Research: 10 autonomous-research avenues for brainctl

[TSchonleber]

Summary

Generative-end memo answering the second half of the overnight ask: 'autonomous avenues of research'. While the brain-region codification chain (PRs #121-#124) was codification of canonical neuroanatomy, this is generative — staking out 10 directions brainctl could explore that don't already have a canonical answer.

Each avenue:

• Connects to at least one existing brainctl subsystem
• Has a concrete empirical probe against live brain.db
• Has a Phase 1 schema sketch
• Includes an explicit research question to answer before committing

The 10 avenues

1. Sleep architecture as first-class state machine (NREM/REM staging on dream_cycle)
2. Memory aging as synaptic tagging-and-capture
3. Cross-modal binding (claustrum-analog over multiple retrieval modalities)
4. Multi-agent brain federation (shared bg_striatal_weights)
5. Connectome as first-class graph (which subsystems talk to which)
6. Dream-as-hypothesis-testing (active validation, not passive)
7. VTA/SNc as first-class dopamine source
8. Septum + theta rhythm as hippocampal pacemaker
9. Colliculi as orienting-salience pre-cortex
10. Allostatic load → trust decay correlation

Tractability ranking

• Highest (sit on existing infra): 1, 2, 5
• Most ambitious (new coordination): 4, 6
• Speculative probes (worth a check, not commitment): 3, 7, 8, 9, 10

Rejection log

Ideas that didn't survive write-through: glial cells, quantum consciousness, mirror neurons (ToM covers), pontine nuclei (cerebellum exists).

Pairs with tonight's chain

• PR Locus Coeruleus Phase 1: schema + read+CRUD tools (issue #116 follow-up) #121 LC, PR Nucleus Basalis Phase 1: ACh attention broadcaster (issue #116 follow-up) #122 NB, PR ARAS Phase 1: ascending reticular activating system (global arousal gate) #123 ARAS, PR Habenula Phase 1: lateral habenula / anti-reward channel #124 Habenula — the codification side
• This PR — the generative side

🤖 Generated with Claude Code

[TSchonleber]

Superseded by #138 (brainctl-consolidation-v2), which already merged this work into main as part of the v2 tool surface consolidation. Closing as redundant.