BASIN MV RESEARCH JOURNAL
← Research record

OTB21 / EVIDENCE SNAPSHOT 2026-11-05

OTB21 · Learn while retaining the past

Public reproducible study

Does a retention condition transfer to changed phases and query durations?

What was found

The guard learned the new relation and retained the unchanged one in 128/128 conditions in every duration group. At the shortest duration, immediate and fixed waiting retained only 64/128.

128/128Guard · old relation retained
30,040.25 stepsGuard · shortest-query total time

Task, conditions and denominator

1,152 paired conditions: eight saved memories × four shared phase draws × two old contexts × two new contexts × three durations × three policies. Not 1,152 independent worlds.

Controls and failures

New relations were learned in every group. At 4,096-step queries, fixed waiting also retained 128/128. For the shortest query, mean total time was 14,632 (immediate), 18,728 (fixed) and 30,040.25 (guard).

External dependencies and limits

No matched-time superiority over a well-tuned fixed wait has been established. Error calculation, routing, state checks and preparation remain external. This is the first fully bundled public study.

Read the interactive results & reproduction guide ↗

Read the primary report

The source snapshot is in Japanese; this page is a bilingual editorial summary. A report snapshot is not an independently reproduced result. Original relative links may require files from the source repository.

Open the preserved report (Japanese) ↗
Source identity and verification scope

Source commit and path:

https://github.com/type37c/oscillator-mind/blob/1c70be3bc7df1bce568e6dc102dde7a573e00597/results/otb21_readiness_history_validation/REPORT.md

The original development repository may require access. The report snapshot above is public. Only OTB21 has a full runnable public package; other entries are report-level evidence.

SHA-256: 413e21a930e59fa035c4073034a67ae7df0fd3e9c479637a0e0c820c34589d15

Source manifest ↗

Concepts to learn first

WHAT THIS RESULT MAKES US ASK

Which state information is necessary for retention at a matched time budget?

Discuss a test or counterexample ↗

Textbook revision 1 · Evidence snapshot: 05 Nov 2026 · Changes & corrections ↗