The clinical model · canonical

Seven phases. One arrow. The data drives the therapy.

"XAI Therapy is a therapist-led clinical model that uses explainable artificial intelligence to illuminate between-session behavior, strengthen therapeutic insight, and improve outcomes — without replacing human judgment." — Lance Nabers, LPC
The Intern, explaining

The Intern explains "This page is the whole playbook — no secret version behind it. What you read here is what your therapist practices and what the tools enforce. That's the entire trick: there is no trick."

The stance

The 45-minute problem

A therapist sees forty-five minutes of a week that holds 10,080. Classical practice has no choice but to discount the rest of the client's lived cognition — the sudden spiral, the Tuesday win, the avoided call — because nothing captured it. XAI Therapy's claim is architectural: when the between-session hours produce structured, framework-tagged, client-authored data, the session no longer has to guess at the week.

The model does not drive the outcomes. The data drives the therapy. Present-moment behaviors and cognitions create the clinical flow; the plan bends to the week the client actually lived, not the week the modality assumed.

Foundations

What the model stands on

Every foundation is already embodied in shipped tools — this table is an inventory, not a wishlist.

FoundationIn the modelIn the tools (built)
Behavioral learning & conditioningBehavior change is learning; reinforcement is engineered, not hoped forBehavioral activation, activity tracking, habit loops, the worlds' reward economies
Conflict learning theory"We cannot learn or grow if we don't struggle or experience conflict. There is no learning without a conflict in the mix." — the founder. Struggle isn't the obstacle to growth; it's the mechanism. The model never engineers conflict away — it engineers it to be survivable and visibleGraded exposure (fear ladder), window of tolerance, distress skills — every challenge is a deliberately held struggle
Narrative & reparative therapyThe client's story is clinical material; re-authoring is interventionShadow journal, letters across time, inner-child work, the story-worlds themselves
Unconditional positive regardThe relational floor: no shame in the data, everThe no-shame laws coded across the apps — an unfinished task is simply unfinished
Present-focused dataPresent-day data guides the reading of the past and the building of the futureCapture at the moment of occurrence; timestamped patterns
Mechanisms of change

What the model moves — and measures

MechanismWhat it means here
Active engagementThe delivery system — a tool no one opens has an effect size of zero; the story-worlds exist for this
AwarenessOf state, trigger, pattern — the first product of capture
Connection to selfThe client sees their own data and recognizes themselves in it
Connection to othersConjoint work: together-firsts, bids, repair — every hand that helped is named
Alignment & acceptanceThe gap between stated values and the lived week, made visible without shame
Purpose & passionThe "why" that powers tasks — values clarification, life vision
Problem solving · processes & systemsClients learn the system of their behavior, not just its episodes
Key languageThe client's own words are data; the exact phrase of the automatic thought is preserved verbatim
What drives — needs not metBehavior is read as need-signal, not defect
Task completionThe between-session act — the model's unit of momentum
Data collectionContinuous, client-authored, framework-tagged — never surveillance; the client sees everything collected
Telehealth-nativeDesigned for distributed care from the first line of code
The clinical flow

The seven phases

The same spine already runs the apps — five acts and thirty-one challenges in the flagship, five seasons and thirty-one together-firsts in the family world. One step at a time, never a flood.

Gullhaven at lamplight
Phase 1

Onboarding & informed transparency

Goal · consent that actually informs; the client meets the glass box
Stance · supportive, direct
Steps · welcome → what is captured and why → who sees it → the story begins
Client task · first capture — a single mood log, the smallest possible yes
Data signal · first engagement only; nothing clinical yet
The feast under lanterns
Phase 2

Therapeutic rapport

Goal · the relational floor — unconditional positive regard, in person and in code
Stance · supportive, neutral
Interventions · engagement-first tools: check-ins, gratitude, grounding
Client task · return — that's the whole assignment
Data signal · engagement rhythm: when, how often, which doors they open
Hands at work
Phase 3

Assessment — continuous, not episodic

Goal · baseline the person, not the intake hour
Stance · objective, neutral
Interventions · biopsychosocial assessment, mood/sleep/activity baselines, values work, genogram where systemic
Client task · complete the baseline instruments in-world
XAI law · assessment never closes — every week of data is re-assessment. Data signal: first patterns — time-of-day risk, avoidance clusters, language.
The workshop
Phase 4

Behavioral model learning

Goal · the client learns their own model — trigger → thought → behavior → consequence
Stance · objective, direct
Interventions · thought records, emotion wheel, worry parking, the pattern-review ritual ("here is Tuesday, in your own words")
Client task · capture honestly; predict their own patterns and check the prediction
The namesake phase: the client becomes explainable to themselves. Data signal: distortion frequencies, belief ratings, trigger topology.
The first robot's eyes light
Phase 5

Task completion

Goal · momentum — the between-session act as the unit of change
Stance · supportive, direct
Steps · one task at a time → completion honestly verified by the tool's own signals, never a self-checked box → reinforcement paid immediately
Interventions · graded exposure steps, behavioral activation, skills practice, conjoint together-firsts
Data signal · completion telemetry — avoidance has a timestamp now
The village
Phase 6

Skill building & generalization

Goal · from tool-use to trait; skills leave the app and enter the life
Stance · neutral, supportive
Steps · re-run mastered tools in harder contexts → teach-back → reduce prompting
Interventions · assertiveness (DEAR MAN), wise mind, resilience map, couples alignment, practice replays
Data signal · unprompted tool use; declining latency between trigger and skill
The island of lanterns
Phase 7

Integration & ownership

Goal · the client keeps their story — literally
Stance · supportive
Steps · outcome review against Phase-3 baselines → integration statement → data sovereignty: the record is theirs to hold and take
Interventions · letters across time, integration statement, the finale ceremonies
Data signal · the full arc — exportable, explainable page by page
Session structure

The weekly rhythm

  1. The data opening (5 min). The week as it actually was — the organized, cited between-session record replaces "so, how was your week?"
  2. The present-moment focus (10 min). What the data surfaced, and what the client brings today. The flow is created here, not pre-scripted.
  3. The work (25 min). The intervention the pattern calls for — modality-agnostic; the therapist's own clinical school rides on the XAI chassis.
  4. The task (5 min). One between-session commitment, chosen from the pattern, scheduled into the world.
The therapist stance

Four words from the founder's board

Neutraltoward the client's choices
Objectiveabout the data
Supportiveof the person
Directabout the pattern

The data makes directness kind: "the record shows Tuesday" lands softer — and truer — than "I think you avoid."

The Intern, pointing

The Intern, pointing at the important part "Boundaries, because we mean them: this is not a crisis service. It never diagnoses. And nothing an AI produces reaches a client without a human who can explain it. Efficacy claims wait until studies earn them — three are proposed, with the conditions that would prove us wrong. On purpose."

The framework names the standard. This model practices it.

The apps deliver it. The Lighthouse lets the clinician steer it. The client keeps the story.

Request early access