Home / Research / Research and Explorations

Identity Research, mapped to the engine (what is built, what is a gap)

Updated Jun 27, 2026 · Affirmology_IdentityResearch_StructureMap_v1.md

Summary. Companion to AffirmologyIdentityResearchv1.md. For each technique: where it lives today (structure / prompt / production rule), status, and the concrete next move. Status legend: SOLID (in the prompt and enforced), PARTIAL (present but not enforced or only in

Identity Research, mapped to the engine (what is built, what is a gap)

Companion to Affirmology_IdentityResearch_v1.md. For each technique: where it lives today (structure / prompt / production rule), status, and the concrete next move. Status legend: SOLID (in the prompt and enforced), PARTIAL (present but not enforced or only in some audios), GAP (not implemented). Grounded in the actual engine prompts (SCRIPT_SYSTEM_PROMPT_ORIGIN, SCRIPT_SYSTEM_PROMPT_ORIGIN_RECOGNITION, SCRIPT_SYSTEM_PROMPT_FLEX) and api/structures.py, 2026-06-26.

Tier 1 - the non-negotiables

# Technique Where it lives Status Next move
1 Self-Relevance Personalization Every prompt forces "ONLY this person's chart"; the personalization brief + full chart JSON are injected per render SOLID Keep. This is the spine and it is enforced.
2 Prediction-Error Reduction (name the pattern in the first 60s) Recognition opens by "NAMING THEIR ACTUAL PATTERN, not a greeting"; Origin opens identity-first but leads with an aspirational line PARTIAL (SOLID in Recognition) If the A/B says recognition wins, fold the "open on their real pattern" rule back into Origin.
3 First-Person Present-Tense "I am" All identity prompts mandate first person; "I am [name]" required in the first 3 sentences SOLID Keep.
4 Recognition over instruction Recognition is built entirely on this ("I am the one who...", never "you should"); Origin uses it but allows becoming SOLID (Recognition) / PARTIAL (others) This is the variable the_recognition isolates. Decide via the A/B.
5 Listener Language Mirroring Recognition prompt explicitly says "MIRROR their own words/metaphors when provided" GAP in practice We do not yet CAPTURE the person's own words. Add an intake / chat "their words" field and thread it into the brief. Highest-leverage unbuilt move.

Tier 2 - the depth layer

# Technique Where it lives Status Next move
6 Mechanism over Label Origin/Recognition both say translate placements to felt experience, never keyword salad; HD numbers OUT SOLID Keep; strongest in readings (The Mirror).
7 Chart-Specific Symbol metaphor Same prompts; "translate every placement into visceral metaphor" SOLID Keep.
8 Believable Gap Incrementalism Not stated in any prompt GAP Add an "audit for overshoot, compress by one step" rule to the transformation/flex prompt (not Origin, which is already-true by design).
9 Values-Aligned Reframe Not explicit GAP Add to the flex/transformation prompt: tie the new identity to a value they already revere, repeat the link 2-3x. Good for confidence/abundance + founder audios.
10 Future-Oriented (becoming) Origin's open-loop close; flex prompt SOLID in Origin This is exactly what The Recognition dials DOWN. The A/B tells us the right dose of becoming vs recognition.
11 Body-Anchored Integration The hand-on-heart breath + gratitude beat (required in Origin and Recognition) PARTIAL Beat exists; the "anchor each placement to a body location" idea (Sun in hands, Moon in chest) is NOT enforced. Add as an option for transformation audios.

Tier 3 - delivery and dose (production, not script)

# Technique Where it lives Status Next move
12 Prosodic Safety (slow, low, warm) C26 pacing (pace/bed_profile per structure); Charlotte at the safe 0.96 speed PARTIAL The safe-speed + bed is in; the "lower register, 8-12s phrases, warmth at 100-500Hz" is not a measured target. Optional mix pass.
13 Slow Voice for Theta (90-110 wpm) Effective rate lands ~117-146 wpm depending on the breath beats PARTIAL Our identity audios already run slow; if we want true theta (90-110), trim word count and lean on inter-sentence silence, not just the speed knob.
14 Sleep-Consolidation Dosing (nightly, 3-8 weeks) The app's auto-replay + a "play nightly" program; not a script rule GAP (product, not engine) Build the nightly cadence into the app program + before-bed structures; this is the retention spine.
15 Emotional Self-Referential Anchor (closing incantation + long exhale, delta sleep variant) Origin's gratitude-breath close is adjacent PARTIAL Add a short repeatable closing incantation to transformation audios, with a delta-bed variant for the before-bed versions.

The three highest-leverage gaps to close next

  1. Capture the person's own words (Tier 1 #5, Listener Language Mirroring). The prompt is ready; the intake/chat field is missing. This is the biggest unbuilt "this gets me" lever.
  2. The A/B itself (Tier 1 #4 vs Tier 2 #10): the_recognition vs Origin decides the recognition-vs-becoming dose. Run it (scripts rendered; audio for both exists).
  3. Nightly dosing in the app (Tier 3 #14): one listen is a thought, nightly for weeks is a prior shift. This is the program/retention spine, currently a product gap.