Discoveries

Every session teaches the methodology something.

As practitioners log sessions, the system looks for patterns across hundreds and then thousands of them and brings the strongest to you. Nothing changes in EBDT until you approve it.

Open your review queue

Sunset over calm water with rings spreading out

EBDT Hub · Discoveries · Cindy’s viewAdmin only

The learning set

Sample data

612sessions in the learning set

38practitioners

92%consent on file

3patterns waiting for you

Frequencies and layers

Sample data

How often each frequency appears with each layer. Tap a cell.

Emotional with Adaptation: together in 41% of sessions.

Common pathways

Sample data

Share of sessions that follow each path through the architecture.

Predictions → Protective Responses → Regulation → Reactions18%

Experiences → Encoded Adaptations → Energetic Patterns → Symptoms11%

Meanings → Adaptive Operating Rules → Physiological Mechanisms → Behaviors9%

Your review queue

Sample data

Candidate patterns. Nothing reaches a practitioner until you approve it.

Candidate A

“Predictions and Protective Responses often appear together when the expression is a Reaction.”

Seen in 214 sessions from 27 practitioners.

Candidate B

“When Body Messages are identified, the Emotional frequency is usually present too.”

Seen in 163 sessions from 22 practitioners.

Candidate C

“Sessions that begin in Origin reach a remedy record in fewer steps.”

Seen in 98 sessions from 15 practitioners.

Methodology v1.4 draft

Sample data

Approved patterns land here. When you publish v1.4, every practitioner’s guide updates.

A pattern appears only after it shows up across at least 25 sessions from 5 practitioners, so no single client or practitioner can be picked out.

How it learns

Four steps, and you hold the last one.

Capture

Every step saved as structured data, not loose notes.

Protect

Names removed, consent checked.

Find

Each night the engine compares sessions and counts what appears together.

You decide

Approved patterns become a new methodology version, tracked in GitHub, and every practitioner’s guide updates.

An honest answer

Why not train a custom AI model now?

Training a model on your sessions would bake your methodology into something you can’t fully inspect or take back. Structured data plus your review queue keeps every change visible, reversible and yours. A custom model can come later, when there are enough sessions to justify it.

Sunlight through oaks in a woodland clearing

“I believe some of our most meaningful discoveries begin when we’re willing to question what we think we know, remain curious about what we don’t yet understand, and explore the possibilities that emerge when different perspectives come together.”

Cindy Costley See the plan