- gadm04.py: cohort capability split into durable (education-formed, slow
depreciation, birth dilution) and current components; learning begets
learning; starts identical to GADM-0.3 (tested)
- L3 now supported: education reaches working-age capability with a 17-year
half-time and 3x the effect; L1 still holds, more weakly
- life_cycle.py selectable per model; GADM-0.3 results unchanged
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 1494246@bnt-lap001
Assistant-Session: caf6de6f-f9e4-47ce-a584-49baa271fa4b
- gadm03c.py: four cohorts with demography, cohort capability, mentoring,
care load on attention; stationary at replacement fertility (tested)
- life_cycle.py: L1 supported after fixing an inert mentoring term; L2
conditional on who carries care per person, care relief via participation
claims protects carers' attention; L3 not supported (H lacks a durable
component)
- GADM-0.3.md complete; ASSEMBLY-WP-0006 (GADM-0.4, proposed residual)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 1494246@bnt-lap001
Assistant-Session: caf6de6f-f9e4-47ce-a584-49baa271fa4b
- aext recalibrated to 0.275 against ATUS and social-media time-use data;
baseline attention headroom now ~0.49 instead of rising to 0.65
- T03/T04 stages stay reproducible (t03_params, t04_params; verified)
- organism_grid.py: factorial analysis; lineage ownership is the strongest
organism lever, then membrane control, then attention protection (which
acts on A, not on power); lineage x membrane complementary at baseline;
organism governance without attention protection raises attention draw
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 1494246@bnt-lap001
Assistant-Session: caf6de6f-f9e4-47ce-a584-49baa271fa4b
- Wealth-concentration stock W with wealth-tax lever; converts into market
and political channels only, moderated by checks (Q*S)
- wealth_on switch reproduces the T03 model exactly (verified)
- wealth.py: P3 supported and robust to conversion strength (full wealth
tax lowers P by only 0.01-0.05 because membrane/lineage control does not
run on wealth); effect on reachable channels depends on k_w
- Tests for wealth switch and tax direction
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 1494246@bnt-lap001
Assistant-Session: caf6de6f-f9e4-47ce-a584-49baa271fa4b
