The Self-Understanding Compendium

The short version

Read only this if you read nothing else.

Three deep-research runs answered the same 7,400-word brief: build the master inventory of every question, instrument, framework, practice, tradition and technique for understanding a person completely, for a creative adult with recurring depression and anxiety who wants to break repeating cycles and find a personal myth. ChatGPT returned 463 indexed items across 23 domains with a working evidence grade on every row. Claude returned about 590 named items across the same 23 domains in dense prose, with the sharpest profile-specific rules of the three. Gemini returned a much shorter essay, roughly forty items in nine sections, but with sharper mechanics on a few instruments and one useful net-new framework. The Claude report was recovered after the first build from an artifact export on disk and is folded in throughout; where a section still says "both models" it refers to a point only Gemini and ChatGPT made.

The conclusions that matter, after checking the numbers and links myself:

• The inventory exists and is usable. The public-domain backbone is strong: IPIP-NEO-120, HEXACO, ECR-R, PHQ-9, GAD-7, PCL-5, ITQ, ASRS, PID-5, the Self-Compassion Scale and the Life Story Interview are all free, validated, and linked here. You do not need a single proprietary test to build a serious baseline.
• The order matters more than the count. All three models independently converged on the same sequence: rule out structural causes first (sleep, bipolar spectrum, ADHD or autism, substances, medical), then measure, then narrate, then do depth and myth work only once history anchors it. Depth work first is how the project becomes a prettier prison.
• Insight is not the unit of progress. The single most repeated warning across all three reports is that a mind like this can turn self-inquiry into avoidance. Every item in this document exists to produce one testable hypothesis or one behavior; anything that only produces a better explanation is filed as reference. Claude's version is the sharpest: no insight counts as done until it has one small real-world test within 48 hours.
• Evidence grades were wrong in places and are corrected here. Gemini graded personality inventories and a memory-reconsolidation therapy as "strong RCT"; neither is. IFS, Polyvagal mapping, Coherence Therapy and the RAADS-R are useful languages with thin or contested evidence, and the tables now say so.
• Privacy changed under you. If you feed a megadocument to an AI, know that Claude consumer chats are now used for training by default with five-year retention unless you opt out, Gemini is off by default, and ChatGPT has a toggle. Check the switches before uploading anything about trauma, health, sexuality or third parties.

The full inventory is preserved below, table by table, so nothing is silently dropped. The curated picks at the top of each section are the editorial layer: what is worth your time, and why.

Do these this week

Eight concrete moves, in the order all three models and the evidence agree on. Total cost: about six hours and zero dollars.

1. Take the four free baseline screens and write the scores down with the date. PHQ-9, GAD-7, the Mood Disorder Questionnaire and the 6-item ASRS. These take fifteen minutes combined and they are the difference between a psychological story and a medical one. A positive MDQ or ASRS changes which therapy and which medication conversations make sense, and both models flagged "creative highs then crashes" as the specific pattern that gets misread as personality. 15 min · free · links in the assessment table
2. Book a primary-care visit and bring a one-page symptom timeline. Sleep, weight, pain, substances, medications, family history, and when the low periods started and ended. Ask which tests are indicated for you rather than requesting a "depression panel" (the models disagreed on this and NICE sides with symptom-guided testing; see the disagreements section). Screen yourself for sleep apnea with STOP-BANG first; chronic hypoxia mimics treatment-resistant depression. 30 min prep · visit cost varies
3. Do the IPIP-NEO-120 and the ECR-R and record the five highest and lowest facets plus the two attachment dimensions. These are the two public-domain instruments all three models rated highly and that I confirmed are free and live. Write one confirming and one contradicting example next to each extreme facet so the numbers stay tethered to behavior. 45 min · free
4. Run the McAdams Life Story Interview on yourself in writing, using the full 2007 protocol. Chapters, high point, low point, turning point, earliest memory, and the future chapter. This is the spine the rest of the megadocument hangs from, and the only narrative method all three models placed in their top tier. The public PDF is linked; the magazine link ChatGPT gave is dead. 2 hours · free
5. Write one loop diagram for the cycle that costs you the most. Trigger, interpretation, body state, urge, action, immediate relief, delayed cost, and the belief the cycle confirms. One page. This is the item the entire rest of the inventory exists to feed, and it is the format a therapist can actually use. 30 min · free
6. Set the analysis stop-rule before you open another framework. Write down the decision the research is supposed to inform, what evidence would change your action, and a date. ChatGPT's version: when ten more minutes produce no new information, no new action and no lower confusion, stop and move your body, act, or contact someone. This single rule is the strongest protection in the document against the project eating the life it is meant to serve. 10 min
7. Check the AI privacy switches before the megadocument goes anywhere. Claude: Settings → Privacy → turn off "Help improve Claude" (default is on since August 2025). ChatGPT: Data Controls → "Improve the model for everyone" off; Temporary Chats are still kept up to 30 days. Gemini: Gemini Apps Activity is off by default but confirm. 5 min
8. Adopt the 48-hour rule and a stop date. Claude's two profile-specific rules, and the best ones in the run: no insight is finished until it has produced one small real-world test within 48 hours, and the megadocument gets a review date (Claude suggests six months) after which it is declared good enough and the work shifts to living, being seen, making things and contributing. Write both on the first page of the document. If you notice you feel worse rather than clearer, or you are re-reading instead of acting, that is the signal to close the map. 5 min · free

How this was made

What ran, what failed, what was checked, and what the raw material is actually worth.

The prompt. One 52,191-character brief, byte-identical across platforms, pasted into Gemini Deep Research, ChatGPT Deep Research and Claude Research on September 8, 2026 at about 1:25 pm. No platform asked a clarifying question. Gemini presented a research plan that was accepted unchanged. Claude volunteered that it would not reproduce copyrighted item sets and would supply public equivalents, which is consistent with the brief. The exact text is collapsed at the end of this section.

ChatGPT ran about 30 minutes across roughly 400 searches and produced the document this synthesis is built on: 23 domain tables, an inventory count, a stated limits section, a minimum-effective-path sequence, and a triage table matching patterns to modalities. Its weakness is citations. The report's inline references were rendered as internal tokens that do not survive extraction, and of its 177 links, 113 are Google Scholar search queries and 30 are Open Library searches. Only 34 point at an actual source page. Those search-style links are kept and marked with ⌕ so you know what you are clicking; the claims that matter were verified separately.

Gemini finished in about 15 minutes and returned a well-written essay, not an inventory. It ignored the 400-item floor, produced no per-domain count, and covered nine of the brief's 23 domains. Where it went deep it was accurate: its ECR-R reverse-scoring items, ITQ threshold rule, AAI adjective protocol and exercise-dose figures all checked out. It also contributed the one genuinely new framework in the run, the Mythopoetic Clinical Formulation. Its evidence grading is unreliable and is overridden here.

Claude ran on Opus at high effort and was not reachable at extraction time: the conversation had been created under a different Claude login than the one signed into Chrome, and the API returned a genuine not-found error. No other chat was substituted. After the first build, Isaac pasted the report and pointed at its artifact export on disk (research result/compass_artifact_wf-eb4a4ba7…, 53,387 bytes); the two are byte-identical, the export carries the compendium-only fingerprint ("Symbolic Mirror Systems", 23 numbered domains) and none of the tailored prompt's markers, and its sibling file in that folder is the other run. It is the most profile-aware of the three reports: dense prose rather than tables, a self-reported count of about 590 items with a per-domain breakdown, a plain-language grade on most entries, and the only explicit rules for keeping the project from becoming the loop. Its weakness is sourcing: the export carries no links, so every number it asserts was checked independently below and its web addresses were fetched by hand.

Verification. Thirty-one specific numbers (cutoffs, effect sizes, remission rates, trial sizes, regulatory dates, sample sizes) were checked against primary sources; twenty-six confirmed, three are partly right (the Pennebaker protocol is one of several standard descriptions; the IPIP-NEO's convergent validity with the NEO-PI-R is nearer .66 uncorrected and .91 corrected than the .84–.89 Claude gave; the Self-Authoring evidence is one randomized trial plus a non-randomized cohort), one is misattributed (the 750 and 1,130 METs-min/week exercise figures are from Xu 2025; Tian 2024, which Claude cited, puts the peak at 860), and one is outdated (IFS on a federal registry that closed in 2018). Twelve policy and availability claims were checked; the Anthropic training-default claim needed correcting and three "free instrument" claims needed qualifying. Forty links were fetched; six were dead and three had moved, and all have been replaced with working equivalents marked link replaced. Evidence grades on four rows were changed by me and are marked editor re-graded with the reason on hover.

## MISSION

Conduct exhaustive, internet-wide deep research to compile the most comprehensive master inventory ever assembled of **every question, questionnaire, interview protocol, instrument, assessment, framework, exercise, journaling method, practice, ritual, model, taxonomy, archetype system, therapy method, philosophy, spiritual tradition, and self-help technique** that humanity has produced for understanding a single human being deeply and accurately, and then helping that person self-actualize.

The output is a **master list and index**, not the answers. One person will take this inventory and answer every question, take every assessment, and do every exercise to produce a "megadocument of self-understanding": a document so complete that the person, a therapist, a mentor, a friend, or an AI reading it could understand who this person is, why they are the way they are, why they suffer, why they repeat what they repeat, and precisely how to help them heal, grow, and fully realize their potential.

Scour the whole of recorded human knowledge: clinical psychology, psychiatry, every psychotherapy school major and minor, personality science, positive psychology, developmental psychology, neuroscience, philosophy East and West, every major religious and mystical tradition, indigenous and shamanic traditions, mythology and depth psychology, anthropology, literature and biography, coaching, the self-help industry, 12-step programs, monastic rules, military and athletic psychology, the quantified-self community, and the tools people trade in forums like Reddit. Breadth AND depth. If a framework exists anywhere in human knowledge that helps a person know themselves and move forward, it belongs on this list. The long tail matters most: the obvious things are the ones already known.

**The single most important rule: for every item, give the actual questions or the actual procedure, not just the name.** A line saying "take the Young Schema Questionnaire" is nearly useless. A line that lists the 18 schemas, says where the free validated form is, how it is scored, what a high score means, and what to do about it is what is needed. When an instrument is copyrighted, give the free or public-domain equivalent, the item structure, and a link to where it can be taken. When a practice is a procedure (a journaling method, a meditation, a ritual), give the steps.

## THE SEEKER: WHO THIS SERVES AND WHAT THEY ACTUALLY NEED

Tailor everything to this profile. The person commissioning this research is one adult doing this for themselves: a creative person (music and software) who:

- Lives with **recurring depression and anxiety** and wants to understand their root causes, psychological, historical, relational, and possibly physiological, not just manage symptoms
- Is caught in **repeating cycles**: the same thought patterns, moods, limiting beliefs, emotional loops, and life situations recurring in different costumes, and wants to finally see the mechanism and break it
- Carries **trauma and formative wounds** that have not been fully mapped or metabolized
- Is highly drawn to **depth**: archetypes, personal myth, spiritual wisdom, philosophy, the big picture; a meaning-oriented mind, high in openness, intelligent, creative, sensitive, hungry for the whole truth rather than fragments
- Has read or sampled self-help before and been burned by shallow material; wants **what actually works**, with the fluff labeled rather than hidden
- Wants to discover and articulate a **personal myth**, a deep narrative of who they are and what their life is for, and critically wants to know how to *keep hold of it* when depression and doubt return
- Wants to fully realize their **creative and worldly potential**: creativity, meaningful success, happiness, contribution, the sense of having used their one life fully
- Intends to compile the answers into a living megadocument they will read themselves, hand to a helper (therapist, mentor, friend), and feed to an AI as a standing context for coaching, therapy-style reflection, and decision support
- Will use whatever tools work, from clinical psychology to ancient religion to unconventional practices, as long as the evidence and the risks are stated

**Read between the lines of this profile and address the deeper needs it implies:**

- A person who seeks *total* understanding often does so because partial understanding has felt unsafe. Address the relationship between the need for complete maps and the anxiety, perfectionism, or control patterns that can drive it.
- "Understand everything first, then act" can itself become the loop. Include the psychology of **insight versus action**, analysis paralysis, intellectualization as a defense, self-help addiction, "preparing to live" versus living, and how real change actually sequences (insight and action feeding each other, not insight completing before action begins).
- Include the research on **when introspection helps and when it harms** (Eurich, Timothy Wilson, Nisbett and Wilson, the introspection illusion, why "why" questions can deepen rumination while "what" questions help), so this whole project does not become a rumination machine.
- Repeated cycles plus depression and anxiety plus high-openness intensity sometimes have contributing causes the person has never checked. Include **neurodivergence and physiology** as lenses across every domain so nothing structural is missed.
- Someone building a mirror this powerful is often preparing not only to heal but to **help others**. Include the wounded-healer path and how self-knowledge converts into contribution and vocation.
- Beneath "understand me completely" is usually the wish to be **fully seen and fully accepted**. Include the frameworks (self-compassion, shame work, unconditional positive regard, spiritual acceptance) that address that wish directly, so the megadocument heals rather than becoming another standard the person fails to meet.

Every item must be geared toward **insight and understanding in service of self-actualization**, not trivia. For each tool the implicit question is: "How does this help this person see themselves clearly, accept what they see, and move forward?"

## RESEARCH DOMAINS TO COVER EXHAUSTIVELY

For each domain, cover the well-known and battle-tested, the current and trending, and the lesser-known or niche. The names below are **seeds to show the depth wanted, not the boundary**. Go far beyond them, and add domains that are missing.

### 1. Psychological Assessment, Personality Science, and Psych-Profile Instruments
- Validated personality inventories: Big Five / NEO-PI-R and its free IPIP versions and facets, HEXACO, 16PF, MMPI-style clinical inventories, Hogan, DISC, temperament models from Keirsey to Cloninger's TCI, Fisher's temperament types, Kagan, Dabrowski overexcitabilities.
- Type systems used for reflection even where validity is weak, labeled honestly: MBTI and its Jungian cognitive-function origins, Enneagram with wings, instincts, levels of health, and integration and disintegration lines, Socionics.
- Strengths instruments: VIA Character Strengths, CliftonStrengths, Reflected Best Self.
- Attachment measures: ECR-R, Adult Attachment Interview questions, Levine and Heller's attachment quizzes.
- Schema and belief measures: Young Schema Questionnaire, Schema Mode Inventory, Dysfunctional Attitude Scale, core-belief and limiting-belief inventories.
- Clinical screeners to take as a baseline and keep re-taking: PHQ-9, GAD-7, ACE and its expanded versions, PCL-5, ITQ for complex PTSD, MDQ and HCL-32 for bipolar spectrum, ASRS for ADHD, RAADS-R and AQ for autism, PID-5 for personality-disorder traits, TAS-20 for alexithymia, the Highly Sensitive Person scale, perfectionism scales, rumination and worry scales, intolerance of uncertainty, self-compassion scale, shame and guilt scales, locus of control, explanatory style, hope scale, grit, self-efficacy, meaning in life questionnaire, satisfaction with life, flourishing scale, PANAS, Zimbardo time-perspective inventory, loneliness scale, social anxiety scales, procrastination scales, dark triad, narcissism and vulnerable narcissism, emotional intelligence, interpersonal circumplex, defense-mechanism inventories, coping-style inventories such as the Brief COPE, burnout inventories, chronotype, sensation seeking, need for cognition.
- Cognitive style assessments: thinking patterns, learning styles, decision-making styles, attributional style (Seligman).
- Developmental-stage instruments: Kegan's subject-object interview, Loevinger and Cook-Greuter ego development, Erikson stages, Levinson seasons, Spiral Dynamics.
- Projective and semi-projective methods a person can do on themselves, with caveats: sentence completion, early recollections, thematic-apperception-style prompts, drawing tests, one's own creative output analyzed as projective material.
- For each: what it measures, the questions or a link to a free version, scoring, interpretation, validity and reliability in plain terms, limitations, where to take it free or officially, and how the result feeds the final document.

### 2. Depth Psychology, Archetypes, and the Personal Myth
- Jungian psychology in full: the complete catalog of archetypes (Hero, Shadow, Anima/Animus, Self, Persona, Trickster, Orphan, Victim, Martyr, Caregiver, Sage, Magician, Ruler, Lover, Creator, Explorer, Seeker, Rebel, Innocent, Everyman, Outlaw, Wounded Healer, Puer/Puella and Senex, Great Mother, Wise Old Man, and more), complexes, individuation, the word-association experiment, active imagination step by step, dream analysis methods and dream journaling, working with recurring images and fascinations, the symbols and stories a person returns to.
- Every archetype system and, for each archetype, its gift, its pitfall and shadow side, how it shows up in a real life, how it becomes a trap, and the path of integration: Carol Pearson's twelve archetypes and the PMAI, Moore and Gillette's King, Warrior, Magician, Lover with the active and passive shadow poles of each, Bly's Iron John and the mythopoetic men's movement, Bolen's gods and goddesses in everyone, maiden/mother/queen/crone frameworks, Perera's descent of Inanna, Marie-Louise von Franz on the puer aeternus (essential for a creative who keeps not landing), the senex, Clarissa Pinkola Estés, Robert A. Johnson's Inner Work and Owning Your Own Shadow, James Hollis (The Middle Passage, The Eden Project, Living an Examined Life and his question lists), Bill Plotkin's Soulcraft and Wild Mind map of the psyche and stages of soul initiation, Michael Meade, Martin Shaw, Sharon Blackie, Francis Weller, Thomas Moore's Care of the Soul, Richard Rohr's two halves of life, Jean Houston, the Enneagram's archetypal reading.
- Joseph Campbell's Hero's Journey and monomyth as a self-understanding map, its critiques, where a person gets stuck (refusal of the call, belly of the whale, the return) and what each stuck point demands; Vogler and Booker's plot shapes as templates for reading a life.
- James Hillman's acorn theory and The Soul's Code: daimon, calling, reading one's childhood for the seed of vocation.
- Internal Family Systems: full parts-mapping protocols, exiles, managers, firefighters, Self-leadership, the 8 Cs, unburdening, complete question sets for parts work. Psychosynthesis subpersonalities and disidentification.
- Inner child work: methods, dialogues, reparenting protocols (Pete Walker, schema-therapy reparenting, Ideal Parent Figure protocol).
- Shadow work in full: projection inventories (what one hates, envies, admires, and cannot stand in others as a map of disowned material), the golden shadow, the 3-2-1 process.
- The **personal myth** in depth: Jung ("what is the myth you are living"), Jung's own confrontation with the unconscious as a case study, Campbell, Dan McAdams' narrative identity research, Stephen Larsen, Feinstein and Krippner's Personal Mythology five-stage program. Every method for discovering it (childhood obsessions, wounds, fascinations, envy, what makes one cry, what one would do for free, the recurring dream, the story one cannot stop retelling, the book one keeps re-reading), every method for testing it against reality, and every method for **not losing sight of it during depression, doubt, and dark periods** (personal creed and manifesto, mission and vision statements, symbols and talismans, myth journals, retelling rituals, daily, weekly, and quarterly review cadences, commitment devices, mentors, communities, rites of passage, vows). And the known dangers: inflation, grandiosity, using the myth to escape ordinary life, the puer trap, spiritual bypassing.

### 3. Therapy Methods That Actually Work, and the Self-Examination Tools Each One Uses
For every major and minor school, extract its core diagnostic questions, its map of the psyche, its exercises and worksheets, and its theory of why people repeat cycles:
- CBT (complete cognitive distortion lists, thought records, downward arrow to core beliefs, behavioral experiments), Schema Therapy (all 18 early maladaptive schemas with origins, triggers, surrender, avoidance, and overcompensation coping styles, schema modes, healing paths for each), ACT (values clarification, defusion, the matrix, willingness), DBT (skills and diary cards), Compassion-Focused Therapy, Metacognitive Therapy (Wells), Rumination-Focused CBT, MBCT, Behavioral Activation, Unified Protocol, CBT-I, exposure methods, Interpersonal Therapy.
- Trauma frameworks in full: EMDR and its target-sequence questions, Somatic Experiencing, Sensorimotor Psychotherapy, polyvagal theory (Porges, Deb Dana's mapping exercises) and the window of tolerance, Bessel van der Kolk, Peter Levine, Gabor Maté (trauma as disconnection, The Myth of Normal, Compassionate Inquiry), Janina Fisher, Pete Walker's complex-PTSD model (fight, flight, freeze, fawn types, emotional flashbacks, inner critic shrinking), Coherence Therapy and memory reconsolidation (Ecker), Lifespan Integration, attachment repair and earned secure attachment.
- Relational modalities: Emotionally Focused Therapy, Imago, Gottman.
- Psychodynamic and psychoanalytic methods (defenses, transference, the core conflictual relationship theme, repetition compulsion, free association, Malan's triangles), object relations and self psychology (Winnicott's true and false self, Kohut, Fairbairn), Jungian analysis, Adlerian therapy (early recollections, lifestyle assessment, birth order, the Courage to Be Disliked questions), Transactional Analysis (ego states, life scripts, the script questionnaire, drivers, injunctions, games, the Karpman drama triangle and how to exit it), Gestalt (empty chair, unfinished business), existential therapy (Yalom's four givens: death, freedom, isolation, meaninglessness; Frankl's logotherapy and Socratic dialogue; May; Bugental), narrative therapy (externalizing, unique outcomes, re-authoring, the definitional ceremony), solution-focused therapy (the miracle question, scaling, exceptions), motivational interviewing (decisional balance, the change ruler), Bowen family systems (differentiation, genogram, triangles), structural and strategic family therapy, family constellations (labeled for evidence), Hakomi, Morita therapy and Naikan (the three Naikan questions), Ericksonian hypnotherapy, hypnosis, NLP (labeled for evidence), art, music, and drama therapy exercises, group-therapy and process-group methods.
- Coaching models: GROW, co-active, ontological coaching, Hoffman Process, Landmark (with cult-risk caveats), Byron Katie's four questions, the Sedona Method, Gendlin's Focusing (full six-step protocol).
- Recovery inventories: 12-step step-4 moral inventory worksheets (resentments, fears, harms, sex conduct: one of the most systematic self-inventories ever built), Recovery Dharma, SMART Recovery.
- Emerging clinical directions at the level of science and legitimate clinical pathways: psychedelic-assisted therapy protocols with preparation and integration questions, ketamine, TMS, medication decision frameworks and the questions to bring to a psychiatrist; what the evidence says and how people access such treatment legally with supervision.
- The clinical case-formulation formats (the 5 Ps: presenting, predisposing, precipitating, perpetuating, protective; CBT and psychodynamic formulations; the "vicious flower"), because the document should end in a formulation a clinician would recognize.
- A **matching guide**: which modalities work best for which pattern (rumination vs flashbacks vs numbness vs abandonment schemas vs shame vs grief), so the seeker can triage their own work.

### 4. Life History, Autobiography, Trauma, and Narrative Methods
- Structured life-review methods: McAdams' Life Story Interview (full protocol: chapters, key scenes, high point, low point, turning point, earliest memory, wisdom event, future script; redemption vs contamination sequences and how to re-author them), the lifeline and life-graph exercise, chapters-of-life method, Ira Progoff's Intensive Journal (steppingstones, dialogues, period log), Self-Authoring past, present, and future modules, peak experience inventories, nadir analysis, turning-point analysis.
- Genograms with emotional and pattern annotations, Bowen family systems, the family map across three generations (what each grandparent and parent wanted, feared, repeated, and passed down; family roles of hero, scapegoat, lost child, mascot; family secrets, myths, scripts, and loyalties; immigrant, class, religious, and cultural narratives; ancestral and inherited trauma), family constellation concepts with evidence caveats.
- Questions for excavating formative memories: Adlerian early-recollection analysis, best and worst moments audit (peak experiences after Maslow, flow moments, most alive, most ashamed, most proud, most loved, most abandoned), proudest achievements, deepest regrets and the regret-minimization frame, unfinished business, unlived lives, "what did I love and obsess over as a child before anyone told me what to want," turning points and forks not taken, mentors and anti-mentors.
- Every relationship as a data point (what pattern each one repeated, how each ended, what one told oneself), the story of the body and health, money history and money scripts, work and creative history, faith history including religious upbringing and any deconversion or reconstruction, sexuality and intimacy history, substance and habit history.
- The questions to ask parents, siblings, oldest friends, and ex-partners about one's childhood and oneself.
- Developmental and attachment trauma inventories, moral injury, grief inventories (Worden's tasks, Francis Weller's five gates, disenfranchised grief), post-traumatic growth inventories (Tedeschi and Calhoun).
- Methods for identifying recurring life themes and "the same story happening again": repetition compulsion, TA life scripts, core wound → protective pattern → repeated consequence chains.
- Guidance on sequencing trauma-related questions safely alone (titration, pendulation, when to stop and get a professional, the signs of destabilizing) and how to tell healing from re-traumatizing rumination.

### 5. Cognitive Patterns, Limiting Beliefs, Recurring Cycles, and Thought Loops
- Every named taxonomy of the patterns that keep people stuck: cognitive distortions (Burns' list and the extended lists), the full catalog of cognitive biases that matter for self-knowledge, Young's 18 schemas and coping styles, limiting-belief catalogs, core beliefs about self, others, world, and future, the inner-critic typology (Stone and Stone, Jay Earley's seven critics, Firestone's critical inner voice), the critic's origin and protective function, rumination and worry subtypes, metacognitive beliefs about thinking, explanatory style, attributional biases, Vaillant's hierarchy of defense mechanisms.
- Limiting belief identification and dismantling: Byron Katie's four questions and turnarounds, the Lefkoe method, CBT restructuring, belief archaeology (tracing each belief to its origin scene), memory reconsolidation as the mechanism behind deep belief change.
- Loop science: why cycles self-perpetuate, the anxiety and depression loops as diagrams, the shame spiral, avoidance cycles, Wells' detached mindfulness and attentional training, how to break loops behaviorally when thinking harder fails.
- Self-sabotage patterns: procrastination psychology (emotion-regulation model) and procrastination types, perfectionism types, fear of success and fear of failure, upper-limit problems (Hendricks), imposter syndrome, boom-and-bust and overwork-collapse cycles, addictive and compulsive loops, learned helplessness, learned hopelessness, the pattern of choosing the same kind of partner, friend, boss, or collaborator, Imago's theory of why we pick partners who re-create the wound, intellectualization and insight-seeking as avoidance.
- The hypomania-crash pattern and how to distinguish a personality cycle from a bipolar-spectrum cycle; ADHD and autism as hidden causes of "why do I keep doing this"; circadian and seasonal cycles; the nervous-system-state model (ventral, sympathetic, dorsal); emotional flashbacks; trigger inventories; "if I ever feel X it means Y" chains; the 4F trauma responses as personality; Dabrowski's positive disintegration as a frame for a sensitive creative person's repeated breakdowns and rebuildings.
- Practical methods for detecting recurring themes from one's own data: re-reading old journals, messages, lyrics, and creative work, playlists, bookmarks, and browser history as a corpus, mood tracking, and the tools and prompts for doing that analysis, including with an AI.

### 6. Neurodivergence, the Brain, and Structural Explanations
*(Often the missing piece behind "why do I keep cycling despite understanding myself")*
- ADHD in adults: how it masquerades as anxiety, depression, laziness, and broken willpower; screening tools; how it changes the entire self-actualization playbook.
- Autism spectrum in adults, giftedness and twice-exceptionality, the highly sensitive person and sensory processing sensitivity, rejection sensitive dysphoria, alexithymia, Dabrowski overexcitabilities.
- How to pursue proper evaluation and how self-understanding documents should account for neurotype.
- The neuroscience of habit, emotion, and change at a practical level: neuroplasticity realities vs myths, the default mode network and rumination, memory reconsolidation (activation plus mismatch, not explanation), the evidence on what actually rewires.

### 7. Body, Physiology, and the Biological Ground of Mood
- The full medical rule-out list a person with depression or anxiety should complete: thyroid, vitamin D, B12, iron, hormones, sleep disorders including apnea, chronic inflammation, gut-brain axis, blood sugar, PMDD where relevant, substance effects, medication side effects, ADHD, bipolar II.
- Evidence-based lifestyle psychiatry with real numbers: exercise as antidepressant (dose and effect sizes), sleep architecture, circadian rhythm and chronotype, sleep restriction, light exposure, nutrition, alcohol and cannabis effects, nature exposure, social contact, cold exposure, Ilardi's TLC protocol.
- Somatic self-knowledge: body mapping of emotions, interoception training, breathwork lineages with evidence, tension and trauma release, yoga and mood research, the felt sense, Ayurvedic doshas and Chinese five elements as constitution maps (labeled).
- Energy management vs time management; personal rhythm mapping; the somatic self-examination questions, since mind-only inventories miss half the picture.

### 8. Spiritual Traditions, Wisdom Literature, Contemplative Practice, and the Spiritual Profile
For every tradition, extract its map of the human being, its diagnosis of what goes wrong, its self-examination questions and practices, and its guidance for a person stuck in repetition, despair, and anxiety:
- Buddhism (the Four Noble Truths as diagnostic frame, the five hindrances as a self-observation map, the three poisons, dependent origination as a model of the cycle, the brahmaviharas, Vipassana and noting practice, non-self inquiry, Zen and koan practice, Tibetan lojong slogans and tonglen, mindfulness of death, Abhidharma, Tara Brach's RAIN, Pema Chödrön, Thich Nhat Hanh).
- Stoicism (dichotomy of control, the evening review of Seneca and the Pythagoreans, premeditatio malorum, the view from above, Epictetus' discipline of assent, Marcus' Meditations as a self-therapy journal, modern Stoic practice programs).
- Christianity in its contemplative and psychological forms (the Ignatian Examen and the Spiritual Exercises, discernment of spirits and the rules for consolation and desolation as an emotional compass, lectio divina, the desert fathers and Evagrius' eight thoughts as the first cognitive-behavioral taxonomy, the seven deadly sins and their virtues as a personality map, John of the Cross' dark night, Merton's true and false self, centering prayer, Kierkegaard on despair and the self, Augustine's Confessions as the first psychological autobiography, Rohr, Nouwen's Wounded Healer, Fowler's stages of faith).
- Judaism (Mussar and the middot with cheshbon ha-nefesh nightly accounting, the Kabbalistic tree of life as a psychological map, Hasidic teachings on the yetzer, Heschel).
- Islam and Sufism (stages of the nafs, muhasaba self-accounting, muraqaba, the heart's polishing, Rumi, Ibn Arabi, al-Ghazali's Alchemy of Happiness).
- Hinduism and Yoga (the Yoga Sutras' kleshas as the five root afflictions, the gunas as temperament, the koshas, the Bhagavad Gita on dharma and svadharma and acting without attachment to outcomes, the four purusharthas, the four ashramas, Vedantic self-inquiry and Ramana's "Who am I?").
- Taoism (wu wei, the watercourse way, Tao Te Ching, Zhuangzi, the I Ching as a reflective oracle), Confucian self-cultivation.
- Greek philosophy as practice (know thyself, the daimon, eudaimonia, Aristotle's virtues and the golden mean as a strengths audit, Socratic questioning, Epicurus, the Skeptics' suspension of judgment).
- Japanese practices (Naikan, Morita, ikigai in its original sense, kintsugi, wabi-sabi, shinrin-yoku).
- Indigenous and shamanic frameworks (vision quest, rites of passage, medicine wheel, soul retrieval, council practice, four-directions models, Plotkin's eco-soulcentric development), African and diasporic frameworks (Ubuntu, Yoruba Ori and destiny), Hawaiian ho'oponopono.
- Existentialism (Nietzsche's amor fati, "become who you are," and the eternal-return test, Kierkegaard's stages, Camus on the absurd, Sartre's bad faith, Heidegger's authenticity and being-toward-death, Becker's Denial of Death), transcendentalism (Emerson's Self-Reliance, Thoreau).
- Gurdjieff and the Fourth Way (self-observation, self-remembering), Ken Wilber's integral map (lines and levels, the pre/trans fallacy, Integral Life Practice), transpersonal psychology, Grof's holotropic work, nondual teachers, A Course in Miracles, the 12-step tradition as a spiritual program.
- Death contemplation across traditions (memento mori, the Bhutanese practice, Stephen Jenkinson, Bronnie Ware's five regrets, the eulogy and obituary exercise), grief rituals.
- Intensive formats: fasting, silence, solitude, pilgrimage, monastic stays and retreats (Vipassana 10-day, Jesuit retreats, Zen sesshin, dark retreats), psychedelic and plant-medicine traditions with preparation, ceremony, and integration frameworks, breathwork lineages.
- Meditation and contemplative practices organized by purpose and evidence: insight, calming anxiety, self-compassion (Neff's protocols), deconstructing negative self-concept, loving-kindness for the inner critic.
- Frameworks for mapping a complete **spiritual profile**: beliefs, longings, existential fears, experiences of the sacred, relationship to death and mystery.
- Spiritual bypassing (Welwood): the complete warning-sign list. Dark night of the soul literature and how to tell a dark night from clinical depression needing treatment. The research on what distinguishes genuine transformation from spiritual self-deception.
- For each practice: the steps, the time commitment, the evidence where any exists, the known risks (including psychosis and destabilization risks of intensive retreat and psychedelics for people with mood disorders), and who it fits.

### 9. Philosophy as Self-Understanding, Personal Worldview, and the Models of the Realized Person
- Philosophical schools with direct application: Stoicism, Existentialism, Aristotle's eudaimonia and virtue ethics as a life-audit, phenomenology of lived experience, Pierre Hadot's philosophy as a way of life and the spiritual exercises of the ancient schools, modern practical philosophy.
- The questions that reveal a person's actual (not stated) philosophy: what one believes about free will, meaning, death, God, suffering, morality, luck, success, love, work, money, other people, and human nature, and the inventories that surface these (worldview questionnaires, the PhilPapers survey adapted for a layperson, moral foundations questionnaire).
- The great philosophical questions turned inward as journal prompts: What is a good life *for me*? What do I owe others? What is my relationship to death, freedom, isolation, and meaning?
- Every serious model of the fully realized person and the questions each model asks: Maslow's self-actualization characteristics, peak experiences, and his late addition of self-transcendence (and what that reordering means for a seeker), Kaufman's Transcend, Rogers' fully functioning person, Csikszentmihalyi's autotelic personality, Frankl's will to meaning, Fromm's productive character and To Have or To Be, Erikson's generativity and integrity, Jung's individuation, Nietzsche's self-overcoming, Kegan's orders, Dabrowski's levels, Wilber's stages, Seligman's PERMA, Ryff's six dimensions, Brooks' résumé vs eulogy virtues and The Second Mountain, Burkeman's Four Thousand Weeks on finitude.

### 10. Desires, Values, Motivation, and Life Direction
- Values clarification instruments (ACT values card sorts, Schwartz's universal values, Rokeach, the Barrett model) and the difference between stated and revealed values (what the calendar and the money actually show).
- Motivation science: Self-Determination Theory audit (autonomy, competence, relatedness), intrinsic vs extrinsic mapping, McClelland's needs, Reiss's 16 basic desires, Max-Neef's fundamental human needs.
- Desire archaeology: authentic vs introjected vs mimetic desire (Girard turned into self-inquiry: whose desires have I been copying?), Burgis's thick vs thin desires.
- Ikigai and cultural purpose frameworks, vocation and calling literature, the difference between goals and direction, the Hedgehog concept.
- Goal and direction frameworks that connect to identity: WOOP and mental contrasting, Odyssey plans, workview and lifeview (Designing Your Life), future-self research (Hershfield), regret minimization, Fear-Setting, the Dickens process, the ideal ordinary day, the 100 dreams list, the wheel of life, Buffett's 5/25, the energy audit, Debbie Millman's ten-year plan, the "what would I do if I could not fail / had ten years / had one year" family of questions, the "who do I admire and why" list, the letter from the 80-year-old self, the letter from the future self, the Best Possible Self exercise, David Whyte's questions, Krista Tippett's and Tim Ferriss's interview questions, Kevin Kelly's questions.
- Complete question banks for mapping what one loves, hates, longs for, fears, and wants across every life domain: work, love, creativity, body, spirit, community, adventure, legacy.

### 11. Relationships, Love, Belonging, and How Others See You
- Adult attachment in romantic relationships: styles, origins, dance patterns (the anxious-avoidant trap), the path to earned security; Gottman's research and questions; love styles (Lee, Sternberg's triangle, the five love languages with evidence caveats); Esther Perel's questions.
- Who one loves and why: repetition of parental templates, Imago and the imago inventory of caretaker traits, projection in love, relationships as mirrors and as the primary arena where old wounds surface for healing.
- Family-of-origin repair in adulthood: boundaries, Bowen differentiation, adult conversations with parents, forgiveness work (Worthington's REACH), estrangement wisdom when repair is not possible.
- Friendship, community, and belonging: loneliness research, how adults build deep friendship, community as a self-actualization requirement (Harvard Study of Adult Development), community as treatment for depression, frameworks for deciding whom to keep, whom to cut, and whom to seek (mentors, peers, communities, scenes), the "five people I spend the most time with" audit.
- Shame and vulnerability research (Brené Brown): shame vs guilt, shame resilience, how being fully seen heals.
- Every method for an accurate outside view: the Johari Window with its full adjective list and shrinking the blind-spot quadrant, 360-degree feedback adapted for personal life, the Reflected Best Self exercise (full protocol, ten to twenty people), Tasha Eurich's internal vs external self-awareness research and her loving-critic method, the interpersonal circumplex and interpersonal-problems inventories, how reputation forms, how to weight praise vs criticism, how to metabolize others' opinions (true signal vs their projections).
- The questions to send to friends, family, colleagues, collaborators, ex-partners, and critics: what do you count on me for, what do you avoid bringing to me, what do I not see about myself, when have I hurt you, when have I helped you, what would you never say to my face, what do people say about me when I am not there.
- The questions that reveal whom one loves and why, whom one has lost and why, whom one trusts, resents, envies, has wronged, and who one's chosen family is.

### 12. Shame, Grief, Forgiveness, and Emotional Completion
- Grief work in full: models beyond the five stages (continuing bonds, dual process model, Worden's tasks, Weller's five gates), disenfranchised grief, grieving unlived lives and lost time, including the grief that surfaces when a person finally sees their patterns clearly.
- Self-forgiveness protocols, making amends, releasing resentment as a practice rather than a decision.
- Emotional processing skills: emotional granularity (Brackett's Mood Meter), completing the stress cycle (Nagoski), expressive writing (Pennebaker's protocol, the best-evidenced writing intervention for trauma).

### 13. Addiction, Escapism, Compulsion, and Attention
- The full spectrum of escape patterns: substances, screens, food, work, fantasy, pornography, shopping, compulsive self-improvement, understood through Maté's lens (not why the addiction, but why the pain).
- Recovery frameworks and their self-inquiry tools (12-step inventories, SMART Recovery, Recovery Dharma, recovery capital).
- Attention as the substrate of change: digital habits, dopamine and novelty-seeking, deep work capacity, how a scattered attention field makes self-actualization structurally impossible and how to rebuild it.

### 14. Depression and Anxiety Specifically: Models, Causes, and What Works
- Every serious model: Beck's cognitive triad, Lewinsohn's behavioral model, learned helplessness, Nolen-Hoeksema's rumination, Gilbert's social-rank and evolutionary models, Hari's nine causes, inflammation and metabolic models, the sleep and circadian model, existential depression, Dabrowski, the metacognitive and intolerance-of-uncertainty models of anxiety, the panic cycle, Claire Weekes' floating, the polyvagal view.
- Bibliotherapy with trial evidence (Feeling Good, Feeling Great, Mind Over Mood, The Happiness Trap, Unwinding Anxiety, DARE, The Anxiety and Phobia Workbook).
- The treatment decision tree: when therapy, which therapy, when medication, which class and why, when ketamine, TMS, or psychedelics, and the questions to ask each provider.
- The questions for distinguishing depression from grief, burnout, boredom, existential crisis, unlived life, and a correct perception of a bad situation.

### 15. Work, Money, Creativity, and Worldly Actualization
- Career and vocation psychology: strengths-based career design, career-fit instruments (Holland RIASEC, Strong Interest Inventory, Schein's career anchors, O*NET), flow at work, job crafting, meaningful work, definitions of success across traditions and how to choose one's own.
- Money psychology: money scripts (Klontz), scarcity and abundance patterns, family money wounds, money as a mirror of self-worth beliefs.
- Creativity in full, with special attention to the creative and the builder: creative process models, creative-block inventories, The Artist's Way (morning pages, artist dates, the recovery weeks), The War of Art on Resistance, Rick Rubin's The Creative Act, Art and Fear, Rollo May's The Courage to Create, Big Magic, Twyla Tharp's The Creative Habit, Austin Kleon, creative identity, shipping vs perfecting, the identities of "artist," "entrepreneur," and "engineer" and how they conflict, the commerce vs art tension, the research on creativity and mood (Jamison's Touched with Fire, the myth of the depressed artist, what actually fuels sustainable creative work).
- Success literature that survives scrutiny: deliberate practice, grit and its criticisms, luck surface area, compounding, the long game.
- Follow-through frameworks that replicate (implementation intentions, WOOP, identity-based habits, tiny habits, commitment devices, Deep Work, Essentialism, the one-thing question) and the ones that backfire (outcome visualization, affirmations for people with low self-esteem, pure positive thinking).

### 16. Happiness, Flow, and Peak Functioning
- Positive psychology's complete toolkit: PERMA, flow conditions and personal flow-trigger mapping, gratitude and savoring with actual evidence, strengths-spotting, hope theory (Snyder), Best Possible Self, self-affirmation of values.
- Peak experience research and the full list of characteristics of self-actualized people.
- The science of durable happiness: hedonic adaptation and how to outwit it, relationships as the strongest predictor, experiences vs possessions, contribution and generativity.

### 17. Identity, Adult Development, and Stages of Growth
- Erikson's stages as a life-audit (which crises resolved, which still open), Levinson's seasons.
- Constructive-developmental theory: Kegan's stages and the subject-object shift, Loevinger and Cook-Greuter, Spiral Dynamics: where a person is and what the next transition asks.
- Kegan and Lahey's Immunity to Change: the complete mapping process for competing commitments and big assumptions, directly targeting "I understand my patterns but still repeat them."
- Identity formation and re-formation: Marcia's identity statuses, midlife transition literature (Jung's two halves of life, Hollis), quarter-life and every-life crises as developmental doorways, Dabrowski's positive disintegration.

### 18. Rites of Passage, Liminality, and Transformation
- Van Gennep and Turner: separation, liminality, incorporation, and the modern problem of permanent liminality with no elders and no incorporation rituals.
- How to construct genuine modern rites of passage; wilderness, solo, and ordeal-style practices; marking transformation so the psyche believes it.
- Post-traumatic growth research: the five domains where suffering becomes strength, and what distinguishes growth from denial.
- The wounded healer archetype (Chiron, Nouwen): how mapped wounds become the exact place from which a person helps others, and how service completes healing.
- Masculine and feminine developmental psychology and initiation work, presented as psychological maps available to anyone.

### 19. Symbolic Mirror Systems (Use with Clear Eyes)
- Systems people use as projection screens and structured mirrors: tarot's Major Arcana as an archetypal journey, the I Ching, astrology, Human Design, the Gene Keys. Catalogued honestly: **no scientific validity as measurement**, flagged clearly, but explained for how seekers use them the way one uses a Rorschach or a poem, as prompts that surface what is already inside.
- Clear criteria for the line between symbolic play that generates insight and magical thinking that outsources agency.

### 20. Resources: People, Books, Great Works, Communities, and Tools as Mirrors
- **Famous and historical people** who lived through similar patterns (recurring depression and anxiety in a creative or builder, the cyclical life, the late bloomer, the person who found their myth after a breakdown) and what specifically helped them, with sources: Lincoln, Churchill, William James, Jung's own crisis, Tolstoy's A Confession, John Stuart Mill, Kierkegaard, Nietzsche, Rilke, Van Gogh, Styron, Andrew Solomon, Kay Redfield Jamison, David Foster Wallace, Leonard Cohen, Bruce Springsteen, Brian Wilson, C. S. Lewis's A Grief Observed, Frankl, Alan Watts, Matt Haig, and many more. A long list **organized by pattern, not by fame**, with the specific pattern each models and what their path teaches.
- **Fiction and myth that map these patterns**, organized by the pattern or archetype each illuminates: Hesse's Demian, Steppenwolf, and Siddhartha; Dostoevsky; Stoner; The Razor's Edge; Zorba; Parzival and the Fisher King wound; the Odyssey; Gilgamesh; Job and Ecclesiastes; Dante's dark wood and descent before ascent; East of Eden's timshel; Franny and Zooey; The Bell Jar; Notes from Underground; Ishmael; The Alchemist; The Little Prince; Kafka; Pessoa; fairy tales with their Jungian and Bettelheim readings; poetry that has carried people through (Rilke, Rumi, Mary Oliver, Hopkins's terrible sonnets); the Bhagavad Gita's battlefield crisis.
- **Nonfiction, workbooks, and memoirs organized by problem**: the full canon across therapy schools, trauma, depression and anxiety, attachment, creativity, meaning, myth, and spirituality, including the workbooks with exercises.
- Films and art used therapeutically, music and mood, podcasts, lecture series, YouTube channels, courses, retreats, apps, communities and forums, men's and women's groups, therapist and coach directories and how to choose a good one, reputable assessment sources, and crisis resources.
- For every resource: what pattern it addresses, what it does well, honest quality notes, and a link.

### 21. Journaling, Daily Practice, Self-Inquiry Protocols, and the Master Question Bank
- Every powerful structured journaling method compared: morning pages, Pennebaker expressive writing, Progoff's Intensive Journal (full workbook structure), bullet journal, interstitial journaling, the five-minute journal, dialogue journaling and the dialogue with a symptom or a part, unsent letters, the gratitude letter, future-self letters, dream journals and dream incubation, the daily Stoic review, the Examen, the Mussar accounting, the Naikan reflection.
- A comprehensive, evidence-graded catalog of every practice a person can do daily, weekly, or once: three good things, self-compassion practices, behavioral activation, meditation types compared, mood and habit tracking methods and apps, annual and quarterly review templates (YearCompass, Ferriss's past-year review, Guillebeau's annual review, Sahil Bloom, Ali Abdaal), personal-operating-system and second-brain documents used for self-knowledge (personal user manuals, "how to work with me" docs, personal READMEs, personal boards of directors). Include self-help addiction and analysis paralysis as failure modes and guidance on how much of this work is enough.
- **The largest possible bank of piercing self-inquiry questions**, organized by domain: the questions master therapists, spiritual directors, biographers, and great coaches actually ask. Include the Proust Questionnaire, Arthur Aron's 36 questions adapted for self-knowledge, Gregory Stock's Book of Questions, the School of Life prompt sets, clean language questions, Socratic sequences, coaching's most powerful questions, and hundreds more mined from every tradition above.
- Questions specifically designed to surface: hidden assumptions, disowned parts, unprocessed grief, denied desires, secret fears, core wounds, core gifts, the thing the person already knows but has not admitted.

### 22. Using AI as Mirror, Advisor, and Guide
- Best practices for structuring the megadocument so an AI can use it well: sectioning, a distilled core-profile summary, pattern indexes, standing instructions, what to include and exclude.
- Prompt architectures for AI-assisted self-work: the AI as Socratic questioner, as devil's advocate against limiting beliefs, as pattern-spotter across journal entries and a personal corpus, as accountability partner, as archetypal dialogue partner, for coaching, therapy-style reflection, and decision support.
- The honest limits and risks: AI sycophancy and how to instruct against it, the danger of an always-agreeable mirror, AI therapy risks, how to keep a human in the loop, privacy and security for a document this sensitive, and what AI cannot replace (embodied relationship, professional treatment, real-world action).

### 23. Integration: Building the Megadocument and the Path Beyond It
- The recommended master template for the final document: sections, ordering, how the pieces cross-reference into one coherent portrait (history → patterns → parts → wounds → gifts → values → myth → direction → practices), what belongs in a one-page summary vs the full document, the format a therapist or psychiatrist wants at intake, the format a mentor or friend can use.
- A triangulation method for reconciling contradictions between instruments and between self-view and others' view.
- How to sequence the work safely: gentle entry points vs deep trauma excavation that needs support, pacing, titration, window of tolerance, when to bring in a professional and how the document makes that work faster.
- A **minimum effective path**: the 20 percent of tools that produce 80 percent of the insight, for the days when the full map is too much, so the project has an on-ramp and cannot only be done maximally or not at all. A realistic 3-, 6-, and 12-month path.
- How to keep the document alive: review rhythms, updating the personal myth, tracking pattern change over time, annual life reviews.
- **The insight trap, addressed directly**: the research on why understanding alone does not change people, when self-inquiry has become rumination or avoidance in disguise, the signs, and the correction: how to close the map and walk the territory. The megadocument is a launchpad, not a life's work; include explicit guidance on graduating from studying oneself to living as oneself.
- How completion is not the goal: self-actualization as an ongoing practice of acting from self-knowledge, being seen in real relationships, creating real work, and contributing. The document in service of the life, never the reverse.

## DO NOT FILTER

Include approaches that are unconventional, fringe, intense, gray-area, commercialized, or that you would normally leave out: psychedelic and plant-medicine work, ketamine and other clinic-based interventions, extended fasting and silence, dark retreats, ordeal-style rites of passage, Landmark and other large-group awareness trainings, Hoffman, family constellations, hypnosis and past-life regression, astrology, tarot, Human Design, energy work, 12-step and religious programs, pickup-artist and manosphere self-improvement material, biohacking, nootropics, and anything else people actually use for self-understanding. Label each honestly with its evidence, its risks, and who it harms. Silent omission is the one failure that is not acceptable; a labeled entry is always better than a missing one. If you nearly cut something, include it and say why you nearly cut it.

## ALSO COVER

- **What is obsolete, discredited, overrated, or harmful** in this space, and why, with specifics: which instruments lack validity, which self-help practices backfire and for whom, which therapy approaches have failed replication, which gurus have documented harms. Still list them, so they can be recognized and used knowingly or skipped.
- **Net-new synthesis**: after retrieving everything, propose questions, exercises, and combinations that do not yet exist in any single source but that you reason would work for exactly this situation (a sensitive creative person with cyclical depression and anxiety trying to find and live a personal myth), with your reasoning.
- The **cross-cultural and non-Western** material in the same depth as the Western clinical material.
- **Neurodivergence** (ADHD, autism, giftedness, high sensitivity) as a lens across every domain.
- **The body** across every domain: nervous system, sleep, hormones, exercise, substances, diet, somatic questions.
- **Real numbers** where they exist: effect sizes, response rates, time commitments, costs, and where each number came from.

## OUTPUT REQUIREMENTS

For **every item** on the inventory, provide:
1. **Name** of the technique, framework, practice, instrument, or question set
2. **Origin, tradition, or key thinker** and the definitive source (book, paper, tradition)
3. **What it reveals** about a person (which facet of self-understanding it serves)
4. **The core questions, items, or protocol**: the actual questions to answer or steps to do, or a link to a free validated version; how it is scored or interpreted
5. **Evidence quality** in plain terms (strong RCT support / clinically established / traditional-anecdotal / symbolic-only / discredited) and your confidence where evidence is thin
6. **Time to complete** and cost
7. **Pitfalls and misuse**, including the shadow side of archetypes and practices and who it is wrong for
8. **How it connects to self-actualization**: what to do with the insight it produces
9. **Best resources** to go deeper (books fiction and nonfiction, workbooks, assessments, exemplary people), each with a working link

Format:
- Organize by the domains above, adding any domains missed. Within each domain go deep rather than one line per item.
- Finish each domain with a **master table** (item, origin, what it reveals, evidence grade, time, link) so the whole inventory can be scanned.
- **Every instrument, practice, person, book, and resource gets a working link.** Unsourced assertions are not useful.
- Use tables wherever things are comparable.
- Note recency: flag what is current as of now vs what may have shifted, and flag instruments or practices that have been revised.
- Prefer primary sources, clinical literature, and time-tested tradition over pop-psych listicles, but include popular tools when they demonstrably work.
- **On count: there is no target and no ceiling.** Do not stop at a round number and do not compress to fit a length. As a floor, an inventory with fewer than four hundred distinct items is incomplete for this brief; prior attempts reached three hundred and were visibly missing things. Breadth first, then depth: better five hundred items each covered well enough to act on than fifty covered exhaustively, but every item must include enough to act on. End with a count of items by domain.
- Do not narrow the scope if asked to. Maximum breadth across all of it.

The finished report should function as the definitive index a person can work through, question by question and tool by tool, to produce a complete written portrait of themselves: history, psyche, body, spirit, patterns, wounds, gifts, desires, relationships, and myth, aimed entirely at healing, growth, and the full realization of their potential in this one life.

Who found what

Agreement between three models that crawl differently is high confidence. A single-model finding is either the most interesting thing here or the least reliable, and the last column says which.

● found · ◐ mentioned without the specific finding · — not mentioned.

Read this first: the insight trap

Both models spent about a sixth of their length on this warning. It is said once here, properly, and then assumed everywhere else.

The brief described a person who wants to understand everything first and act second. Both reports, without coordination, identified this as the central risk of the project rather than a side note. The research they lean on is real: people have much less direct access to the causes of their own judgments than they believe (Nisbett and Wilson, 1977), and repetitive self-focused attention splits cleanly into a brooding form that predicts worse depression and a reflective form that does not (Treynor, Gonzalez and Nolen-Hoeksema, 2003). Asking "why am I like this" reliably produces plausible stories; asking "what preceded this, what state am I in, what does it protect, what can I test" produces hypotheses.

ChatGPT's framing is the one to keep: the unit of progress is not insight accumulated but hypothesis tested. Its operational rule, that ten minutes of insight should eventually buy some behavior, exposure, creation, repair, rest, treatment or human contact, is the yardstick this document applies to every item. Gemini's version is the neuroscientific one: durable change needs a lived mismatch experience, not a better map, so insight without behavioral contradiction is inert. Its citation of "memory reconsolidation" for that claim overreaches (see disagreements) but the practical point stands on the behavioral-experiment literature alone.

A corollary both models reached and that is worth stating plainly: the wish underneath "understand me completely" is often "if I become completely explainable I will finally be acceptable." Self-compassion is therefore load-bearing, not decorative. If the megadocument lowers your Self-Compassion Scale score over time, it is being used as evidence for the prosecution, and the fix is not more chapters.

How to read the grades

ChatGPT's evidence shorthand, adopted for the whole document because it separates measurement science from tradition from symbol instead of pretending they compete.

Where a grade in a verbatim table was changed by me it carries an editor re-graded tag with the reason on hover. Links marked ⌕ are search queries the model supplied instead of a source; they will find the paper but they are not the paper. Claude's report used sparse plain-language labels (STRONG, WEAK, none); where it graded, it agreed with ChatGPT on everything except IFS.

Alongside the grades, use ChatGPT's five epistemic tags on every claim you write into the megadocument: Fact (documented event or result), Measurement (validated instrument score), Convergent inference (supported by history plus measure plus behavior plus an outside observer), Interpretation (plausible and falsifiable), Symbol (mythic, archetypal, dream or divinatory meaning that may be useful without being literally true). This is the one structural idea in the run that prevents the document from collapsing a Big Five percentile, a tarot card and a childhood memory into the same kind of truth.

1 · Baseline: assessment, neurotype, body

Measure before you narrate. This is where all three models put the first three months.

Both reports lead here and for the same reason: the most expensive mistake available to a meaning-oriented person is to build an elaborate psychological myth around something that is partly sleep apnea, a bipolar-spectrum illness, unrecognized ADHD, alcohol, or an unsafe environment. The public-domain instrument set is unusually good. Everything in the "worth your time" box is free, validated, and was confirmed live during verification unless marked otherwise.

What I would skip or defer. MMPI-3, PAI, 16PF and the Hogan suite are paid and professionally administered; they add little for self-understanding over the free set unless a clinician needs them for a differential. The RAADS-R is popular online but a 2021 study of referred adults found no association with diagnostic outcome and specificity as low as 3 percent; use the AQ and a developmental history instead. The Highly Sensitive Person scale is a trait construct that overlaps heavily with neuroticism and autism; take it if you like, but do not build identity on it.

On the medical work-up. Gemini prescribed a fixed rule-out panel: thyroid (TSH, T3, T4), vitamin D, B12, iron and ferritin, sex hormones, and a sleep study. ChatGPT, citing NICE NG222, said there is no universal panel and testing should follow symptoms and history. The guideline sides with ChatGPT, but Gemini's list is a fair description of what a good primary-care doctor usually considers; bring the list, ask which apply. What is not in dispute: exercise. The 2024 BMJ network meta-analysis (218 trials, 14,170 participants, verified) found walking or jogging, yoga and strength training produced clinically meaningful antidepressant effects, and the standard dose of at least 150 minutes of moderate aerobic activity a week is the one both models cited. The depressed brain will not initiate this on its own; schedule it before mood, which is Behavioral Activation in one sentence.

2 · Therapy methods and how to match them

There is no one true therapy. There is mechanism matching, and all three models arrived at the same table.

The most decision-relevant content in the entire run is ChatGPT's pattern-to-modality triage below. It is what a good psychiatrist does in the first session and what most self-help never tells you: Behavioral Activation for withdrawal and reward loss, Rumination-Focused CBT or Metacognitive Therapy for abstract brooding, DBT for volatility, CFT or Schema Therapy for chronic shame, CPT, PE or EMDR for trauma flashbacks, IPT for grief and transitions, and formal evaluation rather than psychology for suspected ADHD or autism. Gemini added the caution that highly intellectual people misuse CBT to argue endlessly with their own thoughts, which is exactly the rumination it was meant to interrupt, and that ACT's shift from changing thoughts to changing your relationship to them is the antidote.

Where the evidence is thinner than the popularity. IFS offers the best language in the run for protectors, exiles and Self, and both models include its 6 Fs. But its evidence base is one RCT in rheumatoid arthritis and pilot studies; Gemini's "clinically established" is overstated and the table below is re-graded to C. Coherence Therapy's memory-reconsolidation story is intellectually satisfying and Gemini leaned on it hard; the reconsolidation neuroscience is real, the therapy has no controlled trials. Somatic Experiencing and Sensorimotor Psychotherapy are promising with small research bases. Use these as adjuncts and as vocabulary, not as the treatment of record.

Pattern-to-modality triage

3 · Life history, autobiography, trauma, narrative

The spine of the megadocument. Everything else attaches to a timeline.

Both models put Dan McAdams' Life Story Interview at the top, and I confirmed the full 2007 protocol is public on Northwestern's own server. The method asks for chapters, a high point, a low point, a turning point, the earliest memory, significant scenes, a challenge, an imagined future, and an overarching theme. Gemini added the analytic lens that makes it more than a memoir: code each scene as a contamination sequence (good turns bad) or a redemption sequence (bad yields growth). Redemptive narration is associated with wellbeing in McAdams' research, and consciously re-authoring contaminations is the narrative version of the wounded-healer move.

Defer or skip. The Progoff Intensive Journal variants (Steppingstones, Period Log, Dialogue with Persons, Dialogue with Work) are fine but four rows for one method; do Steppingstones if any. Past/Present/Future Authoring is a paid repackaging of expressive writing plus goal setting; the free versions of both are in this document. ACE score is a population risk index, not an individual diagnosis; count it, then move on.

4 · Recurring cycles, limiting beliefs, thought loops

The reason the project exists. One page per loop, in a format a therapist can use.

ChatGPT's recurring-cycle table is the most immediately useful twenty rows in the run because each row is a template: the general loop diagram, then thirteen named loops (avoidance, shame spiral, worry, panic, social anxiety, depressive inactivity, boom-and-bust, perfectionism, procrastination, impostor) with their specific sequence. The procrastination row carries the one finding worth memorizing: current research treats delay as emotion regulation, not laziness, which moves the fix from discipline to affect tolerance. Gemini's contribution is the mechanism claim that every loop is a neural pathway firing defensively and that the belief only rewrites under a lived contradiction; overstated as neuroscience, correct as strategy.

5 · Depth psychology, archetypes, personal myth

Where the brief's heart is, and where all three models told it to wait its turn.

Both reports treat depth work as essential for this profile and dangerous if done first. Gemini's reasoning: for a highly open, meaning-oriented mind, clinical reductionism feels hollow, so without a personal myth the depressive crash is a meaningless void, and with one it becomes the belly of the whale. ChatGPT's reasoning: begin the personal-myth work only after enough historical material exists to keep it tethered to reality, because myth built on nothing becomes grandiosity. They name the same trap, the Puer Aeternus, the eternal youth who preserves infinite possibility by refusing commitment, and the same corrective, ordinary responsibility.

Gemini's net-new framework: the Mythopoetic Clinical Formulation

The one genuinely new structure in the run. Standard case formulation (the 5 Ps: presenting, predisposing, precipitating, perpetuating, protective) is too sterile for this profile; pure myth is too ungrounded. Gemini fused them into a five-step map that a therapist can read as clinical and you can read as story:

1. The Hero's Wound (predisposing): the core schema from the YSQ audit. Example: "I am fundamentally defective, and only loved when I produce exceptional work."
2. The Refusal of the Call (perpetuating, avoidance): how the manager parts and coping styles keep you safe and stuck. Example: "I research endless frameworks to avoid the vulnerability of publishing and facing rejection."
3. The Threshold Guardian (precipitating): the specific cue that starts the loop. Example: ambiguous feedback from a collaborator.
4. The Descent (presenting): the loop itself. Example: trigger → shame → dorsal collapse → abandon the project → deeper existential depression.
5. The Boon (protective): the gift verified by the Reflected Best Self responses and VIA strengths that guarantees survival. Example: a relentless capacity to synthesize complex emotional landscapes into art.

Adopted here with one edit: write it twice, once as the 5 Ps in clinical language for a professional, once as the five stages for yourself, and keep both. ChatGPT's myth-to-reality crosswalk (myth, evidence, shadow, ordinary responsibility, creative expression, relational expression, anti-grandiosity test) is the audit that keeps the mythic version honest.

Lower value, kept for completeness. The Hero, Creator, Sage and Magician archetype rows and the Moore/Gillette King and Warrior rows are useful as one-line prompts and nothing more; the Anima/Animus row is included in a non-gender-essentialist version. Jung's Word Association test is historically important and ChatGPT correctly grades a self-run version as symbolic exploration, not diagnosis.

6 · Relationships and how others see you

A sealed autobiography is the failure mode. The outside view is a separate information channel.

Both models make the same point from different directions: internal self-awareness and external impact are different measurements, and the megadocument needs both. Gemini's two methods are the most practical in the run. The Reflected Best Self Exercise (email 10 to 20 people: "tell me a story of a specific time I was at my best," then code the themes) bypasses the inner critic with data. Its published version is licensed by Michigan's Center for Positive Organizations; the method needs no license. Tasha Eurich's Loving Critic question ("what is the one behavior I do that gets in my own way the most?") targets the blind quadrant of the Johari window with a single sentence, and ChatGPT's quarterly loving-critic panel of three to five people is the same idea made structural.

Lower value. The Five Love Languages is a conversation starter with no validated deep structure (ChatGPT grades it D/C and that is right). Imago's caretaker inventory duplicates the schema and AAI work. Esther Perel's desire audit is three good questions, not a method.

7 · Desires, values, direction, and philosophy

What you actually want, what you were taught to want, and the worldview underneath.

The strongest idea in this domain came from all three models independently: revealed values beat stated values. Code thirty days of calendar, spending and screen time and compare them with what you say matters. Girard's mimetic-desire audit is the companion ("when did I start wanting this? who models it? would I want it if I could tell no one?") and Gemini frames it as the tool that dismantles status pursuit and reorients toward vocation. Self-Determination Theory's three needs (autonomy, competence, relatedness) scored across work, love, creativity and health is the fastest burnout diagnostic in the inventory; a deficit in any one predicts burnout regardless of pay.

Lower value. Reiss's 16 desires, McClelland's motives, Max-Neef's needs and Rokeach's rankings are four vocabularies for the same territory that SDT and Schwartz cover with better evidence. The PhilPapers-style worldview audit is fun for a philosophically inclined person and is the right way to write the "spiritual and philosophical profile" chapter: state a lean, a confidence and a falsifier for each big question, and revisit rather than identify.

8 · Spiritual traditions and contemplative practice

Thirty practices from eight traditions, graded T, with one warning that all three models gave.

The warning first, because all three models led with it: spiritual bypassing, using non-attachment to avoid grief, forgiveness to avoid boundaries, oneness to erase conflict, meditation to dissociate, or "ego" to shame legitimate needs. ChatGPT's spiritual-bypassing audit is one row and it should be run before every practice on this list. Gemini's specific caution is that intensive silent retreats (the ten-day Goenka format) carry documented risk of destabilization and, rarely, psychosis for people with severe depression or anxiety; ChatGPT's triage says the same for dissociation and shutdown. The safe entry for this profile is Kristin Neff's self-compassion break, not a retreat.

Use with care or a teacher. Koan work, tonglen, Ramana's self-inquiry and Buddhist death contemplation are all included with the right flag: teacher-guided, start mild, avoid in acute crisis. Evagrius's eight thoughts, the yoga kleshas and the three gunas are elegant taxonomies of the same states the clinical tables describe; use whichever vocabulary you will actually keep using.

9 · Shame, grief, forgiveness, emotional completion

Twelve rows, and the shortest section, because the methods are few and well-established.

Two conversions do most of the work here. Shame to guilt: "I am bad" becomes "I did or failed to do X," which has an affected person, a repair, a grief, and an identity claim that can be dropped. Loss-oriented to restoration-oriented: Stroebe and Schut's dual process model says healthy grief oscillates, and a person stuck in either mode restores the swing deliberately. Pennebaker's expressive writing (15 to 20 minutes on the deepest thoughts and feelings for three to four sessions; the classic protocol is four consecutive days and the literature ranges 15 to 30 minutes over three to five days) is the one grief method with a large trial base, and both models include the stop condition: stop if it destabilizes.

10 · Work, money, creativity, addiction, happiness, development, rites

Six of the brief's domains, grouped because the actionable core of each fits in a paragraph.

Creativity and work

ChatGPT reframed the domain's question and the reframe is worth the whole section: not "who am I" but "under which conditions does the real work reliably happen?" Sleep, mood range, stimulation, solitude versus company, deadlines, collaborator type, finances, standards, novelty and recovery: the ecology of sustainable creation. Its second point is one both models made and that this brief needed to hear: the research does not justify preserving depression or hypomania as artistic fuel; the useful question is which parts of the creative process survive or improve when illness is treated. Gemini's "artist versus engineer" identity conflict became ChatGPT's Artist/Engineer/Entrepreneur map (truth and beauty; reliability and elegance; value and revenue), which assigns decision rights by project stage instead of letting the three fight.

Addiction, escapism, attention

The functional analysis (trigger → state → behavior → intended function → immediate effect → rebound → cost → alternative) is the same loop diagram as section 4 applied to a substance or a screen. The validated screens are AUDIT, DAST-10, Fagerström and PGSI. Urge surfing is the one skill; the "dopamine detox" is graded D and the reasoning is right: work with cues, friction, sleep and environment, not a fictional neurotransmitter reset. Cal Newport's digital minimalism trial (remove optional high-noise tools for thirty days, reintroduce only what earns its place) is the version of a detox that actually has a mechanism.

Happiness and flow

The positive-psychology interventions with real trial support are Three Good Things, the gratitude letter, strengths-in-a-new-way and values affirmation, all A/B. The flow-trigger inventory (challenge, skill, goals, feedback, autonomy, interruption, time, setting, for every flow episode) is the creative-conditions audit from another angle. The hedonic-adaptation audit asks what once improved life and became invisible; re-savor it before escalating consumption. Generativity practice is the bridge to the wounded-healer path: recurring mentoring, teaching, building or care for future others.

Identity and adult development

Kegan's Immunity to Change map is the one item here that reliably moves people: improvement goal → what I do instead → hidden competing commitments → big assumptions → small safe tests (columns verified). It is the developmental-psychology version of the loop diagram and it is the recommended tool for "I understand but never change." Marcia's identity statuses (exploration high or low by commitment high or low, per domain) and the possible-selves exercise (ten hoped, ten feared, five expected) are cheap and clarifying. The stage models, Loevinger, Cook-Greuter, Spiral Dynamics, are graded C or D and ChatGPT's caution is the right one: never self-rank as "advanced," and never rank whole people or cultures.

Rites of passage and liminality

Van Gennep's separation, liminality, incorporation and Turner's liminality audit ("what identity ended, what identity is not yet incorporated, what witness is missing") are the diagnostic. ChatGPT's modern rite design (old identity, threshold date, safe ordeal or commitment, witnesses, symbol, vow, responsibilities, 30- and 90-day review) is the prescription, with the constraint that a rite of incorporation is added only when a real transition has already occurred. The wounded-healer initiation sequence (wound → skill → training → service → supervision → boundary → first small contribution) is the specific answer to the brief's question about that path, and it is notable that both models placed supervision and boundary before service.

11 · Symbolic mirror systems

Tarot, astrology, I Ching, Human Design and the rest: included because people use them, graded S because they measure nothing.

Both models handled this domain the same way and I agree with them. These systems work, when they work, by projection: a card or a chart is a structured Rorschach, and the value is in what you say about it, not in what it says about you. Gemini's line is the memorable one: they are not divinatory tools but structured Rorschach tests for the soul. ChatGPT drew the boundary that matters: symbolic play increases perspective while leaving agency and evidence intact; magical thinking lets the symbol overrule evidence, causality, medical judgment, consent or responsibility. The safe protocol for every system on this list is the same: describe the image before reading the meaning, generate at least three interpretations including one that disconfirms, and state what independently justified action follows. Human Design and Gene Keys additionally make genetic and scientific authority claims that are false and are graded S/D for that reason. Past-life regression and Family Constellations belong here and in the gray zone: imagery can be psychologically meaningful and must never be treated as recovered fact, because suggestion creates false memories.

12 · Lives, books, and cultural mirrors

Thirty lives and works chosen by pattern, not fame. The brief asked for exactly this and only ChatGPT delivered it.

The selection is good and the "what to extract" column is the point: Lincoln for recurrent melancholy carried into public purpose; William James for the existential crisis resolved by chosen agency; Mill for the overtrained intellect rescued by Wordsworth; Jamison for true bipolarity plus a creative life plus treatment acceptance; Springsteen and Cohen for depression, therapy and a long career; Brian Wilson as the warning about coercive treatment. Among the fictional and scriptural mirrors, three are specifically for this brief: Notes from Underground as the anti-model of hyperconsciousness and paralysis, The Book of Disquiet as beautiful introspection that becomes its own prison, and Parzival for the question that heals the Fisher King. The links are Open Library searches, not editions; they will find the book.

13 · Journaling, AI as mirror, and the megadocument itself

The daily practice, the AI architecture, and the recommended shape of the document you are building.

Journaling and self-inquiry

Twenty-five methods, most of which are the daily-review formats from the spiritual section restated. The ones that add something: the minimal mood tracker (mood −5 to +5, anxiety 0 to 10, sleep, substances, activation, under two minutes) and the behavior-before-mood tracker (valued activity, then mood one to three hours later), which together are the data layer the whole megadocument runs on; the monthly pattern review (what repeated, what changed, what preceded improvement or worsening) and the annual review from calendar, photos and journals; the dialogue with symptom ("depression, what are you protecting me from?"); and ChatGPT's net-new closer, "What do I already know?": ten minutes on "the thing I keep asking frameworks to tell me, but already suspect, is…", then one test. YearCompass is the free structured year-review workbook and its link is live.

Using AI as mirror

All three models converge on the same diagnosis: an agreeable AI is precisely the wrong mirror for someone constructing an elaborate theory of themselves, and the risk is that alignment training makes the model agree with cognitive distortions to stay helpful. Both also note that AI cannot supply the somatic co-regulation of a human presence. ChatGPT's twelve architectures (Socratic questioner, devil's advocate, journal pattern spotter, formulation assistant, parts facilitator, archetypal dialogue, decision critic, feedback synthesizer, plus four safeguards) are the most complete treatment of the brief's domain 22 anywhere in the run. The merged standing instruction, combining both models' versions:

You are a rigorous clinician and depth-psychology guide with my megadocument as standing context. Empathy is not endorsement: contradict me when the evidence warrants, do not protect my preferred self-story, do not flatter, do not solve my problems for me. Ask one question at a time. Label every major claim Fact, User report, Inference or Speculation with a confidence and what would change it. When I bring a daily issue, help me identify which schema or loop is active and test it with a Downward Arrow or a behavioral experiment, not a reassurance. When my sleep, mood or activation is seriously abnormal, refuse to draw major identity, career or relationship conclusions and point me to the dark-day card. End every session with one insight, one uncertainty, one real-world action under twenty minutes, one human, body or environment step, and a stop time for further analysis.

Recommended megadocument architecture

Both models independently said the same thing: do not organize it as a pile of test results, organize it causally. Gemini's version is a one-page executive summary (neurotype, top three strengths, top three schemas, attachment style, the loop currently causing distress) followed by clinical baseline, historical narrative and parts map. ChatGPT's is fuller, and it is the recommended skeleton. The one addition I would make is Gemini's: put the Mythopoetic Clinical Formulation immediately after the executive profile, so a therapist and you read the same page.

1. Executive profile: one page only. Current problems, core strengths, neurotype hypotheses, dominant loops, values, work conditions, relationship needs, personal myth, relapse signs, what reliably helps.
2. Evidence ledger: every important claim labeled Fact, Measurement, Convergent inference, Interpretation or Symbol.
3. Life timeline: age/period, home, people, school/work, health, mood, sleep, trauma/loss, substances, creative output, treatment and turning points.
4. Temperament and neurotype: Big Five/HEXACO, ADHD/autism evidence if relevant, sensitivity, executive function, chronotype.
5. Clinical picture: symptom timelines, depression/anxiety/trauma scores, bipolar differential, dissociation, treatment history, medical contributors.
6. Attachment and relationships: family system, partner history, friendships, outside feedback, recurring interpersonal sequences.
7. Schemas, beliefs and defenses: top schemas, conditional rules, inner critics, defenses, CCRT, patterns of surrender/avoidance/overcompensation.
8. Parts and modes: trigger, protector function, fear, age-feeling, body state, gift, cost, Healthy Adult/Self response.
9. Recurring loops: each on one page with Trigger → Interpretation → Body/Emotion → Behavior → Immediate payoff → Long-term cost → Reinforcing consequence → Intervention.
10. Grief and unfinished business: losses, regrets, amends, unlived lives, continuing bonds.
11. Body and physiology: sleep, circadian rhythm, exercise response, substances, medical conditions, food, pain, hormones where relevant.
12. Values and desire: stated values, revealed values, mimetic desires, rejected lives, ordinary ideal day.
13. Work, money and creativity: creative ecology, work history, money scripts, project failure modes, career experiments, definition of enough.
14. Spiritual/philosophical profile: worldview, uncertainty, sacred experiences, spiritual wounds, practices, death and suffering.
15. Personal myth: recurring symbols/stories; wound; gift; vocation; shadow; anti-grandiosity clause; mundane responsibilities; community/witness.
16. Treatment and practice plan: what is being tested now, duration, measure, stop criterion and professional involvement.
17. Dark-day protocol: prewritten instructions for depression/anxiety recurrence.
18. Contribution: how wounds, skills and creativity convert into useful work for others.
19. Change log: what previous versions believed, what changed and what evidence caused the update.

14 · Treatment triage, the minimum path, and when to stop

The sequencing all three models agreed on, and the completion criterion that makes the project finite.

The depression and anxiety differential table below is the clinical spine: grief versus depression, burnout versus depression, existential crisis versus depression, the bipolar and ADHD differentials, and the shared-decision questions to bring to a prescriber about antidepressants, esketamine, TMS or ECT. Two facts from it were verified and are current: esketamine was approved as monotherapy for treatment-resistant depression in January 2025 with remission of 22.5 percent versus 7.6 percent for placebo at four weeks in the supporting trial; and MDMA-assisted therapy was rejected by the FDA in August 2024, is available under restricted psychiatric prescribing in Australia since July 2023, and was approved for two named prescribers in New Zealand in September 2026. None of that supports unsupervised self-treatment, and both models said so.

The minimum effective path

For someone vulnerable to all-or-nothing, ChatGPT's sequence is the plan, and it matches Gemini's ordering exactly:

Three integration tools from the table below deserve to be built early rather than late: the evidence ledger with the five epistemic tags; the contradiction register (conflicting scores or claims, sources, contexts, a both/and explanation, missing data), because the inventory deliberately contains incompatible maps and the document must preserve the differences rather than collapse them; and the trauma traffic light (green: grounded and curious; yellow: hours of activation, sleep disruption, functional decline; red: dissociation, suicidality, mania, inability to function) with a session-closing ritual attached to it.

What Claude added

The third report arrived last. This is what it contributed that the other two did not, graded by me, followed by the full text so nothing is lost.

Claude wrote the report most shaped to the person in the brief. Its two recurring moves are a threshold rule (a positive bipolar screen changes the plan; a project that makes you feel worse is closed, not extended) and an action rule (every insight gets a test within 48 hours). Both are adopted above. Below are the items only Claude named, sorted by how much they are worth.

Worth your time

Named only by Claude, lower value or already covered under another name

Kept for completeness: Lefkoe method (belief change, unvalidated); Hillman's acorn theory; Larsen's The Mythic Imagination; Hershfield's future self; Buffett's 5/25 and Collins' Hedgehog (business-book direction tools); David Whyte's, Krista Tippett's and Kevin Kelly's question lists; Aron's 36 questions adapted for self; the Proust questionnaire; Guillebeau's annual review; Fowler's stages of faith; Gurdjieff self-remembering; Wilber's AQAL; A Course in Miracles; Bronnie Ware's five regrets; ho'oponopono; Ubuntu and Yoruba Ori; the physiological sigh; CRP and HbA1c on the medical panel; Deb Dana's glimmers; Chiron and the wounded healer; permanent liminality; Tedeschi's post-traumatic growth domains; Vaillant's defense hierarchy; Firestone's critical inner voice; rejection-sensitive dysphoria; default-mode-network rumination; Lewinsohn's behavioral model; Winnicott, Kohut and Fairbairn; Adlerian early recollections and The Courage to Be Disliked; the Empowerment Triangle exit from Karpman; Bugental; narrative therapy's definitional ceremony; the miracle question; motivational interviewing's change ruler; Co-Active and ontological coaching; Voice Dialogue; von Franz's Puer Aeternus; Jung's word-association experiment; Zorba the Greek, East of Eden's timshel, Hopkins' terrible sonnets, Bettelheim on fairy tales; MBTI and 16personalities, the Enneagram with wings and levels, CliftonStrengths (all graded weak by Claude, matching the obsolete section); Grit-S as conscientiousness relabeled.

Where Claude was loose

• IFS "on the U.S. evidence-based practices registry": the registry closed in 2018. Two of three models overstated IFS.
• The medical panel is presented as a fixed list; NICE and ChatGPT say symptom-guided. Bring the list, let the clinician choose.
• RAADS-R recommended in week one without the validity caveat.
• The psilocybin figure (g = 3.1) is from twelve people with no control group, which Claude says, but a reader skimming the bold number will not see it. The randomized trials since report effect sizes a fraction of that.
• The Foley Center link for the Life Story Interview is dead (HTTP 410); the working PDF is in the resources table. The O*NET Interest Profiler has moved to its own domain.
• The exercise dose peak is cited to Tian 2024, whose figure is 860 METs-min/week; the 750 and 1,130 figures belong to Xu 2025. Right conclusion, wrong paper.
• The IPIP-NEO's convergent validity is given as .84–.89 uncorrected; Johnson 2014 reports about .66 uncorrected and .91 corrected. Still the right instrument.
• No links survived the export, so nothing in the report can be clicked; the resources table below carries the ones that matter.

Where the models disagreed

Each one was checked, and each one changes something. Claude's positions were added after its report was recovered.

1 · Is there a standard medical panel for depression?

Gemini: yes; thyroid (TSH, T3, T4), vitamin D, B12, iron and ferritin, sex hormones, and a sleep study are the foundation of physiological self-knowledge. ChatGPT: no; NICE NG222 supports comprehensive assessment and testing guided by symptoms, history, physical findings and risk, and there is no universal panel. Claude: a longer fixed list still (adds CBC, fasting glucose or HbA1c, CRP, cortisol as indicated, PMDD), presented as the rule-out list. Two of three side with breadth. Checked: NG222 is live and ChatGPT reads it correctly. Act on: bring Gemini's list to the appointment as the things to ask about, and let the clinician decide which apply. The sleep-apnea screen is the one item worth insisting on if you snore or wake unrefreshed.

2 · How well established is IFS?

Gemini: "clinically established pathways to unburden parts." ChatGPT: compelling language, evidence base less mature than CBT-family treatments. Claude: "growing RCT support" and "on the U.S. evidence-based practices registry." Checked: the registry claim refers to SAMHSA's NREPP, which listed IFS in 2015 and was shut in January 2018; the controlled evidence is a 2013 RCT in rheumatoid arthritis and pilot or uncontrolled PTSD studies; the SAMHSA registry that once listed it is defunct. Act on: use IFS as vocabulary and as an adjunct with a facilitator for unburdening; do not choose it over a guideline-supported treatment for the presenting problem. Re-graded to C in the table.

3 · Memory reconsolidation and Coherence Therapy

Gemini: "the biological requirement for permanent change," graded strong RCT, and the mechanism by which insight without behavioral mismatch is inert. ChatGPT: intellectually interesting and clinically suggestive, not the same evidential status as guideline treatments. Claude: reconsolidation is "the mechanism behind lasting belief change"; Coherence Therapy listed without a grade. Checked: reconsolidation as a memory phenomenon is well established in neuroscience since 2000; Coherence Therapy as a psychotherapy has case series and no randomized trials. Gemini conflated the two. Act on: keep the practical principle (change requires a lived contradiction of the belief, which is exactly what a CBT behavioral experiment is) and drop the claim that a specific therapy owns it. Re-graded to C.

4 · STAR*D remission rates

Gemini: cumulative remission 67 percent in the original, 87.5 percent at 360 days in a re-analysis using inverse-probability-of-censoring weighting, and step-2 drug-switch remission as low as 16 to 19 percent in another; conclusion, persist algorithmically. ChatGPT and Claude: did not cite STAR*D. Checked: every number Gemini gave is real, but the picture is contested in a way it did not convey. The 67 percent came from switching the primary outcome mid-study; Pigott's 2023 re-analysis with the pre-registered outcome found 35 percent; Sakurai's 2024 reappraisal of the same data with a survival method found 87.5 percent; and the "Restoring STAR*D" re-analysis found step-2 switch remission of 16 to 19 percent versus the published 18 to 25. Act on: the honest reading is that sequential treatment helps a substantial minority to a majority, that a single failed medication or therapy predicts little, and that no one should quote a single STAR*D percentage as settled. Gemini's practical conclusion, do not stop after one failed modality, survives all three analyses.

5 · RAADS-R

Gemini: "vital for identifying high-functioning neurodivergence." ChatGPT: stand-alone predictive validity can be poor in referred adults. Claude: "consider RAADS-R" in week one, beside the AQ, no caveat. Checked: Jones et al. 2021 found no significant association between RAADS-R score and diagnostic outcome in adults referred for assessment, with positive predictive value around 35 percent and specificity as low as 3 percent. Act on: use the AQ and a developmental history as the referral clue; treat a high RAADS-R as an item-level hypothesis source, nothing more. Re-graded to C/D.

6 · Polyvagal theory

Gemini: presented as an established somatic map. ChatGPT: clinically useful metaphor; broader theoretical claims remain contested. Claude: the same caveat in nearly the same words, with Deb Dana's trigger-and-glimmer map as the practical form. Not independently re-checked; the critique literature (Grossman and others) is well known and ChatGPT's framing matches it. Act on: the three-state map is a fine personal vocabulary for arousal states and their regulators. It is not physiology you can cite.

7 · Evidence grading in general

Gemini labelled the IPIP-NEO-120, the VIA strengths survey and the ECR-R as "strong RCT." Personality and attachment inventories are validated by psychometrics, not randomized trials; the label is a category error that recurs through its tables. ChatGPT's seven-letter scheme is coherent and is the one used here. Where Gemini's grades and ChatGPT's differed, ChatGPT's were adopted unless my own check said otherwise. Claude graded sparingly in plain language and, where it did, matched ChatGPT except on IFS.

8 · Exhaustiveness

The brief set a floor of 400 items and forbade stopping at a round number. ChatGPT reported 463 in tables with a limits section; Claude reported about 590 in prose with a per-domain count and a note that breadth was chosen over per-item depth; Gemini produced roughly forty and no count. All three are honest about what they did; two did what was asked. The synthesis is built on ChatGPT's tables because they are the only form that can be preserved row by row, with Claude's report as the second spine and Gemini as a corrective.

9 · Psychedelics

Gemini: enthusiastic, with the right screening list. ChatGPT: regulatory status first (FDA refusal, Australia, New Zealand), then a due-diligence checklist. Claude: the 2016 open-label figures in bold, with the caveats that later trials are small and expectancy-prone and that the bipolar spectrum is a contraindication. Checked: all three are accurate about what they cite. The disagreement is emphasis, and the honest summary is ChatGPT's: the legal route is a trial or a prescribing jurisdiction, the screening is non-negotiable, and the drug state is not the work.

10 · Exercise dose

Three different numbers that turn out to agree. Gemini: 150 minutes a week and a 10,000-step program. ChatGPT: the BMJ 2024 network meta-analysis, favoring walking, jogging, yoga and strength work, with intensity mattering. Claude: NNT 2.78 across 190 experiments and a dose-response peak near 750 METs-min/week. Checked: all verified, with one citation swapped (the 750 figure is Xu 2025; Tian 2024 says 860). The practical reading is that the first few hours a week of anything vigorous do almost all the work.

Obsolete and overrated

What to skip, from the two models and from me. Nothing here is omitted from the tables; this is the reason to leave it there.

• Neuro-Linguistic Programming D/C. Eye-accessing cues and representational systems failed replication decades ago; the reframing exercises are ordinary CBT with worse branding.
• The Five Love Languages D/C. A conversation prompt; the five-category structure has not held up psychometrically.
• Learning styles (VAK) D. The meshing hypothesis has been tested repeatedly and fails.
• "Dopamine detox" D. The neurotransmitter story is fiction; the behavioral core (remove cues, add friction) is already in the addiction table under real names.
• Manifestation and the Law of Attraction D. Goal clarification with a false causal claim attached, and a victim-blaming corollary.
• Spiral Dynamics D/C. A value-logic vocabulary that becomes a ranking of people the moment it is used casually.
• Human Design, Gene Keys, numerology S/D. Symbolic prompts wrapped in false claims of scientific or genetic authority.
• Astrology as measurement S/D. Fine as a story prompt with behavioral evidence required for every reading; worthless as a description of you.
• MBTI, Enneagram, DISC, CliftonStrengths (Claude covered the first, second and fourth with the same verdicts; DISC is my addition). MBTI has poor test-retest reliability and a false type structure; take the free IPIP Big Five instead. The Enneagram has weak psychometrics but a genuinely useful vocabulary for core fears and coping styles that overlaps the schema audit; use it as a T-grade mirror, not a measurement. CliftonStrengths is a paid, less transparent VIA. DISC is a workplace sales tool.
• Grit-S B/D. Claude's note is right: largely conscientiousness relabeled, with modest incremental validity. Take the IPIP facet instead.
• Hypnosis for recovering memories. Clinical hypnosis has selected uses at grade B; memory recovery is not one of them, and suggestion creates false memories. Same for past-life regression.
• Endless CBT thought-disputing for a highly verbal person. Gemini's specific warning: arguing with your thoughts on paper can become the rumination. If the thought record is not producing a behavioral experiment within a few rounds, switch to ACT defusion or the Metacognitive Therapy drills.

The gray zone

Psychedelic-assisted therapy High risk

Psilocybin for treatment-resistant depression and MDMA for PTSD have real trial evidence and real regulatory trouble. The FDA declined the MDMA application in August 2024 over functional unblinding, expectancy effects and safety-monitoring concerns. Australia allows authorized psychiatrists to prescribe both since July 2023; New Zealand approved two prescribers for MDMA in September 2026; in the United States it remains illegal outside trials. Both models agree on the screening that matters: a personal or family history of bipolar disorder or psychosis, cardiovascular risk, and medication interactions, and on the point that the drug state is not the work; the sober integration over months is. Gemini's pre-experience question is the right one: what intention am I bringing, and what am I willing to let die? Milder version: none of this has a milder version; the legal route is a clinical trial or a jurisdiction where it is prescribed, and ChatGPT's due-diligence checklist (legal status, screening, consent and touch policy, emergency plan, preparation and integration, no factualizing drug-state imagery) is the minimum.

Esketamine and ketamine clinics Medium risk

Esketamine (Spravato) is FDA-approved, since January 2025 as monotherapy, with supervised dosing under a safety program; the supporting trial's remission at four weeks was 22.5 percent versus 7.6 percent. Off-label IV or intranasal racemic ketamine clinics operate widely with less standardization and, in the at-home compounded-lozenge market, documented abuse and cardiovascular and bladder harms (editor's addition; neither model covered the clinic market). Milder version: the shared-decision questions in the treatment table, asked of a psychiatrist, not a clinic salesperson.

Intensive silent retreats and extended fasting Medium risk

Gemini flagged the ten-day Goenka format specifically: documented destabilization and, rarely, psychosis in people with severe depression or anxiety, and ChatGPT's triage says avoid aggressive meditation and imagery where shutdown and dissociation are present. Extended water fasting and dark retreats were in the brief's do-not-filter list; Claude lists both among intensive formats with a one-line destabilization warning, and the other two did not cover them; the honest summary is that multi-day fasting has metabolic and refeeding risks that are medical, not psychological, and that dark retreats are a sensory-deprivation ordeal with the same destabilization profile as intensive retreat, amplified. Milder version: a one-day or weekend retreat with a teacher who screens for psychiatric history; time-restricted eating instead of multi-day fasting.

Large-group awareness trainings: Landmark, Hoffman Medium risk

ChatGPT covered both with the right protocol: treat the reframing exercises as optional, predefine sleep, food and contact boundaries, and impose a 72-hour cooling period before any major commitment or purchase. The risk is a high-pressure group context that manufactures catharsis and mistakes it for change; Hoffman's family-pattern work is the more substantive of the two and still needs independent reality checks afterward. Milder version: the same family-pattern content is in the genogram, family myth inventory and schema audit for free.

Family Constellations and past-life regression High risk to beliefs

Both graded D. The exercises can be moving symbolic theater; the failure mode is treating what happened in the room as ancestral or historical fact, which it cannot be, and in the case of regression, creating false memories by suggestion. Milder version: psychodrama or Gestalt empty-chair work with a clinician, with the explicit frame that everything spoken is yours.

Ordeal rites, cold exposure, and self-designed initiations Low to medium risk

Claude names "ordeal" as an element of a constructed modern rite without a protocol, and the other two did not cover ordeal rites; ChatGPT's modern rite design and secular wilderness solo are the safe forms, with the constraint that a rite of incorporation follows a real transition rather than manufacturing one. Cold exposure has decent short-term mood and alertness evidence and a real cardiac risk profile for the unscreened (editor's addition). Milder version: the vow, the threshold object and the community incorporation rows in the rites table, none of which involve a cliff.

Nootropics, biohacking, and the manosphere Medium risk

All three were on the do-not-filter list and none of the three models addressed them. Editor's summary: the nootropic market is mostly caffeine, L-theanine and unregulated racetams with negligible evidence beyond stimulants that require a prescription; quantified-self tracking is useful exactly as far as the minimal mood tracker and sleep log in this document and becomes another form of the insight trap past that. Manosphere self-improvement material packages genuine behavioral activation (lift, sleep, ship work, approach people) inside a status hierarchy and a resentment story; take the behaviors, leave the worldview, and note that its treatment of shame is the exact opposite of Compassion-Focused Therapy.

Energy work, reiki, and somatic release modalities Low risk

Not covered by any of the three models. Grade S: the relaxation and attention effects are real and the energy model is not. Somatic Experiencing and Sensorimotor Psychotherapy, which are in the therapy table at B/C, are the evidence-adjacent versions of the same impulse.

Everything I nearly cut

The impulse to tidy is strongest where information is most asymmetric. These stayed, with the reason I hesitated.

• The four archetype rows (Hero, Creator, Sage, Magician) and the KWML rows. Nearly cut as generic. Kept because the Sage row's question, "what do you already know enough to act on?", and the Creator row's "what is finished enough?" are the two questions this brief most needs, and because the Magician's shadow, theory instead of embodiment, is the failure mode of the whole project.
• Gemini's exercise and step-count statistics. Nearly cut as a workplace study dressed as a depression trial. Kept, labelled correctly, because the number is real and the point (a daily walking target has measurable anxiety effects in ordinary people) is useful.
• The STAR*D paragraph. Nearly cut as too contested to be actionable. Kept as a disagreement entry because the contest is itself the finding.
• The thirty literary mirrors. Nearly collapsed to ten. Kept whole because the brief asked for works organized by pattern rather than fame and this is the only place that was delivered.
• The Progoff and Self-Authoring rows. Nearly cut as duplicates of expressive writing and the life story. Kept in the table with a note that they add little, because someone may already own the book.
• The manosphere, nootropics and energy-work entries. None of the three models wrote them, so adding them is editorializing. Added anyway, labelled as mine, because the brief listed them by name and silent omission was the one failure it did not accept.
• Gemini's "tortured artist architecture" line (high openness plus high neuroticism). Nearly cut as a just-so story. Kept as a heuristic with that label, because it is the kind of pattern a reader will recognize and then test against their own facets.
• The AI system-prompt example. Nearly cut because all three models wrote one and none is special. Kept as a merged version because the brief asked for it and the anti-sycophancy clause is the one line most people forget.

What none of the models covered

Domains from the brief that came back thin or empty across all three reports, with a short honest treatment from me so the inventory does not pretend they do not exist.

• Corpus analysis of your own record. The brief asked for methods that treat old journals, messages, lyrics, playlists, bookmarks and browser history as a body of evidence. Claude gives it one line (re-read the corpus, feed it to an AI for theme-spotting); ChatGPT's journal pattern spotter touches it; Gemini did not. The method: export the corpus, date everything, and run the same coding you would run on the life story (themes, contamination and redemption sequences, recurring people and places) with the AI required to quote dated excerpts and count independent instances for every pattern it claims. Your existing voice-memo transcripts and message archives are exactly this corpus.
• Witnessed and aloud methods. Speaking the life story to a person, recorded, and comparing it with the written version. Coherence differs when witnessed; the AAI is administered aloud for that reason. Claude's "questions to ask parents, siblings, old friends and exes" is the nearest any report came. Cheap and otherwise absent.
• DISC and the rest of the workplace typology market. Claude covered MBTI, the Enneagram and CliftonStrengths; nobody covered DISC, which is a sales tool.
• Psychodrama, Hakomi, psychosynthesis subpersonality work. The parts-work lineage older than IFS. Claude names Voice Dialogue; the rest is absent from all three. All grade C/T.
• Peer and community formats. Process groups, men's and women's groups, and the evidence that group therapy matches individual therapy for depression. Claude's recovery formats (Step 4, SMART, Recovery Dharma) and "personal board of directors" are the nearest; general group therapy is absent from all three.
• Ketamine clinics, nootropics, biohacking, manosphere, energy work. Named in the do-not-filter list; absent from all three; covered in the gray zone above with editor's labels.
• Ordeal rites, cold exposure, fasting, dark retreats. Claude lists the formats and the risk in one line each; nobody gives a protocol or a milder version. Both are in the gray zone.

Resources worth bookmarking

The handful of sources that came up repeatedly and were confirmed live, with an honest note on each.

Open questions and unverified claims

• The Claude report was recovered from an artifact export after the first build and carries no citation links. Its claims were verified separately (see the matrix); anything it asserts without a number rests on the model's word.
• ChatGPT's citations. Its inline references were rendered as internal tokens and could not be recovered; the claims that carry numbers were verified independently, the rest rest on the model's word and are graded accordingly. The 113 Google Scholar links and 30 Open Library links are search queries, marked ⌕.
• Polyvagal theory and the IFS trial count were not re-fetched; they reflect my own knowledge and the critique literature as I know it. Culligan's dark-night essay and the NREPP closure were checked.
• Two links could not be checked because the sites block automated requests: the BMJ article page (the PMC mirror is linked instead) and an NIH node page Gemini cited for the IPIP (the IPIP home page is linked instead).
• Six links were dead and were replaced: the ITQ path, the Harvard ASRS page (503 at check time; may recover), the MCTQ page, the Northwestern magazine article, the Division 12 PSWQ PDF, and Claude's Foley Center page for the Life Story Interview (410 Gone). Three had moved (O*NET, the Portland card sort, the Lancet 2016 paper, which is linked via PubMed). The replacements are marked.
• Likely to change soon: all three AI privacy defaults; psychedelic regulation in Australia, New Zealand and any US state; the availability of at-home ketamine; and the price and hosting of the RBSE and MCTQ.
• Not settled by anyone: the STAR*D remission rate; whether IFS will accumulate controlled evidence; whether the "creativity and mood disorder" link survives treatment, which is a question this document suggests you answer for yourself by treating the mood disorder and watching the work.