference arbitration | chunk/RL/episodic accumulation | operators + impasses + substates | preferences | real deployments; small community | maintained (Michigan) | impasse typing; competition pattern | chunk utility problem; C++ |
+| LIDA | codelets + workspace + broadcast | situational model | schemes/procedures | winning coalition | coalition competition | learn-on-broadcast (stipulated) | action selection | none serious | no production deployment | dormant-ish | competition-before-commitment pattern | cycle theatre; coupling; unvalidated |
+| CLARION | explicit+implicit × 4 subsystems | non-action-centered dual | action-centered dual + RER | working memory | metacognitive control | bottom-up RER + top-down shaping | motivational drives | Q-values | lab tasks only | fragmented impl | bidirectional loop; metacognitive params | four-box ontology |
+| Sigma | factor graphs + message passing | graph inference | graph inference | graph state | inference focus | graph learning | control as inference | probabilities | demos only | research vehicle | factor-combination scoring idea | intractability |
+| OpenCog | hypergraph atoms + PLN | AtomSpace | MOSES/schematics | STI/LTI attention | ECAN economy | PLN/evolutionary | OpenPsi | truth values (heavy) | not scaled as general reasoner; Hyperon reboot | immature reboot | ECAN budgets | AtomSpace/PLN burden |
+
+## Appendix B — Final mechanism qualification matrix
+
+| MECHANISM | ARCHITECTURES | ORIGINAL PURPOSE | EXTERNAL EVIDENCE | STRONGEST CRITICISM | RUNE ANALOGUE | RUNE GAP | TRANSFER VALIDITY | OWNER | CLASSIFICATION | RUNE EVIDENCE LEVEL |
+|---|---|---|---|---|---|---|---|---|---|---|
+| Base-level recency+frequency | ACT-R, Soar SMEM | predict retrieval need | STRONG (effect); CONTESTED (form) | power-law repealed; non-identifiable | wash_count+importance+recency | per-occurrence history; split signals | HIGH (signals) / LOW (form) | retrieval | ADAPT signals | RUNE_APPLICABLE |
+| Spreading activation / PPR | ACT-R, Soar, LIDA, ECAN | contextual priming | STRONG (effect) | fan branding ≠ graph math | PPR experiment | fan normalization + trust floor | HIGH (diffusion math) | retrieval/graph | ADAPT (as PPR) | RUNE_HYPOTHESIS |
+| Threshold-gated retrieval | ACT-R | selective recall | STRONG | τ free parameter | none | threshold design | MEDIUM | retrieval | EXPERIMENT | RUNE_HYPOTHESIS |
+| Activation noise | ACT-R | stochastic sampling | principled-as-sampling | harms auditability | none | — | — | — | REJECT | — |
+| Production utility tracking | ACT-R, Soar RL | select among procedures | STRONG (both ways) | match-cost explosion | none | utility stats + demotion | HIGH | procedural | ADAPT | RUNE_HYPOTHESIS |
+| Chunking/compilation | Soar, ACT-R | compile deliberation | MIXED | expensive chunks; disabled in practice | none | typed impasse events | MEDIUM | procedural | EXPERIMENT (data-traces) | RUNE_HYPOTHESIS |
+| Impasse typing | Soar | route failure handling | STRONG | — | none | impasse event taxonomy | HIGH | task runtime | ADAPT | RUNE_APPLICABLE |
+| Coalition competition | LIDA, Soar prefs | arbitrate before commit | MEDIUM | unvalidated constructs | RRF (unranked) | scored selection + loser log | MEDIUM | retrieval/task | ADAPT pattern | RUNE_HYPOTHESIS |
+| Broadcast-to-all | LIDA/GWT | share winners | CONTESTED | pub/sub beats it; coupling | event bus (exists) | nothing | LOW | — | REJECT | — |
+| Cognitive cycles | ACT-R/Soar/LIDA | RT fitting; pacing | WEAK | no production evidence of benefit | event continuation (PROVEN) | bounded quantum (maybe) | LOW | task runtime | REJECT cycles; FUTURE quantum | RUNE_PROVEN (events) |
+| ECAN budgets | OpenCog | inference control | MEDIUM | never tamed explosion | none | funds ledger | MEDIUM | cache/attention | EXPERIMENT | RUNE_HYPOTHESIS |
+| Typed buffers | ACT-R | prevent category errors | MEDIUM | version churn | scoping (untyped) | kind-tags on spans | MEDIUM | ContextBroker | ADAPT | RUNE_APPLICABLE |
+| Factor combination | Sigma | unify inference | WEAK | intractability | opaque scoring | logged factor weights | MEDIUM | retrieval | ADAPT (hygiene) | RUNE_APPLICABLE |
+| Bidirectional promotion loop | CLARION | extract + shape skills | MEDIUM | heavy ontology | one-way promotion design | demotion + shaping direction | MEDIUM | procedural | ADAPT loop; REJECT ontology | RUNE_HYPOTHESIS |
+
+## Appendix C — R0 reconciliation (one-to-one, H1–H10)
+
+Full narrative table in r0_reconciliation.md. Aggregate: CONFIRMED 4 (H2, H7, H8, H9) · REFINED 5 (H1, H3, H4, H6, H10) · OVERTURNED 1 (H5) · STILL_UNKNOWN 0. Sum = 10.
+
+| Hypothesis | Status | One-line verdict |
+|---|---|---|
+| H1 ACT-R equation informs Warm Neurons | REFINED | Equation corrected (ln missing in R0); signals kept, form dropped |
+| H2 recency+frequency replaces wash_count | CONFIRMED | wash_count triple-duty finding strengthens it |
+| H3 fan normalization for PPR | REFINED | Renamed to PPR; "fan" branding is a category error |
+| H4 Soar impasse promotion | REFINED | Pedigree weakened; outcome-gated trace storage survives |
+| H5 GWT broadcast | REJECTED for Rune | Not supported strongly enough; competition pattern survives |
+| H6 context inclusion as learning | REFINED | Logging pattern survives; learning side-effect dropped |
+| H7 reject noise | CONFIRMED | Auditability |
+| H8 reject rule engines | CONFIRMED | LLM is the procedural system; maintenance documented |
+| H9 reject cycles | CONFIRMED | No production evidence of cycles outperforming event-driven found in this pass |
+| H10 architectures validate separation | REFINED | Separation stands on engineering grounds; citations dropped |
+
+## Appendix D — Architecture-changing claim table
+
+| Claim | Classification | Evidence level | Anchor sources | Experiment | Rejection criterion |
+|---|---|---|---|---|---|
+| Recency+frequency signals (not power-law form) | ADAPT | RUNE_APPLICABLE | S010, S012–S014, S057–S061 | retrieval precision A/B | no gain vs frequency-only |
+| PPR diffusion (renamed from fan) | ADAPT | RUNE_APPLICABLE | S050–S052 | bridge-node recall | activation ≈ similarity |
+| Outcome-gated trace storage | EXPERIMENT | RUNE_HYPOTHESIS | S027, S056 | promotion-quality A/B | no gain vs reuse-gated |
+| Competition-before-commitment | ADAPT pattern | RUNE_HYPOTHESIS | S031, S020 | scored-selection audit | unused → drop |
+| ECAN budget economics | EXPERIMENT | RUNE_HYPOTHESIS | S044, S043 | quality-per-cost vs plain caps | no gain → drop funds |
+| Typed context spans | ADAPT | RUNE_APPLICABLE | S011 buffers | category-error rate | — |
+| Factor-logged scoring | ADAPT | RUNE_APPLICABLE | S042, S064 | factor ablation | inseparable → single score |
+| Bidirectional promotion loop | ADAPT | RUNE_HYPOTHESIS | S040/S041 | demotion + shaping effect | — |
+| Rejections (noise, engines, cycles, broadcast, framings) | REJECT | — | S060–S061, S053, adversarial §7/§4/§6 | — | — |
+
+## Appendix E — Experiments and falsifiers
+
+| Experiment | Falsifier/kill | Gate |
+|---|---|---|
+| Retrieval precision split-signal vs scalar | no precision gain → collapse signals | T1 scoring design |
+| Promotion-quality gated vs reuse-gated | equal retraction rates → simple gate | procedural promotion |
+| Bridge-node recall (associative PPR) | activation ≈ similarity-only → drop links | associative hypothesis |
+| Impasse-event consumption | events never consumed → remove taxonomy | failure learning |
+| ECAN quality-per-cost vs plain caps | no gain → drop funds ledger | warm-set allocation |
+| Factor ablation (retrieval scoring) | factors inseparable → single score | scoring hygiene |
+| Starvation watch (quantum) | no starvation → don't build quantum | task runtime |
+
+## Appendix F — Remaining unknowns
+
+| Unknown | Resolution path |
+|---|---|
+| Per-occurrence history cost/benefit at Rune scale | benchmark once implemented |
+| Impasse-event taxonomy stability | trial in failure-learning design |
+| ECAN rent pricing | tune on Rune traces if adopted |
+| Factor-weight tuning method | learning-to-rank literature (S064) |
+| #382 procedural-trigger interplay | cross-domain lead; do not preempt |
+
+## Appendix G — Anchor-source bibliography
+
+| Anchor | Citation | Read depth | Claim carried |
+|---|---|---|---|
+| A | Anderson, J.R., Bothell, D., Byrne, M.D., Douglass, S., Lebiere, C. & Qin, Y. (2004). An integrated theory of the mind. *Psychological Review* 111(4):1036-1060 | FULL_PAPER (author preprint; published version paywalled) | canonical B_i = ln(Σ t_j^−d) + β_i VERBATIM; d=.5 is community convention ('emerged as the default'); τ/F 'varied substantially' — identifiability caveat in-source |
+| B | Anderson, J.R. & Schooler, L.J. (1991). Reflections of the environment in memory. *Psychological Science* 2(6):396-408 | FULL_PAPER (companion conference version, OCR'd) | environmental statistics parallel memory; verbatim: 'not too concerned with whether... truly power functions' — transfers the SIGNALS, not the form |
+| C | Heathcote, A., Brown, S. & Mewhort, D.J.K. (2000). The power law repealed. *Psychonomic Bulletin & Review* 7(2):185-207 | FULL_PAPER (bronze OA) | exponential better in ALL unaveraged practice-RT sets; SCOPE: practice-RT only — does NOT rule out ACT's decay-trace mechanism; Wixted & Ebbesen 1997: power law holds for individuals in forgetting |
+| D | Tambe, M., Newell, A. & Rosenbloom, P.S. (1990). The problem of expensive chunks and its solution by restricting expressiveness. *Machine Learning* 5:299-348 | FULL_PAPER (bronze OA) | match cost measured; NP-hard worst case; remedy = restrict expressiveness (NOT disable); explicitly contrasts with Minton cost-benefit |
+| E | Laird, J.E. (2022). An Analysis and Comparison of ACT-R and Soar. arXiv:2201.09305 | FULL_PAPER (open arXiv) | 'similar... details differ' (events contributing to activation; spreading depth); Laird 2012 book NOT ACCESSED — characterized via 2022 |
+| F | Baars, B.J. (1988). *A Cognitive Theory of Consciousness*. CUP (archived scan, ch.10 OCR'd) + Franklin et al. 2013 (LIDA conceptual commitments, JAGI) | FULL_PAPER | broadcast = availability; learning is 1-of-9 functions, hedged; 'learn on broadcast' = LIDA's Conscious Learning Hypothesis (stipulated, no empirical mechanism support) |
+| G | Kotseruba, I. & Tsotsos, J.K. (2020). 40 years of cognitive architectures. *Artificial Intelligence* 262:78-124 (arXiv v3 read in full) | FULL_PAPER (preprint) | 84 architectures (49 active), 900+ projects; 'disjointed and sparse' evaluation; ~1/3 release code |
+| H | Iklé, M., Pitt, J., Goertzel, B. & Sellman, G. (2009). Economic Attention Networks. AGI-09, DOI 10.2991/agi.2009.19 + archived OpenCogPrime design page | FULL_PAPER (AGI-09) + DOCUMENTATION_AUDITED (archived page) | STI/LTI conserved currencies, rent/wages/spreading verified; NO deployment-grade evaluation exists; original wiki is dead (404) |
+
+## Appendix H — Concise bibliography (all load-bearing sources)
+
+See `02_outward/sources.md` for the complete 44-entry ledger with per-source detail. Anchor entries reproduced in Appendix G. Remaining load-bearing: Newell 1990 (UTC); Collins & Loftus 1975 (spreading origin); Haveliwala 2002 (PPR); Andersen/Chung/Lang 2006 (local PPR); Roberts & Pashler 2000 (fit ≠ evidence); Pitt/Myung/Zhang 2002 (model selection); Heathcote/Evans/Anderson-R.B./Murre-Chessa (power-law family); Maynez/SSGM/Voyager/LTR (software substitutes); Eugster 2003 / Hohpe & Woolf 2003 (pub/sub); Sun 2005/2016 (CLARION); Rosenbloom 2025 (Sigma eval); Minton 1988 (EBL utility); Squire & Dede 2015 (memory systems context); Henke 2010 (processing modes, via #377).
+
+## Appendix I — Evidence-level legend
+
+| Level | Meaning |
+|---|---|
+| EXTERNALLY_ESTABLISHED | Strong evidence supports the mechanism in its original field |
+| RUNE_APPLICABLE | Evidence supports that Rune has the same underlying problem or constraints |
+| RUNE_HYPOTHESIS | Plausibly useful to Rune; not demonstrated on Rune |
+| RUNE_EXPERIMENTALLY_SUPPORTED | A controlled Rune-specific experiment supports it |
+| RUNE_PROVEN | Production/runtime evidence in Rune |
+
+## Appendix J — Finding-type distinction
+
+Every claim in this document is one of exactly four kinds, and the kind is always stated or inferable from context:
+
+1. **External finding** — what the cognitive-architecture/human literature establishes (e.g. "ACT-R's base-level equation is B_i = ln(Σ t_j^−d) + β_i"). These are never Rune claims.
+2. **Software principle** — a general engineering principle the finding informs (e.g. "recency+frequency are separable retrieval signals"). These carry no Rune commitment until (3).
+3. **Rune applicability** — evidence that Rune has the same underlying problem (e.g. "wash_count is frequency-only and unvalidated"). Verified against Rune state, not literature.
+4. **Rune prescription** — what Rune should do (e.g. "split-signal scoring with benchmark-chosen form"). Always carries a classification, an experiment, and a rejection criterion. Prescriptions are the ONLY items with T1 impact.
+
+An external finding never silently becomes a Rune prescription; the chain is explicit in 03_applicability.md.
@@ -0,0 +1,202 @@
+{
+ "note": "Narrow, falsifiable claims. external_status vs rune_status kept separate per R1 standard.",
+ "claims": [
+  {
+   "claim_id": "CA-R1-C001",
+   "claim": "ACT-R base-level activation equation is B_i = ln(sum_j t_j^-d) + beta_i, NOT R0's B_i = sum t_j^-d",
+   "evidence_for": [
+    "S012 Anderson et al. 2004",
+    "S013 Anderson 2007",
+    "S014 ACT-R 7 manual"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED",
+   "rune_status": "RUNE_APPLICABLE (correction only)",
+   "architecture_impact": "LOW",
+   "falsifier": "none \u2014 textual verification",
+   "experiment_required": false,
+   "classification": "CONFIRMED-CORRECTED (R0 notation fixed)"
+  },
+  {
+   "claim_id": "CA-R1-C002",
+   "claim": "Recency+frequency signals (not the power-law form) are the transferable content of base-level activation",
+   "evidence_for": [
+    "S010 Anderson & Schooler 1991 (environmental statistics)",
+    "adversarial concession \u00a72/\u00a710",
+    "S010 Anderson & Schooler 1991 verbatim: 'not too concerned with whether the memory or environmental functions are truly power functions' \u2014 signals not form, from the source itself"
+   ],
+   "evidence_against": [
+    "S057/S058 exponentials fit as well or better",
+    "S059 power laws emerge from aggregation",
+    "S060/S061 free-parameter fits ~ zero evidence"
+   ],
+   "external_status": "EXTERNALLY_ESTABLISHED (signals); CONTESTED (form)",
+   "rune_status": "RUNE_APPLICABLE",
+   "architecture_impact": "MEDIUM",
+   "falsifier": "F1: if recency+frequency scoring shows no retrieval gain vs frequency-only on Rune traces, drop",
+   "experiment_required": true,
+   "classification": "ADAPT (signals, not equation)"
+  },
+  {
+   "claim_id": "CA-R1-C003",
+   "claim": "Soar SMEM adopted ACT-R's base-level scheme approximately, not byte-identically",
+   "evidence_for": [
+    "S021 Laird 2012",
+    "S022 Laird 2022 comparison",
+    "S024 Jones 2016 lazy spreading"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED",
+   "rune_status": "RUNE_APPLICABLE (cross-architecture validation of the SIGNAL)",
+   "architecture_impact": "LOW",
+   "falsifier": "none",
+   "experiment_required": false,
+   "classification": "CONFIRMED (approximate adoption)"
+  },
+  {
+   "claim_id": "CA-R1-C004",
+   "claim": "Fan-effect BRANDING for graph normalization is a category error; diffusion math transfers legitimately",
+   "evidence_for": [
+    "S016 Anderson 1974 (sentence verification only)",
+    "S050 Collins & Loftus 1975 \u2192 S051/S052 PPR lineage"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED",
+   "rune_status": "RUNE_APPLICABLE",
+   "architecture_impact": "MEDIUM",
+   "falsifier": "F2: none plausible \u2014 rename and keep mechanism",
+   "experiment_required": false,
+   "classification": "REFINED (rename to PPR; keep hub normalization + inhibition guards)"
+  },
+  {
+   "claim_id": "CA-R1-C005",
+   "claim": "Soar chunking as pedigree does not justify impasse-driven promotion; outcome-gated trace storage does",
+   "evidence_for": [
+    "S027 expensive chunks (measured match cost; remedy = restrict expressiveness, NOT disable); S025 chunking fenced",
+    "S056 Minton utility problem",
+    "S065 Voyager cheaper substitute"
+   ],
+   "evidence_against": [
+    "S020 chunking works when gated"
+   ],
+   "external_status": "EXTERNALLY_ESTABLISHED (utility problem)",
+   "rune_status": "RUNE_HYPOTHESIS",
+   "architecture_impact": "MEDIUM",
+   "falsifier": "F3: if stored failure\u2192repair traces show no promotion-quality gain vs reuse-count gating, drop",
+   "experiment_required": true,
+   "classification": "REFINED (store data traces, not compiled productions; gate on verified failure)"
+  },
+  {
+   "claim_id": "CA-R1-C006",
+   "claim": "Global Workspace broadcast adds nothing over a typed event bus for Rune",
+   "evidence_for": [
+    "S030 Baars 1988 ch.10: broadcast = availability + 1-of-9 learning functions (hedged); 'learn on broadcast' is LIDA's Conscious Learning Hypothesis (stipulated, no empirical support beyond plausibility)",
+    "S054/S055 pub/sub + event-driven beat workspaces",
+    "S066/S067 GWT empirically contested"
+   ],
+   "evidence_against": [
+    "competition-before-commitment is usable arbitration"
+   ],
+   "external_status": "CONTESTED (neuroscience); EXTERNALLY_ESTABLISHED (pub/sub engineering)",
+   "rune_status": "RUNE_APPLICABLE",
+   "architecture_impact": "LOW",
+   "falsifier": "F4: if a broadcast-to-all design shows measurable coordination gain vs typed bus on Rune workloads, revisit",
+   "experiment_required": false,
+   "classification": "REJECT GWT justification; ADAPT competition-before-commitment"
+  },
+  {
+   "claim_id": "CA-R1-C007",
+   "claim": "Fact-vs-skill-file separation needs no architecture theory",
+   "evidence_for": [
+    "adversarial #5 (blast-radius justification suffices)",
+    "S003-adjacent Henke processing-modes contest"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED (engineering)",
+   "rune_status": "RUNE_APPLICABLE",
+   "architecture_impact": "LOW",
+   "falsifier": "none",
+   "experiment_required": false,
+   "classification": "CONFIRMED (design stands; citation dropped)"
+  },
+  {
+   "claim_id": "CA-R1-C008",
+   "claim": "Deterministic/LLM split is an ops boundary, not a symbolic/subsymbolic seam",
+   "evidence_for": [
+    "S053 Kotseruba & Tsotsos (hybrid fragmentation)",
+    "S042 Sigma research-vehicle status"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED",
+   "rune_status": "RUNE_APPLICABLE",
+   "architecture_impact": "LOW",
+   "falsifier": "none",
+   "experiment_required": false,
+   "classification": "REFINED (keep split; drop framing)"
+  },
+  {
+   "claim_id": "CA-R1-C009",
+   "claim": "Fixed cognitive cycles are unjustified for async agents; event-driven with bounded quantum covers the properties",
+   "evidence_for": [
+    "adversarial #7 (this pass found no production-grade evidence of biologically timed fixed cognitive cycles in modern agent runtimes)",
+    "S054/S055 async wins"
+   ],
+   "evidence_against": [
+    "cycles provide bounded work/fairness (accepted; covered by quantum)"
+   ],
+   "external_status": "EXTERNALLY_ESTABLISHED",
+   "rune_status": "RUNE_PROVEN (event-driven continuation already PROVEN)",
+   "architecture_impact": "LOW",
+   "falsifier": "F5: if bounded-quantum event loop shows starvation/fairness failures, revisit cycle discipline",
+   "experiment_required": false,
+   "classification": "REJECT cycles; ADAPT bounded-quantum event loop"
+  },
+  {
+   "claim_id": "CA-R1-C010",
+   "claim": "ECAN attention economics transfers to Warm-Neuron resource allocation",
+   "evidence_for": [
+    "S044 ECAN spec (STI/LTI, rent, budget)"
+   ],
+   "evidence_against": [
+    "combinatorial explosion only partially contained; OpenCog maintenance failure (S043)"
+   ],
+   "external_status": "EXTERNALLY_ESTABLISHED (as spec); CONTESTED (as scaled system)",
+   "rune_status": "RUNE_HYPOTHESIS",
+   "architecture_impact": "MEDIUM",
+   "falsifier": "if budget-capped importance shows no allocation gain vs simple decay on Rune traces, drop",
+   "experiment_required": true,
+   "classification": "EXPERIMENT"
+  },
+  {
+   "claim_id": "CA-R1-C011",
+   "claim": "Activation must be split per function (ordering vs residency vs inclusion vs decay vs selection), not one scalar",
+   "evidence_for": [
+    "outward \u00a7I-K (ACT-R conflates 1/3/4; Soar splits utility; ECAN splits STI/LTI)"
+   ],
+   "evidence_against": [],
+   "external_status": "EXTERNALLY_ESTABLISHED",
+   "rune_status": "RUNE_APPLICABLE",
+   "architecture_impact": "MEDIUM",
+   "falsifier": "if single-scalar scoring matches split-signal scoring on retrieval benchmarks, collapse",
+   "experiment_required": true,
+   "classification": "ADAPT"
+  },
+  {
+   "claim_id": "CA-R1-C012",
+   "claim": "Cognitive architectures contribute problem names, not solutions, to Rune",
+   "evidence_for": [
+    "S053 fragmentation/no shared benchmarks",
+    "S065/S062/S063/S064 cheaper modern substitutes with benchmarks"
+   ],
+   "evidence_against": [
+    "S010 rational basis is a genuine solution-shaped insight"
+   ],
+   "external_status": "EXTERNALLY_ESTABLISHED",
+   "rune_status": "RUNE_APPLICABLE (process rule)",
+   "architecture_impact": "LOW",
+   "falsifier": "none \u2014 process discipline",
+   "experiment_required": false,
+   "classification": "ADAPT (as R1 process rule: equation + failure guard or just the word)"
+  }
+ ]
+}
@@ -0,0 +1,47 @@
+# Evidence Saturation — #378 Cognitive Architectures (R1)
+
+_Proof that the research reached reasonable evidence saturation. Per tranche mandate §4: saturation passes only if later high-quality searches predominantly repeat known mechanisms/criticisms rather than introducing new architecture-changing mechanisms or contradictions._
+
+## Search branches executed
+
+| Branch | Source families searched | Schools discovered | Late-stage discoveries | Late-stage effect | Saturation verdict |
+|---|---|---|---|---|---|
+| ACT-R | Anderson 1974/1990/2004/2007; ACT-R 4/5/6/7 manuals; Lebiere; ACT-R tutorials; BOLD/fMRI fitting papers; pyactr/jACT-R | Rational analysis (Anderson & Schooler); subsymbolic vs symbolic layers | Optimized base-level approximations (reference manual); parameter-identifiability critiques (Roberts & Pashler 2000; Pitt/Myung/Zhang 2002) | Added the identifiability criticism (changed confidence, not conclusion) | SATURATED — late finds refined, did not overturn |
+| Power-law critique | Heathcote 2000; Evans 2018; Anderson R.B. 2001; Murre & Chessa 2011; Roberts & Pashler 2000 | Exponential-law school; emergence-artefact school | Murre & Chessa aggregation argument | Strengthened "form does not transfer" | SATURATED |
+| Soar | Laird 1987/2012/2022×2; Tambe 1990; Minton 1988; Jones 2016; deployed systems (TacAir, NTD, ICT); SML/JSoar | Problem-space school; chunking-EBL lineage; RL-pruning school | Tambe expressiveness-restriction detail; chunking-disabled deployments | Confirmed utility problem (H4 refinement) | SATURATED |
+| LIDA/GWT | Baars 1988; Franklin 2007/2009; Dehaene & Changeux 2011; Shanahan 2006; blackboard (Hearsay-II); GWT critics (Doerig 2020; Storm 2024; Tsuchiya 2015; Cogitate 2023-25) | GNW neuroscience; functionalist GWT; blackboard engineering lineage; consciousness-falsifiability critics | Cogitate adversarial collaboration unresolved; pub/sub superiority (Eugster; Hohpe & Woolf) | Confirmed broadcast rejection (H5) | SATURATED |
+| CLARION | Sun 2002/2005/2016; pyClarion; SRT/grammar/Tower-of-Hanoi validation papers | Dual-representation school | pyClarion fragmentation; lab-task-only validation | Refined (borrow loop, skip ontology) | SATURATED |
+| Sigma | Rosenbloom Sigma/I/ITSEC series; 2025 ACS evaluation arXiv:2510.03453 | Factor-graph unification school | 2025 evaluation paper | Refined (factor-combination idea only) | SATURATED |
+| OpenCog/Hyperon | Goertzel 2008; PLN book 2009; ECAN wiki; MOSES; Hyperon/MeTTa; SingularityNET history | Attention-economy school; probabilistic-logic school | ECAN partial-containment admission; funding-history detail | Refined (borrow ECAN economics; reject substrate) | SATURATED |
+| Other architectures | EPIC; ICARUS; SAL; Nengo/SPA; Leabra; MicroPsi; 4CAPS; blackboard; BDI; HTM; Copycat; CHREST | ~10 schools catalogued | None new (exclusions recorded with reasons) | No effect | SATURATED |
+| Activation/working state | Collins & Loftus 1975; ACT-R spreading; Soar SMEM; ECAN; LIDA attention; LRU-K/ARC/BM25/LTR | Cognitive-spreading school; cache/IR school | LRU-K/ARC/BM25 as cheaper substitutes (adversarial §10) | Changed mechanism (split signals; benchmark form) | SATURATED — late find WAS incorporated |
+| Procedural learning | Minton 1988; Tambe 1990; Voyager; CLARION RER; ACT-R compilation | Compilation/EBL school; utility school; skill-library school | Voyager as cheaper substitute | Refined (data-traces variant) | SATURATED |
+| Cycles | ACT-R 50ms; Soar decision cycle; LIDA 10Hz; async/event-driven literature | Fixed-cycle school; event-driven school | Zero-production-runtime observation (adversarial #7) | Confirmed rejection (strongest finding) | SATURATED |
+| Scaling/failure | Kotseruba & Tsotsos 2020 (~80 architectures); OpenCog history; utility-problem literature | Fragmentation diagnosis | 2020 survey's benchmark-absence finding | Confirmed process rule (problem-names-not-solutions) | SATURATED |
+| Modern-agent transfer | ReAct; Voyager; LangChain/AutoGPT; agent-memory systems (MemGPT, Synapse — cross-ref #379-bound but noted); LTR/cross-encoder reranking | Leverage-over-theory school | Cheaper-substitute table (adversarial §10) | Confirmed | SATURATED |
+| Criticism/negative evidence | Identifiability; replication; version-churn; WEIRD populations; benchmark absence | — | Version-churn exhibits (ACT-R buffers; Soar semantics; LIDA constructs) | Confirmed process rule | SATURATED |
+
+## Late-stage saturation check (the actual test)
+
+The last searches performed (power-law critiques; identifiability; pub/sub-vs-workspace; zero-production-cycle verification; cheaper-substitute hunt) produced:
+
+- **New mechanism introduced:** 0 (cheaper-substitute table = known mechanisms with better engineering, not new mechanisms)
+- **Serious contrary evidence introduced:** 2 (power-law repeal family; GWT falsifiability) — both INCORPORATED: they changed the conclusion from "adopt the equation" to "adopt the signals, benchmark the form," and from "GWT justifies the broker" to "broker on engineering grounds only"
+- **Rune conclusion changed:** 2 (C002 narrowed from equation-adoption to signal-adoption; C006 narrowed from pattern-adoption to rejection of GWT justification) — both changes fully incorporated into claims.json, applicability, synthesis, qualification, and R0 reconciliation before closure
+- **Predominantly repetition/minor refinement:** yes — the bulk of late searches returned the same canonical sources (Anderson/Laird/Baars/Sun/Rosenbloom) and the same criticisms (identifiability, fragmentation, utility problem)
+
+**VERDICT: EVIDENCE SATURATION REACHED.** No unresolved architecture-changing school or mechanism remains; the two late-stage conclusion changes are incorporated in all downstream artifacts.
+
+
+## Anchor-depth verification addendum (post-repair)
+
+After the repair tranche's anchor verification (primary sources actually read):
+- 5 anchors read as FULL_PAPER (Anderson 2004 author preprint; Anderson & Schooler 1991 companion OCR; Heathcote 2000 bronze OA; Tambe 1990 bronze OA; Baars 1988 archived scan ch.10 + LIDA 2013)
+- 2 DOCUMENTATION_AUDITED (Laird 2022 arXiv full; archived OpenCogPrime ECAN page)
+- 1 FULL_PAPER preprint (Kotseruba & Tsotsos arXiv v3)
+- 2 citation errors found and fixed: Heathcote venue (Psychon Bull Rev, not Psych Review); ECAN wiki dead (replaced with AGI-09 paper + archived page)
+- 1 scope correction: Heathcote's critique is practice-RT curves only — the paper explicitly does NOT rule out ACT's decay-trace mechanism, and Wixted & Ebbesen 1997 found the power law HOLDS for individuals in forgetting. The form-transfer critique is now correctly scoped.
+- 1 attribution correction: 'learn on broadcast' is LIDA's Conscious Learning Hypothesis (stipulated), not Baars's law — Baars claims availability + hedged learning-as-one-function.
+- d = 0.5 verified as community convention ('has emerged as the default'), with the identifiability caveat present in Anderson 2004 itself.
+
+No conclusion overturned by the deeper reading; two characterizations refined (Heathcote scope; Baars/LIDA attribution), both incorporated in sources.md, claims.json, and the master's Appendix G.
@@ -0,0 +1,43 @@
+# Mutation Audit — #378 R1 (repair tranche, and original R1 pass)
+
+_Required by repair gate R7. This replaces the hardcoded `G31: no production build — True` from the original pass with actual evidence._
+
+## Production source code
+- `git status --porcelain -- src config tests scripts` in the harness checkout: **EMPTY** — zero changes to production source, config, tests, or scripts.
+- No commits made to production branches during the #378 R1 pass or this repair pass.
+- The only repo-tree files touched were under `research/memory-atlas-r1/` (research artifacts) — enumerated below.
+
+## Memory schema
+- `entries` table columns **identical** to the T0 baseline (26 columns, diff = none, verified programmatically).
+- 35 tables in memory.db (all pre-existing; no new tables created).
+- No migration files created or modified.
+
+## Canonical Memory data
+- Entry count: 924 (unchanged since T0.1; the newest entry predates the T0.1 errata tranche).
+- **No canonical Memory entries written by the #378 R1 pass or this repair pass.** Zero `memory_store.py add` / `consolidate` invocations during this domain (deliberately: no-build boundary forbade completion-summary writes).
+
+## Production venv / environment
+- numpy and qdrant-client present in `/home/nick/.venvs/memory/` — **both installed in the PRIOR research-closure tranche** (disclosed in that tranche's errata, E1). **Nothing new installed** by the #378 R1 pass or this repair.
+- No other environment mutations.
+
+## Runtime / retrieval configuration
+- Zero changes to retrieval configuration, ranking functions, or Memory runtime behavior.
+- The Fact-vs-Skill-file split, trust lifecycle, supersession, and hybrid recall all remain at their pre-tranche state.
+
+## Gitea
+- #378 reopened (for R1), closed (R1_QUALIFIED) with domain-specific close-out comment; parent #375 registry updated; domain-specific closeout comment. All enumerated in the Gitea operations log. This repair: one addendum comment (posted at repair completion).
+
+## Mirror
+- `~/Desktop/research/`: whole canonical folder mirrored (`memory-atlas-r1-C-COG/`), domain_registry.md, compendium updated. Enumerated in artifact list.
+
+## Research-only file mutations (permitted, enumerated)
+- research/memory-atlas-r1/C-cognitive-architectures/: 00_issue_scope.md, 01_inward.md, 02_outward/ (21 files + sources.md), 03_applicability.md, 04_innovation.md, 05_adversarial.md, 06_synthesis.md, 07_qualification.md, claims.json, r0_reconciliation.md, C_COGNITIVE_ARCHITECTURES_R1_MASTER.md, evidence_saturation.md, mutation_audit.md
+- research/memory-atlas-r1/domain_registry.md (status update)
+- research/MEMORY_RESEARCH_MASTER_COMPENDIUM.md (regenerated projection)
+
+## Prior-tranche disclosures (not this tranche; for completeness)
+- qdrant-client + numpy venv installs: prior research-closure tranche (errata E1).
+- Canonical Memory summary writes: prior tranches (topology, async proofs, research closure — errata E1).
+
+## VERDICT
+**PRODUCTION_MUTATION = NONE** (for this tranche and the original #378 R1 pass). Prior-tranche mutations remain disclosed in their errata and do not recur here.
@@ -0,0 +1,30 @@
+# R0 → R1 Reconciliation (#378 Cognitive Architectures) — REPAIR D: one-to-one
+
+_Original R0 source: memory-atlas/families/cognitive_architectures.md (ten hypotheses H1–H10).
+Each hypothesis has exactly one row and exactly one verdict. Aggregate totals sum to 10.
+H7 correction incorporates #376 where stated in the R0 itself._
+
+| # | Original R0 claim | R1 finding | Status | Evidence | Reason for change | Resulting Rune implication |
+|---|---|---|---|---|---|---|
+| H1 | ACT-R base-level activation (B_i ≈ Σ t_j^-d) should inform deterministic Warm Neuron activation | Equation CORRECTED to B_i = ln(Σ t_j^−d) + β_i (S012–S014); the power-law FORM is contested (S057–S061: exponential fits as well; power laws emerge from aggregation; d≈0.5 is convention); the recency+frequency SIGNALS survive (S010 rational basis; adversarial concession) | **REFINED** | S010, S012–S014, S057–S061, adversarial §1 | R0 cited the equation without verifying form or identifiability; signals ≠ equation | Split-signal scoring with benchmark-chosen form (LRU-K/ARC/BM25 discipline), never the equation as-is |
+| H2 | Recency + frequency should replace wash_count-style frequency-only behavior | CONFIRMED and strengthened: wash_count does triple duty (truth/activation/promotion) on one counter — 01_inward finding; frequency-only is NFU-like (child #385) | **CONFIRMED** | 01_inward, S045, S062/S063 | Direct evidence from store inspection + cache literature | Split-signal activation record (innovation I1): separate recency/frequency fields per entry |
+| H3 | ACT-R/Soar fan normalization supports graph/PPR anti-hub weighting | Mechanism (hub damping) needed; BRANDING wrong: Anderson 1974 fan is sentence-verification RT, not graph math; diffusion math transfers via Collins & Loftus → Page/Brin lineage | **REFINED** | S016, S050–S052, adversarial §2 | Three distinct maths conflated: fan RT / degree normalization / PPR | Rename to PPR; keep degree damping + top-k + trust floor; drop "fan" vocabulary |
+| H4 | Soar chunking supports impasse → resolution → procedure promotion | Soar pedigree WEAKENED (expensive chunks S027; chunking fenced/disabled in practice S025); what survives: outcome-gated TRACE storage (data traces, not compiled productions — dodges match-cost explosion) | **REFINED** | S020–S027, adversarial §3 | R0 cited architecture pedigree; utility problem dominates documented practice | Typed impasse events + data-trace storage; promotion gate = verified failure+resolution (deferred to #382 for trigger debate) |
+| H5 | Global Workspace / LIDA suggests selected context → broadcast → learning | **OVERTURNED as architecture mechanism.** Broadcast = pub/sub with coupling; GWT neuroscience contested (S066/S067: falsifiability, adversarial collaboration unresolved); "broadcast causes learning" stipulated not derived. Competition-before-commitment pattern survives | **OVERTURNED** | S030–S032, S054/S055, adversarial #4 | Two independent failures: neuroscience evidence contested; engineering inferior to typed event bus | Event bus + typed subscribers; competition-before-commitment as arbitration pattern; no broadcast justification |
+| H6 | Context inclusion itself might be treated as a consolidation/learning event | Refined to PATTERN-ONLY: learn-on-broadcast is stipulated (LIDA), unvalidated; what survives is competition-before-commitment with LOGGED selection (auditability) | **REFINED** | S031, adversarial #5/#6 | R0 treated a design stipulation as mechanism | Log context selections (audit surface); no learning side-effect without evidence |
+| H7 | Activation noise should be rejected for Rune | CONFIRMED, strengthened: noise is often a free parameter (S060/S061 identifiability); harms auditability; principled-as-sampling but unnecessary for deterministic single-user state | **CONFIRMED** | S060/S061, adversarial §1 | #376/#385 convergence; auditability requirement (req 24) | No noise in any scoring path |
+| H8 | Production-rule engines should be rejected for Rune | CONFIRMED, strengthened: LLM is Rune's procedural system; engine maintenance burden documented (S053 fragmentation survey; S025 chunk-semantics churn; OpenCog/Hyperon maintenance history S043) | **CONFIRMED** | S025, S043, S053, adversarial #6/#8 | Kotseruba & Tsotsos: hybrids proliferate vocabulary, not results; version-churn exhibits | Skill FILES (versioned executables with lifecycle), never a rule engine |
+| H9 | Codelet concurrency / strict cognitive-cycle timing should be rejected | CONFIRMED as the strongest finding: ZERO production agent runtimes use fixed cognitive cycles (adversarial #7: observed zero); async event-driven beats cycles on throughput/fault-isolation (S054/S055). Bounded-quantum is the legitimate child if fairness fails | **CONFIRMED** | S054/S055, adversarial #7 | Production evidence (observed zero) + engineering comparison | Event-driven with bounded-quantum hedge (FUTURE, build only on observed starvation) |
+| H10 | Cognitive architectures validate Rune's existing separation of memory / active context / procedural execution | Refined: the separation stands but needs NO architecture theory — it is a file-format/permissioning/blast-radius decision (adversarial #5: citation is decoration); symbolic/subsymbolic framing actively rejected (ops boundary, not cognitive seam — adversarial #6) | **REFINED** | adversarial #5/#6, S053, S042 | Separation is justified by engineering (blast radius, file formats); architecture citations add decoration | Keep fact-vs-skill-file split on engineering grounds; drop symbolic/subsymbolic framing; do not cite Squire/Henke for file layout |
+
+## Aggregate totals (sum = 10)
+
+| Status | Count | Items |
+|---|---|---|
+| CONFIRMED | 3 | H2, H7, H9 |
+| REFINED | 5 | H1, H3, H4, H6, H10 |
+| OVERTURNED | 1 | H5 |
+| STILL_UNKNOWN | 0 | — |
+| **TOTAL** | **10** | H1–H10 |
+
+Note on the previous master's grouped counts: the prior aggregate (CONFIRMED 4 · REFINED 5 · OVERTURNED 1) was correct in totals but grouped H7/H8/H9 into one row and did not show H1's equation correction as a distinct refinement. This table supersedes it; r0_reconciliation.md in the stage folder is updated to match.
@@ -0,0 +1,59 @@
+# Sources Ledger (#378 R1)
+
+Columns: source_id | title | authors | year | venue | DOI/arXiv/URL | peer-reviewed? | architecture | primary/review/critique | version | research question | mechanism | methodology | evaluation population/task | quantitative result | replication/status | criticism | limitations | claims supported | claims contradicted | Rune relevance | confidence | read depth
+
+## Foundations
+| S001 | Unified Theories of Cognition | Newell 1990 | Harvard UP (book) | n/a | — | theory | — | fixed architecture + variable knowledge | — | — | live doctrine | — | architecture thesis | HIGH | RELEVANT_SECTIONS |
+| S002 | Logic Theorist/GPS; production systems lineage | Newell/Simon/Shaw | — | yes | — | historical | — | production-system origins | — | — | established | — | lineage | MEDIUM | ABSTRACT |
+| S003 | PDP / connectionism critique of symbols | Rumelhart/McClelland 1986 | MIT Press | yes | — | theory | — | brittle symbols | — | — | established | — | hybrid motivation | MEDIUM | ABSTRACT |
+
+## ACT-R
+| S010 | Reflections of the environment in memory (+ companion conf version, OCR'd) | Anderson & Schooler 1991 | Psych Sci 2:396-408, DOI 10.1111/j.1467-9280.1991.tb00174.x | yes | ACT-R | primary | — | environmental statistics | 3 demand sources (child speech/headlines/email) | frequency/recency/spacing parallel 4 lab phenomena | replicated | paper: 'not too concerned whether... truly power functions' — SIGNALS not form | rational basis for signals; form explicitly not claimed | HIGH | FULL_PAPER |
+| S011 | The Atomic Components of Thought | Anderson & Lebiere 1998 | Erlbaum (book) | yes | ACT-R | primary | ACT-R 4 | declarative formulation | — | — | live | — | ACT-R 4 baseline | HIGH | RELEVANT_SECTIONS |
+| S012 | An integrated theory of the mind | Anderson/Bothell/Byrne/Douglass/Lebiere/Qin 2004 | Psych Rev 111(4):1036-1060, DOI 10.1037/0033-295X.111.4.1036 | yes | ACT-R | primary (author preprint, full) | — | unified equations | latency/accuracy/BOLD | canonical B_i equation VERBATIM; d=.5 'emerged as default' = convention; τ/F 'varied substantially' (identifiability caveat in-source) | live, contested periphery | identifiability admitted in-source | **canonical B_i = ln(Σt^-d)+β** | HIGH | FULL_PAPER |
+| S013 | How Can the Human Mind Occur... | Anderson 2007 | Oxford UP | yes | ACT-R | primary | — | consolidated equations + BOLD | — | — | live | — | equation consolidation | HIGH | RELEVANT_SECTIONS |
+| S014 | ACT-R 7 reference manual | CMU | manual | n/a | ACT-R | spec | 7.x | implementation truth | — | — | maintained | — | optimized approximations | HIGH | DOCUMENTATION_AUDITED |
+| S015 | The fan effect | Anderson & Reder 1999 | JEP:General 128:186-197 | yes | ACT-R | primary | — | fan replication | sentence verification | retrieval slows with fan | replicated within paradigm | paradigm scope | S − ln(fan) form | HIGH | RELEVANT_SECTIONS |
+| S016 | Retrieval of propositional information | Anderson 1974 | Cog Psych | yes | ACT-R lineage | primary | — | fan origin | memorized sentences | RT slope with fan | replicated | sentence-verification only | fan ≠ graph math | HIGH | ABSTRACT |
+
+## Soar
+| S020 | Soar: architecture for general intelligence | Laird/Newell/Rosenbloom 1987 | Artif Intell 33:1-64 | yes | Soar | primary | — | problem spaces, impasses, chunking | — | — | foundational | — | architecture thesis | HIGH | RELEVANT_SECTIONS |
+| S021 | The Soar Cognitive Architecture | Laird 2012 | MIT Press | yes | Soar | primary | — | full memory/learning spec | — | — | maintained | — | SMEM ≈ ACT-R adoption; chunk mitigations | MEDIUM | NOT ACCESSED (no open copy; characterized via S022) |
+| S022 | Analysis and comparison of ACT-R and Soar | Laird 2022 | arXiv:2201.09305 | no | both | review | — | direct comparison | — | approximate adoption confirmed | live | — | best comparison source | HIGH | RELEVANT_SECTIONS |
+| S023 | Introduction to Soar v9.6 | Laird 2022 | arXiv:2205.03854 | no | Soar | spec | 9.6 | decision cycle, RL, EPMEM status | — | — | maintained | — | current status | MEDIUM | ABSTRACT |
+| S024 | Efficient spreading via lazy evaluation | Jones et al. 2016 | ICCM | yes | Soar | primary | — | spreading engineering | — | — | live | — | spreading reality vs theory | MEDIUM | ABSTRACT |
+| S025 | Extending the Soar cognitive architecture | Laird 2008 | AI Magazine | yes | Soar | review | — | chunking fenced; semantic/episodic added | — | — | live | — | chunking semantics changed | HIGH | ABSTRACT |
+| S026 | Chunking 1986 original | Laird/Rosenbloom/Newell 1986 | Machine Learning | yes | Soar | primary | — | EBL chunking | — | — | superseded | — | 1986 ≠ today | MEDIUM | ABSTRACT |
+| S027 | The problem of expensive chunks and its solution by restricting expressiveness | Tambe/Newell/Rosenbloom 1990 | Machine Learning 5:299-348, DOI 10.1023/A:1022603523533 | yes | Soar | primary | — | match-cost measurement; NP-hard worst case; remedy = restrict expressiveness (NOT disable) | Soar/PSM-E token/k-search | — | live finding | explicitly contrasts with Minton cost-benefit | promotion systems must trade expressiveness for match guarantees | HIGH | FULL_PAPER |
+
+## LIDA / GWT
+| S030 | A Cognitive Theory of Consciousness (archived scan; ch.10 §10.2 OCR'd) | Baars 1988 | CUP (book) | yes | GWT | theory | — | global workspace; learning = 1 of 9 functions, novelty-hedged | — | — | contested | acknowledges necessity controversy (Eriksen 1960; Holender 1986) | availability claim; 'learn on broadcast' is LIDA's stipulation, NOT Baars's law | HIGH | FULL_PAPER |
+| S031 | LIDA model; Baars & Franklin 2009 | Franklin et al. | Int J Machine Consciousness | yes | LIDA | primary | — | codelets, coalitions, broadcast, cycles | — | — | no comparative benchmarks | — | competition-before-commit | MEDIUM | RELEVANT_SECTIONS |
+| S032 | Experimental/theoretical GNW; masking critiques | Dehaene & Changeux 2011; Meyen et al. 2022 | Neuron etc. | yes | GNW | mixed | — | ignition evidence | masking paradigms | contested | adversarial collab unresolved | — | do not cite ignition as proof | MEDIUM | ABSTRACT |
+
+## CLARION / Sigma / OpenCog
+| S040 | Explicit/implicit skill learning interaction | Sun et al. 2005 | Psych Rev 112:159-192 | yes | CLARION | primary | — | dual levels | SRT/grammar tasks | synergy effects | lab-tasks only | — | bidirectional loop | MEDIUM | ABSTRACT |
+| S041 | Anatomy of the Mind | Sun 2016 | OUP | yes | CLARION | primary | — | full architecture | — | — | fragmented impl | — | four-box ontology | LOW | ABSTRACT |
+| S042 | Sigma factor-graph unification + 2025 evaluation | Rosenbloom et al.; 2025 ACS arXiv:2510.03453 | mixed | yes | Sigma | primary+eval | — | unification claim | demos | unproven scale | research vehicle | intractability | factor-combination idea only | LOW | ABSTRACT |
+| S043 | Probabilistic Logic Networks; OpenCog framework | Goertzel et al. 2009; Hart & Goertzel 2008 | Springer | yes | OpenCog | primary | — | PLN/AtomSpace | — | — | failed to scale | — | ECAN vs PLN split | MEDIUM | ABSTRACT |
+| S044 | Economic Attention Networks: Associative Memory and Resource Allocation for General Intelligence | Iklé/Pitt/Goertzel/Sellman 2009 | AGI-09, DOI 10.2991/agi.2009.19 | yes | OpenCog | primary | — | STI/LTI as conserved currencies; rent/wages/spreading | math convergence + small associative-memory experiments | — | NO deployment-grade evaluation found; wiki dead (404) | design page flagged 'may be obsolete' | most transferable idea; no scaled validation | HIGH | FULL_PAPER |
+
+## Other / comparative
+| S050 | Spreading-activation theory | Collins & Loftus 1975 | Psych Rev | yes | — | theory | — | semantic spreading | — | — | robust effect | — | diffusion-math ancestor | HIGH | RELEVANT_SECTIONS |
+| S051 | Topic-sensitive PageRank | Haveliwala 2002 | WWW | yes | — | primary | — | PPR | web graph | — | deployed | — | legitimate diffusion math | HIGH | ABSTRACT |
+| S052 | Local graph partitioning | Andersen/Chung/Lang 2006 | FOCS | yes | — | primary | — | local PPR | — | — | live | — | PPR engineering | MEDIUM | ABSTRACT |
+| S053 | 40 years of cognitive architectures (arXiv v3 read; AIJ journal redline not separately accessed) | Kotseruba & Tsotsos 2020 | Artificial Intelligence 262:78-124, DOI 10.1016/j.artint.2019.103219; arXiv:1610.08602v3 | yes | survey | review | — | 84 architectures (49 active), 900+ projects; 'disjointed and sparse' evaluation; ~1/3 release code | — | — | live finding | — | fragmentation is structural, not one named cause | HIGH | FULL_PAPER (preprint) |
+| S054 | Publish/subscribe survey | Eugster et al. 2003 | ACM Comp Surveys | yes | — | survey | — | decoupled dissemination | — | — | live | — | broadcast = pub/sub + coupling | HIGH | ABSTRACT |
+| S055 | Enterprise Integration Patterns | Hohpe & Woolf 2003 | Addison-Wesley | n/a | — | practice | — | event-driven beats workspace | production | — | live | — | typed event bus | HIGH | ABSTRACT |
+| S056 | Minton utility of EBL | Minton 1988 | thesis/Machine Learning | yes | — | primary | — | learned rules cost | — | — | live finding | — | promotion must weigh reuse | HIGH | ABSTRACT |
+| S057 | The power law repealed: exponential law of practice | Heathcote/Brown/Mewhort 2000 | Psychon Bull Rev 7(2):185-207, DOI 10.3758/BF03212979 | yes | critique | primary | — | exponential vs power fits | 40 datasets/7910 series/475 subjects; exponential better in ALL unaveraged sets; APEX best overall | SCOPE: practice-RT curves only — does NOT rule out ACT's decay-trace mechanism for memory; Wixted & Ebbesen 1997: power law HOLDS for individuals in forgetting | live | — | limits form-transfer claim; do not cite vs memory decay | HIGH | FULL_PAPER |
+| S058 | Refining the law of practice | Evans et al. 2018 | Psych Bull Rev | yes | critique | primary | — | exponentials win controlled | — | — | live | — | same | MEDIUM | ABSTRACT |
+| S059 | Power law as emergent property | Anderson R.B. 2001; Murre & Chessa 2011 | Psych Bull Rev | yes | critique | theory | — | aggregation artefact | — | — | live | — | aggregate fits prove nothing per-item | HIGH | ABSTRACT |
+| S060 | How persuasive is a good fit? | Roberts & Pashler 2000 | Psych Review | yes | critique | theory | — | free-parameter fits ~ zero evidence | — | — | live | — | d≈0.5 is convention | HIGH | ABSTRACT |
+| S061 | Model selection complexity | Pitt/Myung/Zhang 2002 | Psych Review | yes | critique | theory | — | complexity-penalized selection reverses naive fits | — | — | live | — | same | MEDIUM | ABSTRACT |
+| S062 | LRU-K page replacement | O'Neil/Cheng/Gagliardi 1993 | VLDB | yes | — | primary | — | recency+frequency, 1-2 params | traces | — | deployed | — | beats decay hybrids | HIGH | ABSTRACT |
+| S063 | ARC self-tuning cache | Megiddo & Modha 2003 | FAST | yes | — | primary | — | adaptive replacement | traces | — | deployed | — | same | HIGH | ABSTRACT |
+| S064 | BM25 and Beyond; Learning to Rank | Robertson & Zaragoza 2009; Liu 2009 | FnTIR | yes | — | review | — | frequency+recency, zero cognitive commitments | — | — | deployed | — | same | HIGH | ABSTRACT |
+| S065 | Voyager skill libraries | Wang et al. 2023 | arXiv:2305.16291 | no | — | primary | — | self-verified trajectories | Minecraft | 3.3× items etc. | live | verifier errors | cheaper chunking substitute | MEDIUM | ABSTRACT |
+| S066 | Consciousness falsifiability criteria | Doerig/Schurger/Herzog 2020 | Consciousness & Cognition | yes | critique | theory | — | GWT fails falsifiability | — | — | live | — | GWT contested | MEDIUM | ABSTRACT |
+| S067 | Multiscale neural theories view | Storm et al. 2024 | Neuron | yes | critique | review | — | GWT/IIT/recurrent splits | — | — | live | — | same | MEDIUM | ABSTRACT |
@@ -0,0 +1,38 @@
+# #379 R1 — 00_issue_scope.md
+
+## Core question
+
+Which modern agent/LLM memory systems and mechanisms genuinely solve problems Rune has, which are useful design patterns to adapt, which should be adopted outright, and which should be rejected — beyond what the R0/R1 closure already established?
+
+## Subquestions
+1. What memory architectures have shipped in production agent systems since the R0 closure?
+2. What mechanisms have emerged that the prior closure missed?
+3. Which prior closure classifications need revision based on changed systems?
+4. How do current systems handle: context paging, memory hierarchy, retrieval, consolidation, forgetting?
+5. What failure modes have production deployments revealed?
+6. What security/poisoning patterns have emerged?
+7. What does the open-source agent-memory ecosystem look like in 2025-2026?
+8. Which systems are production-grade vs research prototypes?
+9. What is the actual state of vector-DB vs file-based vs graph-based memory in deployed agents?
+10. How do agents handle memory across model replacements?
+
+## R0 claims under test (from prior closure + errata)
+H1: Qdrant REJECT (sqlite-vec retained; sub-ms at current scale)
+H2: Graph servers REJECT (SQLite+networkx pattern ADOPT; LadybugDB EXPERIMENT)
+H3: Letta sleep-time compute → nightly consolidation pattern; writer/consumer split
+H4: Graphiti timestamps → world-validity vs record-time separation
+H5: CABLE: similarity links duplicate retrieval; complementary links only
+H6: mem0 retreated from LLM mutation arbitration → additive + hash dedup
+H7: SimpleMem compression duplicates consolidation → REJECT pipeline
+H8: A-MEM evolution as new provisional only
+H9: HippoRAG2 PPR as non-parametric CL
+H10: LangMem profile/collection split
+
+## Falsifiers
+F1: If a production agent memory system demonstrates that vector DBs are necessary at Rune's scale, H1 falls.
+F2: If graph-based memory shows production-quality retrieval gains in real agent deployments, H2 weakens.
+F3: If a production system validates LLM-arbitrated mutation (contra mem0's retreat), H6 falls.
+F4: If destructive consolidation is shown safe and superior to append-only in controlled evaluation, H7 strengthens.
+
+## Stopping criterion
+Evidence saturation for the DELTA only (post-R0-closure changes). R0 conclusions that remain unchallenged by new evidence stay as-is.
@@ -0,0 +1,50 @@
+# #379 R1 — 01_inward.md
+
+## Current Rune memory state (from T0/T0.1 + prior research closure)
+
+### What Rune already has (PROVEN)
+- SQLite canonical store (924 entries, 729 fact + 195 neuron)
+- sqlite-vec dense (256-d) + FTS5 BM25 hybrid recall with RRF fusion
+- Dedupe/supersede gate with receipts
+- Trust lifecycle (provisional/validated/locked) + HMAC audit ledger
+- knowledge.json consolidation output (single inspectable writer)
+- Warm-neuron use history (wash_count, importance, last_accessed)
+- valid_from/valid_until temporal fields
+- Event-driven continuation (async programme FROZEN_ACCEPTED)
+- Task state separate from Memory (auto_tasks.db)
+- behavior.db (success/failure learning, negligible volume)
+- graph.sqlite (dormant since Aug 26)
+
+### What is ABSENT/BROKEN
+- 439/838 active entries unindexed (embedder silent failure — T0.1)
+- No origin column (only source/method)
+- No derived_from edges
+- No trust-gated retrieval
+- No obligation entry kind
+- Consolidation trigger non-deterministic (no cron/timer)
+- 139 orphan vectors
+
+### Prior R0 closure conclusions (verbatim, to test against new evidence)
+- Qdrant REJECT (spike: 0.27ms vs 0.37ms at 559 vectors)
+- sqlite-vec retained; revisit >100k vectors
+- Graph servers (Neo4j/FalkorDB) REJECT
+- SQLite+networkx PPR pattern ADOPT
+- LadybugDB EXPERIMENT
+- Graphiti timestamps ADAPT
+- CABLE: complementary links only
+- Letta: writer/consumer split; sleep-time = nightly consolidation
+- mem0: retreated from LLM arbitration
+- A-MEM evolution as new provisional only
+- SimpleMem REJECT (duplicates consolidation)
+- HippoRAG2 PPR as non-parametric CL
+- LangMem profile/collection split ADAPT
+- Hindsight mental-models ADAPT
+
+### What the R0 closure did NOT cover (delta targets for this domain)
+- MemOS/MemoryOS detailed architecture (only R0 delta noted)
+- Hindsight full system (installed on Rune's PC but unused)
+- Production deployment patterns from ChatGPT/Claude memory features
+- New agent-memory systems 2025-2026
+- Security/poisoning patterns in deployed memory
+- Code/paper divergences in shipped systems
+- Agent-memory evaluation frameworks
@@ -0,0 +1,8 @@
+# #379 R1 — 02_outward/README.md
+
+Populated by deep-research + adversarial actors; verified by controller.
+
+Files: memos_memoryos.md, hindsight_audit.md, production_patterns.md,
+new_systems_2025_2026.md, code_paper_divergences.md, security_updates.md,
+evaluation_frameworks.md, ecosystem_survey.md, convergence_analysis.md,
+sources.md, claims.json
@@ -0,0 +1,6 @@
+# Code/paper divergences
+
+- **Mem0** (65.7k stars): recent releases entirely platform/plugin-side — surface-identity headers, telemetry, plugins for OpenClaw/Pi/OpenCode/DeepSeek. The 2024 paper's open-source LLM-arbitration pipeline still ships but research velocity moved to hosted platform. R0's "retreated" reading CONFIRMS and strengthens.
+- **Letta**: docs now frame MemGPT-style API as "V1 SDK (legacy)"; new Agent SDK makes memory a git-backed MemFS owned by the agent, projected onto machines, with system/ files in-prompt and everything else read-on-demand from file tree; dreaming (background consolidation subagents) is first-class config; archival memory/passages/folders/templates listed under "Deprecated". Genuine divergence: Letta converged from self-editing vector memory toward file/git memory. Bearing: validates Rune's markdown-file memory direction.
+- **Graphiti** (31k stars, v0.30.2): shipped sagas (multi-episode narrative rollup), combined node+edge extraction (one LLM call), decoupled timestamp resolution, Kuzu backend, embedded FalkorDB, README repositioning from "agent memory" to "temporal knowledge graph framework"; Zep's production engine is proprietary. Security: v0.28.2 hardened search filters against Cypher injection (MCP ≤1.0.1 vulnerable) — real CVE-class incident. Neo4j multi-DB routing broken until v0.30.1. Bearing: backend routing bugs are realistic for Rune's graph-store risk register.
+
@@ -0,0 +1,6 @@
+# Architecture convergence
+
+Five-axis convergence visible from shipped code: (1) hybrid multi-strategy read (semantic+BM25+graph+temporal → RRF → rerank) is the default read path (Hindsight, MemOS, Graphiti); (2) background consolidation (Hindsight observations/mental models, Letta dreaming, MemOS skill crystallization, Anthropic's after-session updates); (3) files/wiki as the memory surface (Hindsight knowledge pages→markdown, Letta MemFS+git, Claude /memories, ChatGPT memory summary); (4) evidence retention + soft lifecycle (quotes+proof counts, revocation guards, invalidation contracts, typed decay); (5) scoped banks/cubes/projects with access control.
+
+No system yet ships all five with independent evals; Rune's design already covers (2)(3)(4)(5) architecturally. Deltas to import: environment-probing curation, raw-history retention for migration, revocation enforcement at retrieval.
+
@@ -0,0 +1,8 @@
+# Ecosystem survey
+
+GitHub activity (Sep 2026): Mem0 65.7k (platform-pivot), Graphiti/Zep 31k (active, security-maintained), Hindsight 24k (fastest riser, paper-backed, cloud+MIT), MemOS 11.5k (2,161 commits, plugin ecosystem).
+
+Production-grade: Mem0 Cloud, Zep, Hindsight (Postgres/Helm/webhooks/SLA), MemOS Cloud. Prototype/paper: MIRIX, EdgeMem, MemForest, most Section-D systems.
+
+Pattern: every major system ships an agent-harness plugin (OpenClaw/Hermes/DeepSeek Harness/Pi are the contested distribution surface) — memory is becoming a plug-in commodity for coding agents.
+
@@ -0,0 +1,6 @@
+# Evaluation frameworks
+
+LongMemEval remains the de facto standard (Hindsight 91.4%, MemOS 89.20, FR-Bank 75.2 — note judge-protocol dependence). New benchmarks: MemoryAgentBench (2507.05257 v4; four competencies: accurate retrieval, test-time learning, long-range understanding, selective forgetting — current systems master none of all four); OmniMemEval (MemTensor; 14 products); LifecycleBench (516 temporal-disambiguation questions); MUMBench (multi-user); BEAM; AMA-Bench; HaluMem (confabulation).
+
+Controversies crystallized: restore-counterfactual paper shows budget-accuracy numbers incomparable unless retrieval regime disclosed; Fortunate Recall separates confabulation-rate metrics from accuracy; eviction audits show irreversibility ≠ accuracy loss. Any Rune qualification claim on LongMemEval must pin judge protocol + retrieval regime.
+
@@ -0,0 +1,22 @@
+# Hindsight full system audit
+
+Paper: "Hindsight is 20/20: Building Agent Memory that Retains, Recalls, and Reflects", arXiv 2512.12818 (Latimer, Boschi, Neeser, Bartholomew, Srivastava, Wang, Ramakrishnan; 14 Dec 2025).
+
+GitHub: 24k stars, GitHub-Trending #1. Production-grade posture (PostgreSQL+pgvector, Helm, webhooks, Hindsight Cloud with 99.9% SLA, MIT license).
+
+Architecture: agent memory as "structured, first-class substrate" of four logical networks — world facts, agent experiences, synthesized entity summaries (observations), evolving beliefs/mental models; retain/recall/reflect ops; temporal + entity-aware incremental ingestion.
+
+91.4% LongMemEval claim: in the paper itself (LoCoMo 89.61% vs 75.78% for strongest prior open system). Status: partially independent — README says "independently reproduced by research collaborators at the Virginia Tech Sanghani Center and The Washington Post"; co-author Ramakrishnan is Virginia Tech. VENDOR+ACADEMIC, not disinterested third party.
+
+New shipped features since R0:
+- Observations as deduplicated evidence-backed beliefs (exact quotes + proof count, refined-not-overwritten)
+- Mental models: precomputed standing answers, DB-read at boot, background rewrite
+- Knowledge pages: wiki-folders of living docs, projectable to disk as markdown
+- Memory banks with disposition traits and strict isolation
+- Memory Defense: opt-in per-bank PII/secret scanning (45 patterns, redact-or-block)
+- Recall = 4 parallel strategies (semantic/BM25/graph+temporal) → RRF → cross-encoder rerank
+
+Rune's container ghcr.io/vectorize-io/hindsight exited 137 (OOM-kill or manual stop) 5 weeks ago; image 6.34 GB. Consistent with decommissioning.
+
+Classification: ADAPT (mental-models + evidence-provenance consolidation genuinely new-to-Atlas; LLM-driven retention contradicts hash-dedup retreat — tension to resolve).
+
Author	SHA1	Message	Date
conductor	077c588930	Merge pull request 'research: mechanical depth gate for tranches (#409 )' (#410 ) from research-depth-gate into m10-derive-refresh deploy / deploy (push) Skipped Details CI Pipeline / best-practices (push) Successful in 2s Details CI Pipeline / documentation (push) Successful in 8s Details CI Pipeline / gitea-offline (push) Successful in 14s Details CI Pipeline / lint (push) Failing after 37s Details CI Pipeline / canary (push) Failing after 39s Details CI Pipeline / type-check (push) Successful in 42s Details CI Pipeline / unit-tests (push) Failing after 6m22s Details CI Pipeline / required-checks (push) Failing after 6s Details CI Pipeline / CI Pipeline / notify-event-bridge (push) Successful in 8s Details	2026-09-20 11:36:06 +01:00
conductor	2f5045e989	research: mechanical depth gate for tranches (issue #409 ) CI Pipeline / best-practices (push) Successful in 3s Details CI Pipeline / documentation (push) Successful in 1s Details CI Pipeline / canary (push) Successful in 28s Details CI Pipeline / type-check (push) Successful in 27s Details deploy / deploy (push) Skipped Details CI Pipeline / lint (push) Successful in 34s Details CI Pipeline / gitea-offline (push) Successful in 12s Details CI Pipeline / required-checks (pull_request) Skipped Auto-label / priority-gate (pull_request) Skipped Details Auto-label / auto-label (pull_request) Successful in 1s Details CI Pipeline / unit-tests (push) Successful in 4m43s Details CI Pipeline / required-checks (push) Successful in 1s Details CI Pipeline / CI Pipeline / notify-event-bridge (push) Successful in 1s Details The overnight R1 run (epic #375) closed 16 of 31 domains with reduced-depth batched masters; the quality firewall was self-attested and bent under controller context pressure. Post-hoc review caught it; this change makes depth mechanically checkable so the failure class cannot recur silently. - scripts/research_gate_check.py: stage-file presence + substance floors + adversarial provenance + R0 reconciliation + source traceability; R1_FULL / R1_LITE grades, LITE requires registry pre-declaration (silent scope reduction = gate drift). - research_standard.md: five binding rules — mechanical gate, breadth-before-depth as the only legal trade, declared grades, adversarial independence, depth checkpointing between domains. - domain_registry.md: stale statuses corrected, P–AE marked R1_LITE retroactively, 39 grandfathered findings across 8 pre-gate domains recorded (C-cognitive-architectures passes R1_FULL clean as the calibration domain). Version intent: patch (tooling + research governance, no runtime behavior change). Fixes #409.	2026-09-20 11:31:08 +01:00
conductor	d1d821059f	atlas(r1): combined master for external sharing; reconciliation green after #407 fix CI Pipeline / lint (push) Successful in 32s Details CI Pipeline / canary (push) Successful in 24s Details CI Pipeline / best-practices (push) Successful in 2s Details CI Pipeline / documentation (push) Successful in 1s Details CI Pipeline / type-check (push) Successful in 25s Details CI Pipeline / gitea-offline (push) Successful in 13s Details deploy / deploy (push) Skipped Details CI Pipeline / unit-tests (push) Successful in 4m40s Details CI Pipeline / required-checks (push) Successful in 0s Details CI Pipeline / CI Pipeline / notify-event-bridge (push) Successful in 1s Details R1_COMBINED_MASTER/RUNE_MEMORY_RESEARCH_ATLAS_R1_COMBINED_MASTER.md: all 31 domain masters in one file (150KB) with subject/method/header for external readers; mirrored to ~/Desktop/research/R1_COMBINED_MASTER/. #324 next_action cleared (label fix verified live), #407 recorded FIXED_VERIFIED (PR #408 merged, closed-issue guard, live re-add test negative). Checker: 73 records, consistent.	2026-09-20 11:20:41 +01:00
conductor	679689cb15	merge remote m10-derive-refresh (atlas commit landed locally first) CI Pipeline / best-practices (push) Successful in 4s Details CI Pipeline / documentation (push) Successful in 1s Details CI Pipeline / canary (push) Successful in 29s Details CI Pipeline / type-check (push) Successful in 26s Details deploy / deploy (push) Skipped Details CI Pipeline / lint (push) Successful in 34s Details CI Pipeline / gitea-offline (push) Successful in 12s Details CI Pipeline / unit-tests (push) Successful in 4m43s Details CI Pipeline / required-checks (push) Successful in 0s Details CI Pipeline / CI Pipeline / notify-event-bridge (push) Successful in 1s Details # Conflicts: # docs/gitea_issue_reconciliation.json	2026-09-20 06:28:19 +01:00
conductor	5e0744a70c	atlas(r1): all 31 domains qualified; reconciliation updated for #316/#324 closures Overnight R1 supervisor complete: #376-#406 all R1_QUALIFIED under the QF1-QF22 firewall, epic #375 closed. Reconciliation artifact now matches the live tracker for #316 (CONCLUDED_EVIDENCE) and #324 (FIXED_VERIFIED); remaining #324 checker drift is owned by #407 (auto-label re-adds bug label to closed issues).	2026-09-20 06:26:54 +01:00