Corpus & Archive

1. What the Corpus Is

The psychographic corpus is the complete body of material from which the persona is built. It includes everything the technical team will use for RAG (Retrieval-Augmented Generation): session transcripts, voice recordings, written materials, personal documents, and the Personality Report you produce.

The corpus is not merely a collection of recordings. It is a structured archive of voice evidence — organised to give the AI system maximum access to the subject's authentic language, reasoning patterns, emotional register, and biographical context.

2. Word Count Requirements

The minimum viable corpus depends on the service level. These thresholds are based on what the RAG system requires to produce reliable, contextually grounded responses — not arbitrary targets.

Contribution Sources

Word count is cumulative across all sources: session transcripts, written materials the client provides (letters, journals, emails, social media), and any additional archive material. Session transcripts typically contribute 60–70% of the corpus; written materials fill the rest. The client's own writing — in their own hand — is the most valuable source type.

3. The Legacy Protocol — Sessions 1–10

The Legacy protocol is for living clients who are building their own psychograph — preserving their voice for their descendants, for a future version of themselves, or for the historical record. It consists of 6 to 10 sessions, each 90 minutes, conducted over several weeks. Rushing defeats the purpose. People reveal themselves in layers.

Session 1 — The Surface

Goal: Comfort, trust, first voice sample.

Begin with biography — but never through forms. No "where were you born?" Instead open with sensory, grounded prompts: "Tell me about a morning you remember clearly." "What did your kitchen smell like growing up?" "What is the first sound you associate with home?"

While they speak, you are already working. Note sentence length, vocabulary register, pause patterns, filler words. These are the raw materials of voice.

Session 2 — Relationships

Goal: Map the emotional landscape around them.

Probe the relational world: "Who made you laugh the hardest in your life?" "Who do you avoid, and why?" "Describe your mother in three words. Now tell me why those three." "Who would you call at 3 AM?"

Listen for how they talk about others. Note every "but" after a positive description. Do not resolve contradictions. File them.

Session 3 — Values & Beliefs

Goal: Map the gap between stated and lived values.

Ask what they believe — about God, politics, morality, justice, death. But listen more carefully for the distance between what they say they believe and how they act. A man who says he values honesty but lies to avoid conflict is not a hypocrite — he is a human being. That gap is personality gold.

"What is something you believe that most people would disagree with?" "What is a rule you break regularly without guilt?" "What hill would you die on?"

Session 3B — Introducing the Vault & Emotional Wheel

Goal: Give the subject tools for deeper self-expression.

At the end of Session 3 or beginning of Session 4, introduce the Emotional Wheel (run calibration prompts before showing the wheel — capture raw vocabulary first) and the Vault system. Tell them: "There are things in everyone's life that are too private to share — even here. That's not a problem. We have a system for this."

Session 4 — Fears, Failures & The Decision Timeline

Goal: Access the deeper architecture. Begin mapping the Decision Timeline.

This is the hard session. Approach gently. You have earned trust — do not squander it. "What keeps you up at night?" "What do you regret?" "What is the version of yourself you are most afraid of becoming?"

Pay close attention to defence mechanisms. Do they deflect with humour? Go silent? Become philosophical? The defence is as revealing as what it defends.

Begin the Decision Timeline here (see Module IV for the full method). For each major life event surfaced in Sessions 1–4, classify and probe: POSITIVE, NEGATIVE, and GREY decisions, each with the 5-question calibration layer.

Session 5 — Joy & Passion

Goal: Counter-balance. Find where they light up.

After the weight of Session 4, this session should feel like relief. "When were you happiest?" "What could you talk about for hours without getting bored?" "What do you do when no one is watching?"

Watch for the energy shift. Their posture changes. Speech accelerates or slows. Vocabulary becomes more specific. This is where you find the living centre of the person.

Session 6 — Contradictions

Goal: Produce the most authentic material of the entire process.

Present back their own contradictions. Do this with warmth, not confrontation: "You told me you hate conflict, but you also said you always speak your mind. How do those live together?"

Let them wrestle. Do not resolve it for them. The wrestling is the portrait.

Session 7 — Language & Style

Goal: Refine the voice model. Map the personal lexicon.

Read back their own quotes from previous sessions. Ask: "Is that how you would say it?" "What word would you actually use there?" Probe for register shifts, private vocabulary, code-switching patterns.

Session 8 — The Letter

Goal: Create the emotional anchor of the psychograph.

Ask them to write a letter — to someone they love, to their future self, to someone they lost. Handwritten if possible. Do not read it in front of them. Take it sealed. Read it later. This letter will become the emotional core of the persona — proof that a real person stood here.

Sessions 9–10 — Review & Calibration

Goal: Verify the portrait. Final recordings.

Show the client early persona drafts. "Does this sound like you?" "What is missing?" "What would you never say that this version of you might?"

End with a final recording: 30+ minutes of natural, unstructured conversation. This extended sample is essential for voice modelling. End with acknowledgment — they have done something extraordinary.

4. The Memorial Protocol

The Memorial protocol is for deceased subjects — building a persona from memory, archive, and testimony rather than direct engagement. It is fundamentally different from the Legacy protocol in that the subject cannot consent, cannot correct, and cannot provide new material. The psychographer works with the living who knew them.

Working with Families

Conduct 2–3 sessions with close family members and friends. Different people remember different facets of the same person — and that multiplicity is the point. Use prompts that surface the living texture of the person, not a eulogy:

• "How would they react if you told them you were moving to another country?"
• "What would they say about the current state of the world?"
• "What phrase did they always use?"
• "What joke did they tell too many times?"
• "How did they answer the phone?"
• "What made them angry? What did anger look like on them?"

Conflicting Accounts

Cross-reference accounts. Where do memories align? Where do they conflict? Conflicting memories are not problems — they are gifts. They prove the person was complex enough to be perceived differently by different people. A mother who was "always patient" to one child and "terrifying when angry" to another was both. Preserve both. Build the personality from triangulated data. No single source owns the truth.

5. Source Materials

Not all source materials are equally valuable for psychographic purposes. The following hierarchy reflects psychographic quality — the degree to which the material reveals authentic voice, not just biographical fact:

6. The Composite Portrait

For Memorial projects — and to some extent for any project that combines session material with archive material — the psychographer must construct a composite portrait: a unified picture of the person built from multiple sources that may contradict each other.

Triangulation

Treat each account as a partial map. Each perspective reveals different terrain. The psychographer's job is to lay the maps over each other and find what is consistent (the stable personality) and what varies by context (the situational self). Both are true. Both are part of the portrait.

Weighting Sources

Not all sources are equal in authority:

• The subject themselves (in Legacy sessions, or in their own written material) holds highest authority over their own experience
• People who knew them daily — spouse, close friends — tend to have access to the everyday self
• Professional contacts tend to have access to the formal, role-based self
• Distant acquaintances or public materials capture reputation and presentation, not interior life

The goal is not to resolve disagreements between accounts. The goal is to preserve them as part of the portrait's complexity — and to note, in the Personality Report, where the multiplicity exists and why.

7. Reflection