A research program for the science of experience.

The central goal is to identify recurring neural patterns associated with distinct experiential states, and to determine the conditions under which those patterns generalize across individuals and contexts.

Research output isn't an end in itself — every validated archetype sharpens four downstream domains at once:

• → training and alignment data for frontier AI labs
• → ground truth for brain–computer interface state inference
• → consumer neurofeedback and entrainment via Ground
• → empirical anchor points for philosophy of mind

The research pipeline proceeds in five stages, each feeding the next:

1. 01

   Neural data collection

   EEG recordings captured on Ground under standardized session protocols.

2. 02

   Phenomenological reporting

   Structured first-person reports submitted alongside each neural recording.

3. 03

   Machine learning clustering

   Unsupervised methods identify candidate neural patterns across participants.

4. 04

   Architect classification

   The Architect agent evaluates candidate clusters against phenomenological tags and assigns or proposes archetypes.

5. 05

   Taxonomy refinement

   The Architect updates the living taxonomy as new data enters the Archive.

• Q1. Can neural archetypes be reliably detected across individuals?
• Q2. How stable are these archetypes across contexts and over time?
• Q3. Can AI systems infer experiential states directly from neural signals with calibrated uncertainty?
• Q4. What level of resolution does experiential inference require, and where does additional resolution stop helping?