Fix the issues for each RQ instead of a general summary

This learning page is generated for GitHub Wiki. The public portal is managed on mind-upload.com.

• Updated: 2026-04-02 / Role: RQ-by-RQ Deep Focus

This page is a practical note that fixes the core of mind-upload's 60 RQs, which can be confirmed directly with EEG-DATA, application themes for auto-research-funds, and necessary requirements outside of EEG.

A/B/C judgment is the same as mind-upload-eeg-rq60-feasibility-and-funds. C means 'not complete with EEG alone' rather than 'not important'.

• Literature map
• Data & Bench

• Wiki: RQ60 EEG verification availability and subsidy theme - The original version of the judgment and data candidates.
• Wiki: RQ60 Application Playbook - Fixed table of applicant ID and minimum deliverables.

---

1. Select only one RQ and implement only the core to be directly verified with EEG in the table below first.
2. Copy the application theme (grant key) directly to the proposed application title.
3. Don't expand your claim unless you can satisfy the ``required outside the EEG''.
4. Numeric ID in this page is for search guidance only. Submission/reproducibility logs must use DOI + dataset name + access class in the canonical map auto-research-funds/wiki/Mind-Upload-EEG-RQ-Grant-Map.md.

• A: Main claims can be verified with EEG-DATA only.
• B: EEG main analysis is possible. Requires additional external modality/design.
• C: Mainly system/legal/theoretical integration. EEG is an audit log or supporting evidence.

• An actual-pull reread after mind-upload@1ab35b64796f3936e525befb5d9428d789a16263, EEG-DATA@be3d03d1c286dcd37fb056317a693e47c39f7f3a, and auto-research-funds@845e6849a6c36e6e83394e32cf8dc4b5fe6fdc07 still leaves the current public six-question batch, the solve-first order, and the A/B/C reading unchanged.
• The newest pulled deltas still widen support rather than the current practical batch: mind-upload now just carries the earlier 21:05 JST route-hold sync, EEG-DATA adds three more deep EEG dataset overrides while now reporting merged_unique=30918, high_confidence=28881, and access_counts_high.immediate=24528, and auto-research-funds now just carries the earlier 21:05 JST funding addendum without changing the practical shortlist.
• Recounting the ### Ux-y headings in the RQ-by-RQ dossiers still returns 60, the grant/dataset playbook still keeps 60 unique Ux-y rows, and the present practical reading still stays deepen one row at a time.
• The route-safe access read still stays narrow in one specific way: the refreshed local EEG-DATA catalog keeps widening, but the live-access read still leaves D10 as the only clear public-open primary anchor while D03 / D02 / D01 remain gated-support anchors in public wording.
• Browser-level live checks at 2026-04-03 22:04 JST still returned 200 for issue.html, mind-upload-current-public-six-rq-brief.html, mind-upload-rq-solvability-bridge.html, and u13-2-brain-minus-prior-route.html; those public surfaces reported Last-Modified: Fri, 03 Apr 2026 12:24:26 GMT, and their route-facing bodies already surfaced the earlier 21:05 JST hold note before this source sync.
• Official window spot checks still keep the same practical bridge: Kura Fund = 2026-02-02 - 2026-04-17 17:00, Nakatani Foundation graduate scholarship = 2026-04-01 - 2026-05-25 15:00, Kashinome Scholarship = 2026-01-14 - 2026-04-09, while Nakatani Foundation research grant, Secom General Research Grant, and Brain Science Foundation research grant remain outside the current-cycle lane.
• A fresh read-only Todoist api/v1/sync(items, projects) audit still showed enough route coverage (items=205, projects=2, Kura=2, Nakatani family=11, Nakatani graduate scholarship=7, Kashinome=2, Secom=3, Brain Science Foundation=8, Okawa=2, Japan Industrial Science Research Institute=1, Keio graduate scholarship=2), so no new task was added in this session.

• An actual-pull reread after mind-upload@208e59e2180c, EEG-DATA@00d85380f3fd, and auto-research-funds@f4ba35518400 still leaves the current public six-question batch, the solve-first order, and the A/B/C reading unchanged.
• The newest pulled deltas still widen support rather than the current practical batch: mind-upload refreshes the 10:07 route-hold packets, EEG-DATA deep-dives the maintained ChineseEEG-2 release plus the public auditory-eeg family while now reporting merged_unique=30887, high_confidence=28850, and access_counts_high.immediate=24875, and auto-research-funds keeps the same practical shortlist after dossier resync.
• Recounting the ### Ux-y headings in the RQ-by-RQ dossiers still returns 60, the grant/dataset playbook still keeps 60 unique Ux-y rows, and the present practical reading still stays deepen one row at a time.
• The route-safe access read still stays narrow in one specific way: the refreshed local EEG-DATA catalog keeps widening, but the live-access read still leaves D10 as the only clear public-open primary anchor while D03 / D02 / D01 remain gated-support anchors in public wording.
• Browser-level live checks at 2026-04-02 11:05 JST still returned 200 for issue.html, mind-upload-current-public-six-rq-brief.html, mind-upload-rq-solvability-bridge.html, and u8-1-closed-loop-delay-tolerance-route.html; those public surfaces reported Last-Modified: Thu, 02 Apr 2026 01:22:44 GMT.
• Official window spot checks still keep the same practical bridge: Kura Fund = 2026-02-02 - 2026-04-17 17:00, Nakatani Foundation graduate scholarship = 2026-04-01 - 2026-05-25 15:00, Kashinome Scholarship = 2026-01-14 - 2026-04-09, while Nakatani Foundation research grant, Secom General Research Grant, and Brain Science Foundation research grant remain outside the current-cycle lane.
• A fresh read-only Todoist api/v1/sync(items, projects) audit still showed enough route coverage (items=203, projects=2, Kura=2, Nakatani graduate scholarship=7, Kashinome=2, Secom=3, Brain Science Foundation=8, Okawa=2, Japan Industrial Science Research Institute=1, AI for Aging-Society Problem Solving=0), so no new task was added in this session.

• An actual-pull reread after mind-upload@da9644095ba6, EEG-DATA@1ecc5cdc0131, and auto-research-funds@b5cd37febbbb still leaves the current public six-question batch, the solve-first order, and the A/B/C reading unchanged.
• The newest pulled deltas still widen support rather than the current practical batch: mind-upload tightens human-proxy composition evidence roles outside the current six, EEG-DATA adds an Iowa Gambling EEG family deep dive while still reporting merged_unique=30873, high_confidence=28836, and access_counts_high.immediate=24486, and auto-research-funds adds the Nagahisa research grant snapshot plus a README-stat refresh without changing the practical shortlist.
• Recounting the ### Ux-y headings in the RQ-by-RQ dossiers still returns 60, the grant/dataset playbook still keeps 60 unique Ux-y rows, and the present practical reading still stays deepen one row at a time.
• The route-safe access read still stays narrow in one specific way: the refreshed local EEG-DATA catalog keeps widening, but the live-access read still leaves D10 as the only clear public-open primary anchor while D03 / D02 / D01 remain gated-support anchors in public wording.
• Browser-level live checks at 2026-04-01 20:09 JST still returned 200 for issue.html, mind-upload-current-public-six-rq-brief.html, and mind-upload-rq-solvability-bridge.html; those public surfaces reported Last-Modified: Wed, 01 Apr 2026 10:21:50 GMT.
• Official window spot checks still keep the same practical bridge: Kura Fund = 2026-02-02 - 2026-04-17 17:00, Nakatani Foundation graduate scholarship = 2026-04-01 - 2026-05-25 15:00, Kashinome Scholarship = 2026-01-14 - 2026-04-09, while Nakatani Foundation research grant, Secom General Research Grant, and Brain Science Foundation research grant remain outside the current-cycle lane.
• A fresh read-only Todoist api/v1/sync(items) audit still showed enough route coverage (items=203, Kura=2, Nakatani graduate scholarship=7, Kashinome=2, Secom=3, Brain Science Foundation=8, Okawa=2, Japan Industrial Science Research Institute=1, AI for Aging-Society Problem Solving=0), so no new task was added in this session.

• An actual-pull reread after mind-upload@c5a0550, EEG-DATA@82893f6ca6e, and auto-research-funds@ec6f7d068 still leaves the current public six-question batch, the solve-first order, and the A/B/C reading unchanged.
• The newest pulled deltas still widen the support map rather than the current practical batch: mind-upload deepens astrocyte PET route splits outside the current six, EEG-DATA now reports merged_unique=30871, high_confidence=28834, and access_counts_high.immediate=24484, and its experiment-metadata report still covers 18,312 / 28,832 rows with at least one experiment-facing field (63.51% coverage), but those gains still do not replace D10 / D03 / D02 / D01.
• Recounting the ### Ux-y headings in the RQ-by-RQ dossiers still returns 60, the grant/dataset playbook still keeps 60 unique Ux-y rows, and the present practical reading still stays deepen one row at a time.
• The route-safe access read also stays narrow in one specific way: the refreshed local EEG-DATA catalog still shows D03 / D02 as catalog-side immediate rows and D01 as a mixed immediate/check-needed pair, but the live-access read remains stricter, so D10 stays the only clear public-open primary anchor while D03 / D02 / D01 still stay gated-support anchors in public wording.
• Browser-level live checks at 2026-04-01 17:06 JST still returned 200 for issue.html, mind-upload-current-public-six-rq-brief.html, mind-upload-rq-solvability-bridge.html, and u8-1-closed-loop-delay-tolerance-route.html, and those public surfaces now share Last-Modified: Wed, 01 Apr 2026 07:16:09 GMT.
• Official window spot checks still keep the same practical bridge: Kura Fund = 2026-02-02 - 2026-04-17 17:00, Nakatani Foundation graduate scholarship = 2026-04-01 - 2026-05-25 15:00, Kashinome Scholarship = 2026-01-14 - 2026-04-09, while Nakatani Foundation research grant, Secom General Research Grant, and Brain Science Foundation research grant remain outside the current-cycle lane.
• A fresh read-only Todoist api/v1/sync(items) full-item audit still showed enough route coverage (items=203, Kura=2, Nakatani family=11, Nakatani graduate scholarship=7, Kashinome=2, Secom=3, Brain Science Foundation=8, Okawa=2, Japan Industrial Science Research Institute=1, AI for Aging-Society Problem Solving=0), so no new task was added in this session.

• An actual-pull reread after mind-upload@ccef40f, EEG-DATA@deefd5f95c6, and auto-research-funds@f5f76c485 still leaves the current public six-question batch, the solve-first order, and the A/B/C reading unchanged.
• The newest pulled deltas still widen the support map rather than the current practical batch: EEG-DATA now reports merged_unique=30869, high_confidence=28832, and access_counts_high.immediate=24482, and its new experiment-metadata report now covers 18,312 / 28,832 rows with at least one experiment-facing field (63.51% coverage), but those gains still do not replace D10 / D03 / D02 / D01.
• Recounting the ### Ux-y headings in the RQ-by-RQ dossiers still returns 60, the grant/dataset playbook still keeps 60 unique Ux-y rows, and the present practical reading still stays deepen one row at a time.
• The route-safe access read also stays narrow in one specific way: the refreshed local EEG-DATA catalog still shows D03 / D02 as catalog-side immediate rows and D01 as a mixed immediate/check-needed pair, but the live-access read remains stricter, so D10 stays the only clear public-open primary anchor while D03 / D02 / D01 still stay gated-support anchors in public wording.
• Browser-level live checks at 2026-04-01 14:08 JST still returned 200 for issue.html, mind-upload-current-public-six-rq-brief.html, mind-upload-rq-solvability-bridge.html, and u8-1-closed-loop-delay-tolerance-route.html, and those public surfaces now share Last-Modified: Wed, 01 Apr 2026 04:20:02 GMT.
• Official window spot checks still keep the same practical bridge: Kura Fund = 2026-02-02 - 2026-04-17 17:00, Nakatani Foundation graduate scholarship = 2026-04-01 - 2026-05-25 15:00, Kashinome Scholarship = 2026-01-14 - 2026-04-09, while Nakatani Foundation research grant, Secom General Research Grant, and Brain Science Foundation research grant remain outside the current-cycle lane.
• A fresh read-only Todoist api/v1/sync(items) full-item audit still showed enough route coverage (items=203, Kura=2, Nakatani family=11, Kashinome=2, Secom=3, Brain Science Foundation=8, Okawa=2, Japan Industrial Science Research Institute=1, AI for Aging-Society Problem Solving=0), so no new task was added in this session.

• A latest pull reread after mind-upload@31835fcd8daa, EEG-DATA@12eef8ca4ef0, and auto-research-funds@040e5b8798fd still leaves the current public six-question batch, the solve-first order, and the A/B/C reading unchanged.
• The newest pulled deltas still widen the support map rather than the current practical batch: mind-upload split temporal routes for invasive language BCIs outside the current six, EEG-DATA added a semantic-text-relevance deep dive that widens language-related backup coverage without displacing D10 / D03 / D02 / D01, and auto-research-funds refreshed Obayashi / Yokogaku cards without changing the current shortlist.
• Recounting the ### Ux-y headings in the RQ-by-RQ dossiers still returns 60, the grant/dataset playbook still keeps 60 unique Ux-y rows, and the present practical reading still stays deepen one row at a time.
• Browser-level live checks at 2026-04-01 03:09 JST still returned 200 for issue.html, mind-upload-current-public-six-rq-brief.html, mind-upload-rq-solvability-bridge.html, and u8-1-closed-loop-delay-tolerance-route.html.
• Official window spot checks still keep the same practical bridge: Kura Fund = 2026-02-02 - 2026-04-17 17:00, Nakatani Foundation graduate scholarship = 2026-04-01 - 2026-05-25 15:00, Kashinome Scholarship = 2026-01-14 - 2026-04-09, while Nakatani Foundation research grant, Secom General Research Grant, and Brain Science Foundation research grant remain outside the current-cycle lane.
• A fresh read-only Todoist api/v1/sync(items) full-item audit still showed enough route coverage (items=203, Kura=2, Nakatani family=11, Kashinome=2, Secom=3, Brain Science Foundation=8, Okawa=2, Japan Industrial Science Research Institute=1, AI for Aging-Society Problem Solving=0), so no new task was added in this session.

• A latest pull reread after mind-upload@c912fc5ebd4a, EEG-DATA@9bdf3b16bfb4, and auto-research-funds@4f071fcd98fd still leaves the current public six-question batch, the solve-first order, and the A/B/C breakdown unchanged.
• The newest EEG-DATA additions still widen the support map rather than the current practical batch: the current README summary now reports merged_unique=30,861, high_confidence=28,824, and access_counts_high.immediate=24,843, but those gains still do not replace the current D10 / D03 / D02 / D01 family.
• Recounting the ### Ux-y headings in the RQ-by-RQ dossiers still returns 60, the grant/dataset playbook still keeps 60 unique Ux-y rows, and the A/B/C split still stays 17 / 25 / 18.
• Browser-level live checks with a browser user agent still returned 200 for issue.html, mind-upload-current-public-six-rq-brief.html, and u8-1-closed-loop-delay-tolerance-route.html.
• The practical funding reading still stays narrow: Kura Fund, Nakatani Foundation graduate scholarship, and Kashinome Scholarship remain the active bridge for the current six; Secom General Research Grant is already closed for the current cycle (March 31, 2026 at 12:00 JST); Nakatani Foundation research grant stays next-cycle watch; and no new Todoist item was created in this session.

• A latest pull reread after mind-upload@a11240d38415, EEG-DATA@f318ac8ce42d, and auto-research-funds@c95cc61fdfc9 still leaves the current public six-question batch and the solve-first order unchanged.
• The newest deltas widen the support map, not the current practical batch: mind-upload tightened the connectome / timing-state route cards, EEG-DATA added an OSF eucqf deep-dive refresh, and auto-research-funds added new Takeda / Nohken funding-card material, but none of those changes replace the current D10 / D03 / D02 / D01 family or the outside-current-six order U7-1 -> U1-2 -> U4-2 -> U8-1.
• Browser-level live checks still returned 200 for issue.html, mind-upload-current-public-six-rq-brief.html, and u8-1-closed-loop-delay-tolerance-route.html, so this run does not add new public prose.
• The Todoist full-item audit still showed enough family coverage (items=205, Kura=2, Nakatani family=11, Kashinome=2, Secom family=3, AI for Aging-Society Problem Solving=0), so no new task was added.

• This run rechecked the live public route, the current official funding windows, and a Todoist full-item audit after the latest EEG-DATA upstream additions.
• The current public six-question batch still stays U13-2/U13-5/U0-2/U0-3/U14-4/U14-2, and the solve-first order still stays U13-2 -> U13-5 -> U0-2 -> U0-3 -> U14-4 -> U14-2.
• The newest upstream EEG-DATA additions remain outside the current speech / identity / reproducibility anchor family, so they do not replace D10 as the only clearly public-open primary anchor and do not turn D03/D02/D01 into immediate public routes.
• Funding-window recheck in this run still supports the same narrow route: Kura Fund remains open through 2026-04-17 17:00, the Nakatani Foundation graduate scholarship still runs 2026-04-01 - 2026-05-25 15:00, the Kashinome Scholarship remains open through 2026-04-09, and Secom General Research Grant remains a watch route through 2026-03-31 12:00.
• A March 31 access audit at 2026-03-31 00:20 JST also means earlier March 30 notes that called D03/D02/D01 immediate public anchors should now be read as superseded. Those three stay only as subscription-gated support anchors on the public route.
• Todoist still did not need a new item in this run. The full-item audit continues to show the relevant family coverage already present (Kura=2, Nakatani family=11, Kashinome=2, Secom family=3, AI for Aging-Society Problem Solving=0).

• A fresh pull --ff-only recheck still leaves mind-upload, auto-research-funds, and the active EEG-DATA workspace aligned with origin/main, so this run adds no hidden upstream batch switch.
• A live reread of https://mind-upload.com/issue.html still shows the same public six, the same execution order, and the same route-hold framing, so the public source page itself did not need new prose in this run.
• Direct access rechecks now reconfirm the anchor split more concretely: D10 3M-CPSEED remains public-open on OpenNeuro, while the current IEEE DataPort pages for D03, D02, and D01 still show LOGIN TO ACCESS DATASET FILES plus Subscription Required, so those three remain subscription-gated support anchors rather than immediate public anchors.
• A same-run official-window reread still supports the same narrow funding bridge: Kura Fund = 2026-04-17 17:00, Nakatani Foundation graduate scholarship = 2026-04-01 - 2026-05-25 15:00, Kashinome = 2026-04-09, and Secom General Research Grant = 2026-03-31 12:00.
• A same-run Todoist sync audit still finds the practical route already covered (Kura=2, Nakatani graduate=5, Nakatani research=3, Kashinome=2, Secom=3), so no new task was added.

• This run pulled mind-upload@836d4db, EEG-DATA@25f065735da, and auto-research-funds@7619c14a2, then rechecked the live public page plus the current official funding windows.
• The current public six-question batch still stays U13-2/U13-5/U0-2/U0-3/U14-4/U14-2, and the solve-first order still stays U13-2 -> U13-5 -> U0-2 -> U0-3 -> U14-4 -> U14-2.
• The practical reason for holding the batch is unchanged: the narrowest current bridge from mind-upload.com RQs to bounded EEG-ready and fundable work still runs through inner-speech decoding + identity drift + reproducibility auditing, not through a broader thematic spread.
• Anchor strength also stays conservative after the pull. D10 3M-CPSEED remains the only clear public-open primary anchor, D02 CSTE remains the narrowest gated planning anchor for the identity rows, and D03 Multimodal Fusion EEG + Eye-Tracking plus D01 Acquisition delay of wireless EEG remain bounded-support anchors with lower claim ceilings.
• Funding-window recheck in this run: Kura Fund remains open through 2026-04-17 17:00, the Nakatani Foundation graduate scholarship opens 2026-04-01 and closes 2026-05-25 15:00, the Kashinome Scholarship remains open through 2026-04-09, the Secom General Research Grant remains an urgency watch route through 2026-03-31 12:00, and the Nakatani Foundation research grant remains next-cycle watch.
• Large social-deployment programs such as AI for Aging-Society Problem Solving remain outside this current EEG-first route, because their required output is a scaled public deployment program rather than a one-RQ / one-anchor / one-first-artifact EEG result.

• This run rechecked the live public page, the current official funding windows, and Todoist in read-only mode. The batch itself still does not change.
• What does change here is the wording of what counts as real progress right now: not a stronger claim, but a narrower first artifact for each of the six current public RQs.

RQ	What can honestly move forward now	Why the current anchor is enough for that next step	First artifact to close next	Keep outside the public claim
U13-2	Fix the brain-minus-prior control family before any stronger decoder wording.	D10 3M-CPSEED already gives overt / silent-intended / imagined speech, repeated sessions, and transcript-aligned phases, so it is sufficient for a bounded hallucination / inconsistency audit.	Brain-minus-prior control table and mimic-separation note	Do not treat decoder success as preserved internal state or mind preservation.
U13-5	Freeze only the perception-to-recall branch point, not a general mechanism claim.	D03 Multimodal Fusion EEG + Eye-Tracking is enough to keep the transfer-gap interpretation bounded because the shared EEG + gaze streams let the branch stay tied to a concrete perception/recall shift.	Perception-to-recall degradation note and transfer-gap table	Do not present a shared decoder as evidence of preserved internal mechanism.
U0-2	Close the timing budget that prevents identity-like wording from absorbing measurement error.	D02 CSTE already has repeated sessions, cross-task labels, and timestamp/QC structure, so it is enough to lock clock offset + jitter + state-feature collapse into one narrow audit.	Offset/jitter distribution and state-feature collapse audit	Do not elevate synchronization stability into legal or philosophical identity continuity.
U0-3	Fix the threshold-selection rule only after the timing route is bounded.	The same D02 CSTE anchor lets threshold stability and overfit exclusion stay on the same cross-session / cross-task identity-drift floor as U0-2.	Threshold-stability note and overfit-exclusion rule	Do not treat a stable biometric threshold as a sufficient condition for personhood continuity.
U14-4	Freeze the disclosure fields that the current batch actually needs for reruns.	D01 Acquisition delay of wireless EEG is narrow enough to turn Model Card / Dataset Card items into a rerun checklist without overclaiming field-wide standardization.	Filled Model Card / Dataset Card examples for the current public batch	Do not claim organizational adoption or community-wide standardization yet.
U14-2	Make the exploration/confirmation split operational rather than aspirational.	The same D01 latency-first route is enough to define a fixed split plus rerun boundary, even though it is still too narrow for broader reproducibility claims.	Fixed-split rerun contract and negative-case trail	Do not present exploratory outputs as confirmatory evidence after the split boundary is crossed.

• Official-window recheck in this run stayed conservative: Kura Fund remains open through 2026-04-17 17:00, the Nakatani Foundation graduate scholarship detail page still shows 2026-04-01 - 2026-05-25 15:00, the Kashinome Scholarship remains open through 2026-04-09, and Secom General Research Grant remains a watch route through 2026-03-31 12:00.
• Todoist read-only recheck found the current route already covered by existing tasks, including Kura Fund, Nakatani Foundation graduate scholarship, Kashinome Scholarship, Nakatani Foundation research grant, and Secom. No new task was added in this run.

This run changes the active public batch to U13-2, U13-5, U0-2, U0-3, U14-4, and U14-2.

This is a route-setting update, not a solved-claim update. The public batch is now intentionally concentrated on inner-speech decoding + identity drift + reproducibility auditing, because these six RQs are the cleanest current bridge from mind-upload.com research questions to bounded EEG-ready and fundable work packages.

RQ	What is fixed first	Main Dxx + DOI anchor	Practical first / backup route	First-pass KPI	Keep outside the EEG claim
U13-2	A brain-minus-prior hallucination audit for overt/silent/imagined speech.	D10 / 10.18112/openneuro.ds006465.v2.0.0	Kura Fund / Nakatani Foundation graduate scholarship	semantic alignment brain-minus-prior gap misalignment rate	Do not treat language-decoder success as a mind-preservation claim.
U13-5	A perception-to-recall branch analysis for mimicry/mechanism separation.	D03 / 10.21227/6106-6120	Kura Fund / Nakatani Foundation graduate scholarship	perception-to-recall degradation point transfer gap control-condition stability	Do not treat a shared decoder as evidence of preserved internal mechanism.
U14-4	Model Card / Dataset Card fields are wired into the reproducibility contract.	D01 / 10.21227/dv1p-vq18	Kura Fund / Nakatani Foundation research grant (watch)	required-card fill rate rerun success rate leak-audit coverage	Do not claim organizational adoption or field-wide standardization yet.
U0-2	The identity route is limited to clock offset + jitter + state-feature collapse auditing.	D02 / 10.21227/j162-nh61	Nakatani Foundation graduate scholarship / Kashinome Scholarship (student-runway only)	clock-offset p95 state-feature collapse rate resync recovery rate	Do not elevate technical synchronization stability to legal or philosophical identity.
U0-3	Task-specific threshold stability and overfit exclusion are fixed before any stronger identity reading.	D02 / 10.21227/j162-nh61	Nakatani Foundation graduate scholarship / Kashinome Scholarship (student-runway only)	session-out AUC threshold stability overfit exclusion rate	Do not treat a stable biometric threshold as a sufficient condition for personhood continuity.
U14-2	Exploratory and confirmatory tracks are separated inside the rerun contract.	D01 / 10.21227/dv1p-vq18	Kura Fund / Nakatani Foundation research grant (watch)	fixed-split rate rerun success rate explore-to-confirm carryover gap	Do not present exploration outputs as confirmatory evidence after the split boundary is crossed.

• Practical funding alignment in this run is intentionally narrow: Nakatani Foundation graduate scholarship, Kura Fund, Kashinome Scholarship (student-runway only), and Nakatani Foundation research grant already provide active routes or watch routes for this cluster.
• Relevant Todoist tasks already existed before this run, so no new task was needed for the current practical shortlist.
• Funding-window spot check (2026-03-29): Kura Fund remains open in the current cycle, the Nakatani Foundation graduate scholarship remains the main student-facing bridge for this batch, the Kashinome Scholarship stays only as a student-runway side route rather than a core EEG research-funding anchor, and the Nakatani Foundation research grant plus broader neuroscience routes stay in watch mode.

• The current six-question batch does not change here. What changes is the explicit reading order for the anchors: D10 and D02 are the strongest current primary anchors, while D03 and D01 stay in the batch as bounded route-setting anchors with narrower claim ceilings.
• Practical consequence: U13-2/U0-2/U0-3 can carry stronger first-pass artifacts inside EEG-DATA alone, while U13-5/U14-4/U14-2 should remain attached to narrower artifacts until either the DOI-based source package is exercised more deeply or an adjunct benchmark is added.
• Submission and reproducibility references remain fixed to Dxx + DOI + dataset name + access class. Numeric IDs stay search-only because the local mirror keeps re-ranking and remapping those IDs.
• There is also a structural backlog outside this six-question batch: U2/U5/U6/U9 exist in tech_roadmap.md, but they are not yet translated into the current RQ60 practical pages. They should be treated as the next internal translation queue rather than as an already-covered part of the public six.

This run does not change the public six-question batch. It only fixes the execution order more explicitly, so the work can proceed as one-RQ / one-anchor / one-first-artifact instead of drifting back into a broad summary.

The order below is intentionally sequential. Each row closes a blocker for the next row rather than treating the six questions as one blended benchmark.

Order	RQ	Why this comes before the next one	Main anchor / access	First artifact to finish	Practical route now
1	U13-2	It is the shortest route to test whether the neural signal adds anything beyond the language prior. If this fails, stronger speech-decoder wording stays blocked.	D10 / 10.18112/openneuro.ds006465.v2.0.0 / immediate	Mimic separation evaluation script and brain-minus-prior control table	Kura Fund / Nakatani Foundation graduate scholarship
2	U13-5	It becomes meaningful only after U13-2 fixes the control family. Then the perception-to-recall branch point can be read as a bounded mechanism split rather than a generic decoder score.	D03 / 10.21227/6106-6120 / subscription-gated support	Perception-to-recall degradation note and transfer-gap table	Kura Fund / Nakatani Foundation graduate scholarship
3	U0-2	Identity-like wording remains unsafe until synchronization drift is audited at the millisecond level. This is the first hard ceiling for the continuity route.	D02 / 10.21227/j162-nh61 / subscription-gated support	Offset/jitter distribution and state-feature collapse audit	Nakatani Foundation graduate scholarship / Kashinome Scholarship (student-runway only)
4	U0-3	Threshold stability only becomes interpretable after U0-2 fixes the timing budget. Otherwise the threshold can absorb synchronization error.	D02 / 10.21227/j162-nh61 / subscription-gated support	Threshold search notebook and overfit-exclusion rule	Nakatani Foundation graduate scholarship / Kashinome Scholarship (student-runway only)
5	U14-4	Card requirements should be frozen only after the speech and identity rows stabilize what must actually be disclosed.	D01 + D13 / 10.21227/dv1p-vq18 + 10.17632/vyczny2r4w / gated floor + immediate support	Filled Model Card / Dataset Card examples for the current public batch	Kura Fund / Nakatani Foundation research grant (watch)
6	U14-2	The exploration/confirmation split is the last step because it depends on the artifact family above being stable enough to rerun without ad hoc reinterpretation.	D01 + D13 / 10.21227/dv1p-vq18 + 10.17632/vyczny2r4w / gated floor + immediate support	Fixed-split rerun contract and negative-case trail	Kura Fund / Nakatani Foundation research grant (watch)

• Live public check in this run: mind-upload.com/issue.html still presents the same six RQs and still labels them as a route-setting update, so the public source page itself does not need a new batch switch.
• Anchor reality recheck in this run: D10 = 3M-CPSEED (10.18112/openneuro.ds006465.v2.0.0, immediate), while D03 = Data of Multimodal Fusion System for Cognitive Load Assessment in Psychological Experiments: An EEG and Eye-Tracking Study (10.21227/6106-6120, subscription-gated support), D01 = Acquisition delay of wireless EEG (10.21227/dv1p-vq18, subscription-gated support), and D02 = CSTE (10.21227/j162-nh61, subscription-gated support) remain in the current route only as gated support anchors. These four anchors still define the current execution floor for the batch, but only D10 currently counts as a clearly public-open primary anchor.
• Funding-window recheck in this run: Kura Fund stayed open, the Nakatani Foundation graduate scholarship stayed the main student-facing bridge, the Kashinome Scholarship stayed a runway-only side route, and the Nakatani Foundation research grant stayed in watch mode.
• Todoist recheck in this run: the start/deadline tasks for these routes already existed, so no duplicate task was added.

• This run pulled the latest upstream route-taxonomy update and rechecked whether it changes the current public six-question batch. It does not. The public route remains U13-2/U13-5/U0-2/U0-3/U14-4/U14-2, and the solve-first order remains U13-2 -> U13-5 -> U0-2 -> U0-3 -> U14-4 -> U14-2.
• The practical reason for holding the batch is unchanged: the cleanest current bridge from mind-upload.com questions to bounded EEG-ready work still runs through inner-speech decoding + identity drift + reproducibility auditing, not through a broader thematic spread.
• Anchor strength also remains unchanged after the pull. D10 stays the only clear public-open primary anchor for first-pass artifacts, D02 stays the narrowest gated planning anchor for the identity rows, and D03 plus D01 remain bounded-support anchors whose outputs should keep a narrower claim ceiling.
• The upstream route-taxonomy tightening is still useful, but its immediate effect is internal rather than public-facing. It sharpens backlog translation and claim ceilings around uncovered roadmap families; it does not justify a new public batch switch or stronger solved-claim wording.
• Funding and task routing also stay conservative in this run. Kura Fund and the Nakatani Foundation graduate scholarship remain the practical first routes, Kashinome Scholarship stays runway-only support, Nakatani Foundation research grant stays next-cycle watch, and no new Todoist task is needed because the current practical routes are already covered by existing tasks.

The public batch above does not change in this run. This section only records the next bounded queue once the current public six have stable audit artifacts.

The rule stays the same: no broad thematic expansion, and no solved-claim wording. Each row below is a one-RQ / one-anchor / one-minimum-deliverable continuation that stays close to the current inner-speech decoding + identity drift + reproducibility auditing bundle.

RQ	Why it is next	Main Dxx + DOI anchor	Minimum deliverable	Keep outside the EEG claim
U13-1	It is the nearest unresolved split after U13-2/U13-5: semantic restoration must be separated from causal restoration.	D10 / 10.18112/openneuro.ds006465.v2.0.0	Intervention protocol and rebuttal condition definition	Do not read semantic-match success as preserved internal mechanism.
U13-3	It extends the same inner-speech cluster to the same output / different mechanism failure mode.	D10 / 10.18112/openneuro.ds006465.v2.0.0	Imitation separation evaluation report (meaning/mechanism)	Do not treat output equivalence as causal equivalence.
U14-3	It is the cleanest next reproducibility step after U14-2/U14-4: leaderboard leak and report-bias auditing.	D719 / leaderboard leak benchmark anchor used in the RQ dossier	Leak audit results and repartition trail	Do not present internal leaderboard robustness as field-wide validation.
U7-3	It fixes when preprocessing drift becomes release-block relevant instead of a mere implementation detail.	D11 / 10.18112/openneuro.ds007216.v1.0.0	Reproducibility audit report (synchronization/QC/preprocessing difference)	Do not generalize one preprocessing audit rule beyond its tested protocol family.
U1-2	It is the next strongest A candidate near the same verification stack: uncertainty calibration for inverse estimates.	D08 / 10.21227/7may-9q56	Inverse problem reproduction report (error/uncertainty)	Do not promote interval coverage stability into a blanket guarantee across all head models.
U14-1	It turns rerun requirements into an executable contract before broader standardization claims.	D01 / 10.21227/dv1p-vq18	Supplementary test operation report (including negative cases)	Do not claim community-wide agreement on the required rerun pack.

• Funding-wise, the next internal queue splits into two states on purpose: U7/U14 already have concrete reproducibility-facing watch routes, while U13/U1 still lean on G4-style neuroscience watch routes where the near-term public call signal is weaker.
• Todoist-wise, no new task is required just because a row appears here. A task is only added when an actual watch / start-date / deadline route is concrete enough to avoid duplication.

RQ	Judgment	Core to be directly verified with EEG	Application theme (grant key)	Main EEG-DATA	Required outside of EEG
U0-1	A	Two-axis ROC that separates observation matching and intervention response matching	Necessary/sufficient condition design for identity KPI (G2/G3)	11, 29, 696	Normative definition of personality identity
U0-2	A	Correspondence coefficient between ms synchronization deviation and state representation collapse	Standardization of time synchronization audit criteria (G2/G3)	D02 CSTE (10.21227/j162-nh61, subscription-gated support); adjunct: ds004148, ds007216	Inter-sensor clock calibration regular agreement
U0-3	A	External verification difference between task-specific thresholds and overfitting exclusion	Individual identification method with generalized threshold (G2/G3)	D02 CSTE (10.21227/j162-nh61, subscription-gated support); adjunct: ds004148, ds007216	Screening rules for threshold selection
U0-4	B	Lower limit evaluation of technical ID stability after branching	Procedure for determining identity of branching entity (G2/G3)	11, 29, 696	Legal entity definition/responsibility allocation

RQ	Judgment	Core to be directly verified with EEG	Application theme (grant key)	Main EEG-DATA	Required outside of EEG
U1-1	A	Position error and recall for each prior distribution	Inverse problem prior distribution bench (G1/G4)	56, 676, 1839	Model selection rules
U1-2	A	Uncertainty propagation of conductivity/electrode defects/noise injection	Inverse estimation with uncertainty calibration (G1/G4)	56, 676, 1839	Head model identification conditions
U1-3	B	Deviation rate distribution of MNE/beamformer/Champagne	Decision rule that allows method deviation (G1/G4)	56, 676, 1839	Inter-method agreement protocol
U1-4	B	Estimated values and reanalysis concordance rate when releasing backward distribution	Establishment of standards for publishing backward distribution (G1/G4)	56, 676, 1839	Scope of disclosure and anonymization decision

RQ	Judgment	Core to be directly verified with EEG	Application theme (grant key)	Main EEG-DATA	Required outside of EEG
U3-1	B	Predictive improvement of EEG alone vs. integrated model	Estimated effect size of boundary expansion (G2/G6)	13, 16, 39	Additional modality collection
U3-2	B	Lower bound of correlation between connectome proxy and functional prediction	Cross-species comparison design study (G2/G6)	13, 16, 39	Cross-species connectome data
U3-3	B	Performance deterioration when removing the body/environment loop	Verification of necessity of body connection (G2/G6)	13, 16, 39	Real environment intervention log
U3-4	C	Fixed only the proxy KPI for the required minimum configuration	System design for determining the minimum configuration (G2/G6)	13, 16, 39	Agreement on theory adoption criteria
U3-5	C	Search for alternative indicators under unobserved glymphatic conditions	Preparatory research on body fluid system integration (G2/G6)	13, 16, 39	Metabolic/body fluid measurement data
U3-6	C	Extraction of short-term failure signs of immune surveillance exclusion model	Design of long-term failure monitoring (G2/G6)	13, 16, 39	Longitudinal data on inflammation and immunity

RQ	Judgment	Core to be directly verified with EEG	Application theme (grant key)	Main EEG-DATA	Required outside of EEG
U4-1	B	Sign discrepancy rate between correlation estimation and intervention effect	Formulation of correlation → causal transfer conditions (G1/G4)	19, 56, 1839	Ethical review of intervention design
U4-2	A	Reproducibility of minimal intervention claim	Ground-truth causal verification (G1/G4)	19, 56, 1839	Stopping rule when intervention fails
U4-3	B	Counterfactual error comparison of DCM/active inference	Single design comparison of theoretical predictions (G1/G4)	19, 56, 676	Common I/O specifications between theories
U4-4	B	Falsification threshold (ΔAUC/sign reversal rate)	Evaluation contract with falsification precedent (G1/G4)	19, 56, 1839	Operating rules for failure declaration

RQ	Judgment	Core to be directly verified with EEG	Application theme (grant key)	Main EEG-DATA	Required outside of EEG
U7-1	A	Missing rate of synchronization/QC/stimulation log required items	BIDS extended audit infrastructure (G1/G3)	676, 783, 4878	Organizational adoption of standard specifications
U7-2	A	LSL synchronization error jitter distribution and threshold	Synchronization error automatic warning operation (G1/G3)	676, 783, 4878	Inter-device calibration SOP
U7-3	A	Results drift due to preprocessing setting differences	Establishment of preprocessing audit CI (G1/G3)	676, 783, 4878	CI operation governance
U7-4	A	Re-measurement judgment tree in case of alignment failure	Operational criteria for re-measurement/exclusion (G1/G3)	676, 783, 4878	Measurement re-execution cost plan
U7-5	A	False block rate against release block threshold	Stopping rule for decreasing recall rate (G1/G3)	676, 783, 4878	Release responsibility system
U7-6	A	Conclusion preservation rate under missing modality	Minimum observation set design (G1/G3)	676, 783, 4878	Clinical operational standards when missing

RQ	Judgment	Core to be directly verified with EEG	Application theme (grant key)	Main EEG-DATA	Required outside of EEG
U8-1	B	Stability rate/recovery time against delay changes	Identification of closed-loop delay tolerance range (G2/G5)	6, 1972, 2412, 696	Design of mounted system controller
U8-2	B	Online calibration frequency and performance maintenance rate	Optimization of concept drift countermeasures (G2/G5)	6, 1972, 2412, 696	Continuous operation infrastructure
U8-3	B	Variance decomposition of individual differences and diurnal variation	Long-term stability KPI design (G2/G5)	6, 1972, 2412, 696	Long-term tracking protocol
U8-4	B	Abnormality detection delay and failsafe activation rate	Safe shutdown evaluation contract (G2/G5)	6, 1972, 2412, 696	Safety procedures at operational sites
U8-5	B	Malfunction rate and recovery time during override	Human-in-the-loop evaluation (G2/G5)	6, 1972, 2412, 696	Operator training and demarcation of responsibility
U8-6	B	Trade-off curve between relearning interval and safety margin	Relearning scheduling (G2/G5)	6, 1972, 2412, 696	MLOps update control

RQ	Judgment	Core to be directly verified with EEG	Application theme (grant key)	Main EEG-DATA	Required outside of EEG
U10-1	C	Audit consistency with theoretical formula using proxy	Neural calculation and Landauer consistency (G2/G6)	735, 783, 1011	Direct measurement of heat dissipation
U10-2	C	Weak correlation between nonequilibrium index and information efficiency proxy	Verification design of nonequilibrium index definition (G2/G6)	735, 783, 1011	Simultaneous measurement with metabolic measurement
U10-3	C	Observation design validity of neural activity + metabolic proxy	Observation design template construction (G2/G6)	735, 783, 1011	Metabolic/temperature sensor integration
U10-4	C	Theoretical consistency check for calculation cost evaluation	WBE cost evaluation with thermal constraints (G2/G6)	735, 783, 1011	Actual measurement data on WBE implementation side

RQ	Judgment	Core to be directly verified with EEG	Application theme (grant key)	Main EEG-DATA	Required outside of EEG
U11-1	B	PCI approximation/LZ/wSMI rank preservation rate	Unification of theoretical cross-sectional I/O specifications (G2/G4)	735, 842, 859	Clinical operation acceptance criteria
U11-2	C	Comparison of approximation indicators within calculation budget	Evaluation of calculation amount constraints for IIT approximation (G2/G4)	735, 842, 859	Exact IIT calculation platform
U11-3	B	Single-condition comparison of theoretical conflicting points	Single experimentalization of theoretical conflicting points (G2/G4)	735, 842, 859	Joint experimental design agreement
U11-4	B	Advance declaration of failure conditions including false positives/false negatives	Definition of failure conditions for awareness indicators (G2/G4)	735, 842, 859	Clinical responsibility division and accountability

RQ	Judgment	Core to be directly verified with EEG	Application theme (grant key)	Main EEG-DATA	Required outside of EEG
U12-1	C	Technical stability log of post-branch ID	Audit design of branch ID (G2/G6)	11, 509, 696	Legal ID granting criteria
U12-2	C	Technical Event Log for Succession Points	Responsibility/Rights/Consent Succession Rules (G2/G6)	11, 509, 696	Institutional Branching Rules
U12-3	C	Change audit of psychological continuity proxy	Connection of psychological standards and legal standards (G2/G6)	11, 509, 696	Normative judgment frame
U12-4	C	Correspondence log between performance KPI and attribution KPI	Connection between performance evaluation and personality evaluation (G2/G6)	11, 509, 696	Judicial/Ethical Review
U12-5	C	Extracting ID reorganization candidates during resynchronization events	ID reorganization criteria during memory editing (G2/G6)	11, 509, 696	Personal consent and legal validity
U12-6	C	Delayed audit of authority reflection after consent withdrawal	Consent withdrawal and authority revocation audit (G2/G6)	11, 509, 696	Access control system design

RQ	Judgment	Core to be directly verified with EEG	Application theme (grant key)	Main EEG-DATA	Required outside of EEG
U13-1	B	Discrepancy between semantic matching rate and causal matching rate	Separate evaluation of meaning restoration/causal reproduction (G1/G4)	509, 13, 65	External behavior/language evaluation system
U13-2	B	Correlation between hallucination test results and neural state differences	Neural decoding connections for hallucination tests (G1/G4)	D10 3M-CPSEED (10.18112/openneuro.ds006465.v2.0.0, immediate); adjunct: spoken/imaged robot-speech EEG, D03	LLM evaluation platform collaboration
U13-3	B	Identification rate of the same output and different internal mechanisms	Shortcut detection bench (G1/G4)	509, 13, 65	Model internal visualization platform
U13-4	B	Effect of prohibiting passing based on imitation score alone	Imitation evaluation with causal consistency (G1/G4)	509, 13, 65	Causal intervention design
U13-5	B	Degradation branch point estimation from perception to recall	Perception/recall decoder branch analysis (G1/G4)	D03 Multimodal Fusion EEG + Eye-Tracking (10.21227/6106-6120, subscription-gated support); adjunct: D10, ZuCo 2.0, spoken/imaged robot-speech EEG	Unification of task design
U13-6	B	Prompt induction/leak/shortcut separation rate	Leak audit with controlled experiment (G1/G4)	509, 13, 65	Data management governance

RQ	Judgment	Core to be directly verified with EEG	Application theme (grant key)	Main EEG-DATA	Required outside of EEG
U14-1	A	Follow-up success rate for each fixed granularity	Minimum granularity definition for reproduction contract (G1/G3)	6, 56, 719	Contract document template
U14-2	A	Recall rate difference when separating exploration/verification	Operational separation of exploration and verification (G1/G3)	D01 Acquisition delay of wireless EEG (10.21227/dv1p-vq18, subscription-gated support); required support: D13; adjunct: D08	Research management process
U14-3	A	Subject/time series leak detection rate	leaderboard audit operation (G1/G3)	6, 56, 719	Enforcement of publication rules
U14-4	A	Differences in ease of auditing with and without Card description	Model/Dataset Card contracting (G1/G3)	D01 Acquisition delay of wireless EEG (10.21227/dv1p-vq18, subscription-gated support); required support: D13; adjunct: D08	Legal preparation for evaluation contracts
U14-5	B	Retry improvement rate with negative example registry	Negative example disclosure and retry cycle (G1/G3)	6, 56, 719	Disclosure consent/anonymization operation
U14-6	A	Cost increase due to container fixation	Setting the maximum allowable reproduction cost (G1/G3)	6, 56, 719	Computing resource procurement plan

RQ	Judgment	Core to be directly verified with EEG	Application theme (grant key)	Main EEG-DATA	Required outside of EEG
U15-1	C	Technical logging of neural data sensitivity	Neural data law concept mapping (G2/G6)	49, 2412, 3419	Official definition of legal concepts
U15-2	C	Correspondence between neurorights items and technical audit logs	Neurorights audit itemization (G2/G6)	49, 2412, 3419	Consistency with regulatory agency requirements
U15-3	C	Differences in technical items for requirements by jurisdiction	EU/US/JP minimum common operational definition (G2/G6)	49, 2412, 3419	Legal review of each jurisdiction
U15-4	C	Risk indicator candidates linked to suspension/renewal conditions	Governance of suspension conditions/renewal conditions (G2/G6)	49, 2412, 3419	Design of operation supervision system

• For all 60RQ, core to be verified directly with EEG and required outside of EEG are listed separately.
• Application theme is unified to the existing grant key of G1-G6.
• The operating rule is 1RQ = 1 verification proposition = 1 application theme = 1 main data.

• Re-verify mind-upload/research_harvest_50.md as the original and check the number of lines in Ux-RQy is 60, duplicate 0, and missing 0.
• Recalculated the judgment breakdown A/B/C=17/25/18 and confirmed that it is consistent with each line on this page.
• The referenced EEG data ID (6, 11, 13, 16, 19, 49, 56, 65, 509, 676, 696, 719, 735, 783, 842, 859, 2412) is EEG-DATA/eeg_dataset_summary_ja.csv All cases above exist (missing 0).
• The real ID of subsidy key G1-G6 (GR-2026-013, GR-2026-014, 9Lx4dPK6a4k2gOb7, Drbm6vBRDJkn0NGJ, 871pw3rLjNPKgqA0, 46z9VPE4wnkrvEJR) is All items can be viewed on the auto-research-funds side.
• The policy continues: Instead of a general overview, we maintain a deep-dive operation of 1 RQ = 1 verification proposition = 1 application theme = 1 main data.

• main Recounted after updating and reconfirmed RQ_TOTAL=60, A/B/C=17/25/18 (matched both on this page and mind-upload-eeg-rq60-feasibility-and-funds.md).
• Re-extracted the research question decomposition section of research_harvest_50.md as the original, RQ_COUNT=60, missing 0.
• The EEG data ID (6, 11, 13, 16, 19, 49, 56, 65, 509, 676, 696, 719, 735, 783, 842, 859, 2412) referenced on this page is EEG-DATA/eeg_dataset_summary_ja.csv All cases exist (missing 0).
• Grant ID (GR-2026-013, GR-2026-014, 9Lx4dPK6a4k2gOb7, Drbm6vBRDJkn0NGJ, 871pw3rLjNPKgqA0, 46z9VPE4wnkrvEJR) All can be viewed at auto-research-funds/wiki/Mind-Upload-EEG-RQ-Grant-Map.md, grant_queue/*.yaml, and auto-research-funds/grant_eeg_dataset_match.csv.

1. U8-1 Closed loop delay tolerance range identification: Main data = 6, application destination G2 (preliminary G5)
2. U8-2 Online calibration and drift countermeasures: Main data = 696, application destination G2 (preliminary G5)
3. U14-1 Fixed granularity and supplementary exam success rate: Main data = 56, application destination G1 (preliminary G3)
4. U14-3 Leak audit operation: Main data = 719, application destination G1 (preliminary G3)
5. U7-2 Time synchronization error audit: Main data = 6, application destination G1 (preliminary G3)
6. U13-1 Separation of semantic agreement and causal agreement: Main data = 509, applicant G1 (preliminary G4)

Reasons for proceeding in this order:

• It is easy to reach "actual KPI + failure conditions + application draft" within 4 weeks with A/B judgment.
• It will be easier to pass the C judgment (system-driven) if you submit it at a later stage based on these 6 actual measurement logs.

• Reconfirmed RQ_TOTAL=60 and A/B/C=17/25/18 and confirmed that they match the number of RQ lines on this page.
• All IDs referenced in the main EEG-DATA column exist in EEG-DATA/eeg_dataset_summary_ja.csv (missing 0).
• All real IDs of grant keys G1-G6 can be viewed on the auto-research-funds side.
• Continue to operate this page: Dig deeper into each case and clarify the key things that need to be verified directly with EEG'' and required things outside of EEG,'' and avoid general summaries.

• After main pull, we reconfirmed RQ_TOTAL=60 and A/B/C=17/25/18 and confirmed that they are consistent with the RQ line on this page.
• We reconfirmed that all EEG data IDs referenced on this page exist (missing 0) in EEG-DATA/eeg_dataset_summary_ja.csv.
• Grant real ID (GR-2026-013, GR-2026-014, 9Lx4dPK6a4k2gOb7, Drbm6vBRDJkn0NGJ, 871pw3rLjNPKgqA0, 46z9VPE4wnkrvEJR) is auto-research-funds I reconfirmed that all items can be viewed on the side.
• The policy will continue: Instead of a general overview, we will maintain ``1 RQ = 1 verification proposition = 1 application theme = 1 main data'' and give priority to in-depth description of each RQ.

• We re-counted the RQ line in mind-upload/wiki/mind-upload-rq60-deep-focus-notes.md and confirmed 60 (A/B/C=17/25/18).
• In all RQ lines, Application theme (grant key) has been assigned to one of G1-G6, and we confirmed that there is a blank 0.
• There are 24 types of data IDs referenced in the main EEG-DATA column (6, 11, 13, 16, 19, 29, 39, 49, 56, 65, 509, 676, 696, 719, 735, 783, 842, 859, 1011, 1839, 1972, 2412, 3419, 4878), we confirmed the missing 0 on EEG-DATA/eeg_dataset_summary_ja.csv.
• Grant real ID (GR-2026-013, GR-2026-014, 9Lx4dPK6a4k2gOb7, Drbm6vBRDJkn0NGJ, 871pw3rLjNPKgqA0, 46z9VPE4wnkrvEJR) is auto-research-funds I confirmed that it can be re-referenced under it.
• The policy remains unchanged. Rather than general summaries, we will continue to deeply dig into ``1 RQ = 1 verification proposition = 1 application theme = 1 main data.''

• Policy: We did not comprehensively update all RQs, but fixed the 6 RQs that are most likely to be implemented "in depth one by one" to the execution plan.
• Target: U8-1, U8-2, U14-1, U14-3, U7-2, U13-1 (maintains the previously fixed priority order).

RQ	Nucleus that can be solved with EEG-DATA (fixed for this time)	Two-week initial experiment (one experiment each)	Application theme (grant key)	EEG-DATA to be used (main/preliminary)	Pass/fail gate (conditions for proceeding to the next)
U8-1	Identify the relationship between stability rate and recovery time when closed-loop delay is manipulated.	The same preprocessing pipeline was fixed with a delay of 0/50/100/150ms in the first week. In the second week, we remeasured the stability rate, recovery time, and abnormal stoppage rate and created a threshold table.	Identification of closed-loop delay tolerance range (G2 main / G5 spare)	6 / 1972	Relative decrease in stability rate is within 20% and abnormal stop rate is within the prior upper limit.
U8-2	Quantify the impact of online calibration intervals on performance retention.	Three conditions of calibration interval (every trial/every block/no fixation) were set in the first week. In the second week, we simultaneously evaluate the amount of drift and recalibration cost and determine the upper limit of operation.	Optimization of concept drift countermeasures (G2 main / G5 spare)	696 / 2412	Improvement in performance maintenance rate is statistically significant with respect to increase in calibration cost.
U14-1	Measure the correspondence between the fixed granularity (data/code/environment) of supplementary trial contracts and the reproduction success rate.	In the first week, re-execute at three granularity: "Fixed data only", "Fixed data + code", "Fixed data + code + environment". Quantify the recall rate difference and the man-hour difference in the second week.	Minimum granularity definition of reproduction contract (G1 main / G3 reserve)	56 / 719	Recall rate meets target value and man-hour increment is within acceptable upper limit.
U14-3	Automatically detect subject leaks and time series leaks, and enable auditing of leaderboard evaluations.	Implemented leak detection rules (subject/session/time-window) in the first week. Apply it to the existing partition in the second week and create an audit report of the performance difference after re-partitioning.	leaderboard audit operation (G1 main / G3 spare)	719 / 56	Zero leak detection, performance deterioration after repartition can be explained.
U7-2	Estimate the jitter distribution of LSL synchronization error and determine the operational warning threshold.	Fixed offset/jitter extraction processing in the first week. Calculate the event rate exceeding the threshold in the second week and set the remeasurement trigger conditions as rules.	Synchronization error automatic warning operation (G1 main / G3 spare)	6 / 696	Reproducibility of threshold exceeded events and false alarm rate of remeasurement rules are within tolerance.
U13-1	Separately evaluate the discrepancy between semantic matching and causal matching using the same decoder.	In the first week, compute semantic match score and intervention-response match score with separate metrics. In the second week, register deviation cases as failure conditions and prohibit pass decisions based only on imitation score.	Separate evaluation of semantic restoration/causal reproduction (G1 main / G4 preliminary)	509 / 13	Cases where only semantic matching is high can be identified and cases where there is no causal matching can be avoided.

1. U8-1 (delay tolerance area)
2. U8-2 (Calibration interval and drift)
3. U14-1 (Reproduction contract granularity)
4. U14-3 (Leak Audit)
5. U7-2 (Synchronization error audit)
6. U13-1 (Separation of semantic matching/causal matching)

• Re-fix 1RQ=1 verification proposition=1 application theme=1 main data as 6RQ.
• RQs that do not reach "first experiment results + pass/fail gate" within two weeks will not be carried over to the next run, but will be switched to another RQ with a clear reason for failure.
• C Judgment RQ is not subject to this time (the policy is to accumulate technology demonstration first).

• After reflecting the latest main of mind-upload / EEG-DATA / auto-research-funds, re-inspect the RQ line of the deep digging note.
• Reconfirmed RQ_TOTAL=60, A/B/C=17/25/18, Application theme (grant key) missing 0, Main EEG-DATA missing 0.
• The policy remains the same: 1RQ = 1 verification proposition = 1 application theme = 1 main data, and each issue will be treated individually and in depth.

• Revalidated deep-focus operation with the same policy: avoid generic summaries and keep 1RQ = 1 verification proposition = 1 application theme = 1 main data.

• Reconfirmed RQ_TOTAL=60 and A/B/C=17/25/18 against mind-upload/wiki/mind-upload-rq60-deep-evaluation-cards.md.
• Refreshed the fixed deep pack to a non-overlapping 6RQ set for this run: U0-1, U3-3, U4-2, U7-5, U11-3, U14-2.
• Re-fixed the operational policy: avoid broad summaries and enforce 1RQ = 1 verification proposition = 1 application theme = 1 main data.
• Updated synchronized map pages: mind-upload/wiki/mind-upload-eeg-rq60-feasibility-and-funds.md and auto-research-funds/wiki/Mind-Upload-EEG-RQ-Grant-Map.md.
• Rechecked RQ coverage on this page against research_harvest_50.md and related RQ60 pages (coverage maintained).
• Revalidated grant keys G1-G6 and source resolvability in grant_queue/auto-research-funds.
• Confirmed EEG numeric IDs are present in latest EEG-DATA CSV but no longer reliable as fixed dataset identifiers after reindex/remap.
• Fixed submission rule for this run: canonical references must be written using DOI + dataset name + access class in auto-research-funds/wiki/Mind-Upload-EEG-RQ-Grant-Map.md.

• Pulled latest main before this run and revalidated this page in deep-by-RQ mode (1RQ = 1 verification proposition = 1 application theme = 1 main data).
• Recounted unique RQ keys on this page and reconfirmed RQ_TOTAL=60 with no missing key.
• Recalculated judgment split from mind-upload-eeg-rq60-feasibility-and-funds.md and reconfirmed A/B/C=17/25/18.
• Revalidated referenced EEG IDs (24 IDs: 6, 11, 13, 16, 19, 29, 39, 49, 56, 65, 509, 676, 696, 719, 735, 783, 842, 859, 1011, 1839, 1972, 2412, 3419, 4878) exist in latest EEG-DATA/eeg_dataset_summary_ja.csv (missing=0).
• Revalidated grant real IDs G1-G6 (GR-2026-013, GR-2026-014, 9Lx4dPK6a4k2gOb7, Drbm6vBRDJkn0NGJ, 871pw3rLjNPKgqA0, 46z9VPE4wnkrvEJR) remain resolvable in grant_queue and auto-research-funds mapping docs.

• Pulled latest main in auto-startup, mind-upload, auto-research-funds, and EEG-DATA before this run.
• Recounted unique Ux-RQy keys on this page and reconfirmed RQ_TOTAL=60 (missing=0, duplicate=0).
• Rechecked judgment split against mind-upload-eeg-rq60-feasibility-and-funds.md and reconfirmed A/B/C=17/25/18.
• Revalidated the 24 referenced EEG IDs (6, 11, 13, 16, 19, 29, 39, 49, 56, 65, 509, 676, 696, 719, 735, 783, 842, 859, 1011, 1839, 1972, 2412, 3419, 4878) in latest EEG-DATA/eeg_dataset_summary_ja.csv (missing=0).
• Revalidated grant real IDs G1-G6 (GR-2026-013, GR-2026-014, 9Lx4dPK6a4k2gOb7, Drbm6vBRDJkn0NGJ, 871pw3rLjNPKgqA0, 46z9VPE4wnkrvEJR) in grant_queue and auto-research-funds/wiki/Mind-Upload-EEG-RQ-Grant-Map.md.
• Execution policy is unchanged: no generic sweep, only deep one-by-one operation with 1RQ = 1 verification proposition = 1 application theme = 1 main data.

• Pulled latest main in auto-startup, mind-upload, auto-research-funds, and EEG-DATA before this run.
• Rechecked deep-by-RQ operation (1RQ = 1 verification proposition = 1 application theme = 1 main data) and reconfirmed RQ_TOTAL=60 with A/B/C=17/25/18.
• Revalidated the 24 referenced EEG IDs (6, 11, 13, 16, 19, 29, 39, 49, 56, 65, 509, 676, 696, 719, 735, 783, 842, 859, 1011, 1839, 1972, 2412, 3419, 4878) in latest EEG-DATA/eeg_dataset_summary_ja.csv (missing=0).
• Revalidated grant real IDs G1-G6 (GR-2026-013, GR-2026-014, 9Lx4dPK6a4k2gOb7, Drbm6vBRDJkn0NGJ, 871pw3rLjNPKgqA0, 46z9VPE4wnkrvEJR) in grant_queue and auto-research-funds map docs.
• Refreshed EEG inventory baseline from current CSV: TOTAL=28,152, immediate=23,871, uncertain_public=3,692, account_required=585, restricted=4.
• Execution rule is unchanged: avoid generic sweep and maintain one-by-one deep operation for every RQ.

• 本ページの RQ 行を再計数し、RQ_TOTAL=60、A/B/C=17/25/18 を再確認しました。
• main EEG-DATA 参照IDは EEG-DATA/eeg_dataset_summary_ja.csv で全件実在（欠落 0）を確認しました。
• 助成実ID G1-G6 は grant_queue と auto-research-funds/wiki/Mind-Upload-EEG-RQ-Grant-Map.md で再参照可能でした。
• 継続方針: 汎用要約ではなく 1RQ=1検証命題=1応募テーマ=1主データ を維持します。

1. U4-2（主データ 56 / G1）
2. U7-5（主データ 676 / G1）
3. U8-4（主データ 2412 / G2）
4. U11-1（主データ 859 / G2）
5. U13-3（主データ 509 / G1）
6. U14-6（主データ 719 / G1）

• main pull 後に mind-upload/wiki/mind-upload-rq60-deep-focus-notes.md を再検証し、research_harvest_50.md 由来の RQ_TOTAL=60（missing=0, duplicate=0）を再確認しました。
• 判定内訳は A/B/C=17/25/18 を再計数で確認しました。
• 60RQの全行で Application theme (grant key) と Main EEG-DATA が埋まっていることを確認しました。
• Main EEG-DATA の参照IDは EEG-DATA/eeg_dataset_summary_ja.csv 上で全件実在（missing=0）を再確認しました。
• 助成実ID G1-G6（GR-2026-013, GR-2026-014, 9Lx4dPK6a4k2gOb7, Drbm6vBRDJkn0NGJ, 871pw3rLjNPKgqA0, 46z9VPE4wnkrvEJR）は grant_queue と auto-research-funds 側マップで再参照可能でした。
• 方針は継続: 汎用横断要約ではなく 1RQ=1検証命題=1応募テーマ=1主データ を固定し、1件ずつ深掘りします。

1. U0-1（同一性KPI二軸ROC、主データ 11、応募 G2）
2. U1-2（不確実性伝播校正、主データ 56、応募 G1）
3. U4-2（最小介入因果主張、主データ 56、応募 G1）
4. U8-1（閉ループ遅延許容域、主データ 6、応募 G2）
5. U13-1（意味一致/因果一致の分離、主データ 509、応募 G1）
6. U14-3（リーク監査運用、主データ 719、応募 G1）

• mind-upload / EEG-DATA / auto-research-funds を再突合し、RQ_TOTAL=60、A/B/C=17/25/18 を維持したまま、2週間で初回結果まで到達しやすい 6RQ を実行カード化しました。
• 汎用横断ではなく、1RQ=1検証命題=1応募テーマ=1主データ で、各RQに KPI 失敗条件 提出最低成果物 を明示しました。
• このrunでは C 判定RQを拡張せず、まず A/B の実測ログを積む順序に固定しました。

RQ	最初の実験KPI	失敗条件	提出最低成果物	応募先（第一/予備）
U14-1	追試成功率 再現率差 追加工数	最厳格固定条件でも追試成功率が目標未達	Supplementary test operation report (including negative cases)	G1 / G3
U7-2	clock offset jitter p95 閾値超過率	閾値超過イベントの再現性不足、または誤警報率過大	Time synchronization audit report (offset/jitter distribution)	G1 / G3
U14-3	被験者リーク検出率 時系列リーク検出率 再分割後性能差	再分割後もリーク0件を達成できない	Leak audit results and repartition trail	G1 / G3
U8-1	安定率 回復時間 異常停止率	遅延増加で安定率低下が大きく許容域未確定	Closed loop safety KPI dashboard	G2 / G5
U8-2	性能維持率 ドリフト量 再較正コスト	再較正コスト増に対して性能維持率改善が有意でない	Closed loop safety KPI dashboard	G2 / G5
U13-1	意味一致率 因果一致率 乖離ケース率	意味一致高・因果一致低ケースを分離できない	Intervention protocol and rebuttal condition definition	G1 / G4

• 作業開始前に親リポジトリ auto-startup の main で git pull --ff-only origin main を実行し、Already up to date を確認しました。
• mind-upload/wiki/mind-upload-rq60-deep-evaluation-cards.md を基準に RQ_TOTAL=60、A/B/C=17/25/18 を再計数し、整合を確認しました。
• 本ページの方針（汎用横断要約ではなく 1RQ=1検証命題=1応募テーマ=1主データ で1件ずつ深掘り）を継続固定しました。
• 参照EEG ID（6, 11, 13, 16, 19, 29, 39, 49, 56, 65, 509, 676, 696, 719, 735, 783, 842, 859, 1011, 1839, 1972, 2412, 3419, 4878）を EEG-DATA/eeg_dataset_summary_ja.csv の ID/旧ID と照合し、未解決ID 0 を確認しました。
• 助成キー G1-G6（GR-2026-013, GR-2026-014, 9Lx4dPK6a4k2gOb7, Drbm6vBRDJkn0NGJ, 871pw3rLjNPKgqA0, 46z9VPE4wnkrvEJR）の実在を再確認しました。

• 作業開始前に親リポジトリ auto-startup の main で git pull origin main を実行し、Already up to date を確認しました。
• mind-upload/wiki/mind-upload-rq60-deep-evaluation-cards.md の | Ux-y<br> 行を再計数し、RQ_TOTAL=60（欠損・重複 0）を再確認しました。
• 判定内訳 A/B/C=17/25/18 を再確認し、全60RQで 1RQ=1検証命題=1応募テーマ=1主データ の深掘り運用を維持しました。
• 3ページ（mind-upload 側2ページ + auto-research-funds 側1ページ）で参照されるEEG ID一意集合（6, 11, 13, 16, 19, 29, 49, 56, 65, 509, 676, 696, 719, 735, 783, 842, 859, 1972, 2412）を EEG-DATA/eeg_dataset_summary_ja.csv の ID/旧ID と照合し、未解決ID 0 を確認しました。
• 助成キー G1-G6（GR-2026-013, GR-2026-014, 9Lx4dPK6a4k2gOb7, Drbm6vBRDJkn0NGJ, 871pw3rLjNPKgqA0, 46z9VPE4wnkrvEJR）を grant_queue と auto-research-funds 側で再照合し、全件実在を確認しました。
• 方針は不変で、汎用横断要約ではなく「1件ずつ深く理解して書く」運用を継続します。

• RQ_TOTAL=60、A/B/C=17/25/18 を維持したまま、既存6RQバッチとは重複しない新規6RQを追加し、実行順を固定しました。
• 追加した6RQは U0-2 U1-3 U4-4 U11-1 U14-5 U15-4 で、KPI 失敗条件 提出最低成果物 を明示しました。
• U15-4（C判定）を含め、制度系RQでも技術監査ログとして2週間で到達できる入口を固定しました。
• 運用ルールは継続して 1RQ=1検証命題=1応募テーマ=1主データ を適用します。

• Updated the fixed deep-dive batch without broad summary expansion; selected only U0-3/U1-2/U7-3/U8-4/U13-2/U15-2.
• Kept RQ_TOTAL=60, A/B/C=17/25/18, and the same G1-G6 operating keys.
• Locked the submission rule as Dxx + DOI + dataset name + access class (numeric ID is search-only legacy guidance).
• Focused deliverable style remains: 1RQ = 1 verification proposition = 1 application theme = 1 primary dataset.

• mind-upload/wiki/mind-upload-rq60-deep-evaluation-cards.md を正本として再突合し、RQ_TOTAL=60、A/B/C=17/25/18 を再確認しました。
• 本runは汎用横断の追加を行わず、非重複6RQ（U1-3/U3-5/U7-4/U8-5/U11-3/U15-1）を 1RQ=1検証命題=1応募テーマ=1主データ で固定しました。
• 参照主データID（56,13,676,6,735,49）は EEG-DATA/eeg_dataset_summary_ja.csv で実在（欠落 0）を確認しました。
• 助成接続は G1-G6 実ID系（GR-2026-013, GR-2026-014, 9Lx4dPK6a4k2gOb7, Drbm6vBRDJkn0NGJ, 871pw3rLjNPKgqA0, 46z9VPE4wnkrvEJR）を維持しました。

1. U7-4（再計測/除外判定木を固定）
2. U1-3（逆解法乖離の採択規則）
3. U8-5（オーバーライド安全KPI）
4. U11-3（理論比較I/O仕様）
5. U3-5（境界拡張proxyと外部依存境界）
6. U15-1（神経データ機微性の制度接続）

• mind-upload/wiki/mind-upload-rq60-deep-evaluation-cards.md を正本として再突合し、RQ_TOTAL=60、A/B/C=17/25/18 を再確認しました。
• 本runは汎用横断の追加を行わず、非重複6RQ（U0-1/U1-1/U4-3/U8-6/U12-3/U14-4）を 1RQ=1検証命題=1応募テーマ=1主データ で固定しました。
• 参照主データID（11,56,19,6）は EEG-DATA/eeg_dataset_summary_ja.csv の ID/旧ID で実在（欠落 0）を確認しました。
• 助成接続は G1-G6 実ID系（GR-2026-013, GR-2026-014, 9Lx4dPK6a4k2gOb7, Drbm6vBRDJkn0NGJ, 871pw3rLjNPKgqA0, 46z9VPE4wnkrvEJR）を維持しました。

1. U1-1（逆問題の事前分布比較と採択規則）
2. U0-1（同一性KPIの観測一致/介入一致二軸化）
3. U14-4（Model Card / Dataset Card の契約要件化）
4. U4-3（反事実誤差による理論比較固定）
5. U8-6（再学習頻度と安全余裕のトレードオフ固定）
6. U12-3（心理的連続性監査ログの制度接続準備）

• Pulled the latest main in auto-startup, mind-upload, EEG-DATA, and auto-research-funds before editing (0 0, Already up to date in every repo).
• Recounted ### Ux-y headings in mind-upload/wiki/mind-upload-rq60-rq-by-rq-deep-dossiers.md and reconfirmed RQ_TOTAL=60 (missing=0, duplicate=0).
• Recomputed the judgment split from mind-upload/wiki/mind-upload-rq60-deep-evaluation-cards.md and reconfirmed A/B/C=17/25/18.
• Rechecked the referenced legacy ID set (6, 11, 13, 16, 19, 29, 39, 49, 56, 65, 509, 676, 696, 719, 735, 783, 842, 859, 1011, 1839, 1972, 2412, 3419, 4878) in EEG-DATA/eeg_dataset_summary_ja.csv (旧ID, missing=0).
• Corrected the U11-1 anchor drift: the stale D12=10.21227/qqj2-r702 reference was invalid and is now aligned to D14 (10.6084/m9.figshare.1485719) with D16/D17 as support anchors.
• Corrected the U12-5 minimum deliverable drift: this RQ now consistently points to Branch-based audit log design document instead of a synchronization-only report.
• This correction pass did not replace the solve-first execution pack. The active 2026-03-29 pack remains the synced six-RQ set below.

• Synced this note to the existing 2026-03-29 solve-first pack already fixed in docs/wiki/mind-upload_eeg_funding_verification_2026-03-29_0137-jst.md, mind-upload/wiki/mind-upload-eeg-rq60-feasibility-and-funds.md, and auto-research-funds/wiki/Mind-Upload-EEG-RQ-Grant-Map.md.
• Kept the same evidence rule: no broad sweep, no claim expansion, and no numeric-ID-only citation in submission documents.
• Reconfirmed the execution criterion for this sync: prioritize RQs that are easiest to turn into both an initial EEG experiment and a grant-ready problem statement without relaxing the 1RQ = 1 verification proposition = 1 application theme = 1 primary dataset rule.

1. U1-1 - prior-family benchmark with known stimulation ground truth
2. U4-2 - minimum causal claim fixed before analysis
3. U7-1 - BIDS extension contract for synchronization/QC/stimulus logs
4. U11-1 - rank-preserving theory comparison across anesthesia/sleep/DoC
5. U13-5 - perception-to-recall branch point for mimicry/mechanism separation
6. U14-4 - Model Card / Dataset Card wired into the reproducibility contract

• Pulled the latest main in auto-startup, mind-upload, EEG-DATA, and auto-research-funds before editing and reconfirmed Already up to date.
• Kept RQ_TOTAL=60, A/B/C=17/25/18, and the submission evidence rule Dxx + DOI + dataset name + access class.
• Chose only six RQs that directly tighten the public mind-upload program instead of widening the summary surface again.
• Synchronized the public batch in issue.html to the same six RQs, while keeping the stronger dataset/KPI/stopping-rule details inside the wiki.

RQ	Why this is solve-first now	Primary D anchor	Grant lane	First KPI to lock now	External dependency kept separate
U0-1	identity claims drift upward too easily unless observation-match and intervention-response match are forced into the same public pass/fail gate.	D02 + support D11/D23	G2 / G3	observation-match AUC / intervention-response match rate	Legal/philosophical identity criteria
U1-2	inverse-problem claims are not credible unless uncertainty propagation survives repeated conductivity, missing-electrode, and noise perturbations.	D08 + support D11/D19	G1 / G4	interval coverage	Head-model identification conditions
U4-4	causal-equivalence claims should fail fast, so falsification thresholds must be locked before reading any result.	D05 + support D08/D19	G1 / G4	PEHE + sign-reversal rate	Common I/O specification and intervention protocol
U7-1	synchronization/QC/stimulus logs are the shortest path to a public reproducibility claim that still stays conservative.	D11 + support D15/D23	G1 / G3	required-field fill rate	Institutional adoption and governance
U8-4	a closed-loop program without explicit fail-safe latency metrics is not ready for a public safety claim.	D20 + support D01/D21	G2 / G5	abnormal-detection latency p95	Site shutdown ownership and operational safety SOP
U11-1	awareness-index comparison is publicly safer when it is reduced to rank preservation across shared I/O rather than theory-winning language.	D14 + support D16/D17	G2 / G4	rank-preservation rate	Theory adoption and clinical responsibility split

1. U7-1
2. U1-2
3. U4-4
4. U0-1
5. U8-4
6. U11-1

• Pulled the latest main in auto-startup, mind-upload, EEG-DATA, and auto-research-funds, plus master in auto-startup.wiki, and reconfirmed Already up to date in every target repo.
• Rechecked the active public six-RQ anchors against the local EEG-DATA/wiki/datasets cards instead of relying only on routing tables.
• D11 is currently the strongest anchor in the local mirror because its card exposes BIDS 1.7.0, CC0, 47 sessions, 186 EEG runs, 31 EEG + 5 misc, 5000 Hz, and even the README=24 / participants.tsv=25 mismatch that still has to stay visible in any serious audit.
• D02 remains a valid cross-session identity anchor, but the current local card only fixes a conservative minimum set (10 participants, BrainVision / BrainAmp / actiCHamp, motor-imagery framing, cross-session transfer route). It is not yet a substitute for a richer intervention log or a theory-level identity dossier.
• D05, D08, D14, and D20 remain valid route anchors in the current mirror, but their autogenerated local cards are still sparse enough that public claims should stay at the level of what family of test this anchor supports plus which logs are still required, not at the level of a complete protocol/QC dossier.
• Practical consequence for the public batch: keep the current six-RQ selection unchanged, but treat required logs as first-class deliverables before strengthening the claim language around causal equivalence, inverse-problem coverage, closed-loop safety, or awareness-index comparison.

• Rechecked the same six public RQs directly against the local dataset cards and the still-stale live https://mind-upload.com/issue.html output before changing the public source page.
• D08 and D11 remain the strongest current repository anchors: D08 because it exposes a ground-truth source-localization route with intervention, and D11 because it exposes the richest synchronization/BIDS/session-run audit surface in the local mirror.
• D02 and D20 remain usable conservative route anchors, but only when the public claim is kept at the level of cross-session identity drift or closed-loop latency/safety auditing rather than a stronger intervention or deployment claim.
• D05 works as the minimum perturbation anchor for U4-4, but the stronger causal-equivalence reading should stay tied to D08-class ground truth or equivalent intervention detail.
• D14 works as the minimum anesthesia-side anchor for U11-1, but the cross-state wording should stay paired with sleep / DoC support anchors such as D16/D17 rather than being read as a standalone winner.
• Practical consequence for this run: keep the public six-RQ batch unchanged, strengthen the wording around support anchors + required logs, and do not widen the claim ceiling.

• Rechecked the same six public RQs against the live https://mind-upload.com/issue.html wording, the local EEG-DATA dataset cards, the official funding pages, and the live Todoist task inventory before changing any public-facing text.
• The public six-RQ batch stays unchanged: U13-2, U13-5, U0-2, U0-3, U14-4, and U14-2 still form the narrow inner-speech decoding + identity drift + reproducibility auditing route, and the public page still presents them as a route-setting update rather than a solved-claim update.
• The current conservative anchor proof in the local mirror is now fixed more explicitly as D10 -> 000016 3M-CPSEED (OpenNeuro ds006465 v2.0.0, immediate), D03 -> 011012 Data of Multimodal Fusion System for Cognitive Load Assessment in Psychological Experiments: An EEG and Eye-Tracking Study (DOI route, subscription-gated support), D01 -> 016105 Acquisition delay of wireless EEG (DOI route, subscription-gated support), and D02 -> 005558 CSTE (DOI route, subscription-gated support).
• D03, D01, and D02 still have IEEE DataPort-backed routes in the local mirror that now need to be read as subscription-gated support, so the public route no longer treats them as immediate public anchors.
• Funding and task handling also stay conservative in this run: Kura Fund remains the practical first route for U13/U14, the Nakatani Foundation graduate scholarship remains the main student-facing bridge for U13/U0, the Kashinome Scholarship stays a runway-only side route, the Nakatani Foundation research grant stays in watch mode, and no new Todoist task is added because the required start/deadline tasks already exist.
• Practical consequence for this run: keep the batch unchanged, keep the current solve-first order, and treat the next public-facing progress threshold as finished minimum artifacts + required logs, not as a stronger claim ceiling.