The Noetic Archive

The Noetic Archive collects three kinds of information for every neural recording session: neural signals, structured phenomenological reports, and contextual metadata. Together, these form a single, queryable dataset that grows denser and more discriminating with every contribution.

Where most neuroscience datasets capture brain activity in isolation, the Archive insists on a parallel record of what the neural recording was like from the inside. That parallel record is what makes the dataset usable for phenomenological inference — and what makes it the first of its kind at scale.

Neural archetypes are recurring patterns of neural activity associated with particular experiential states. They may exist at multiple levels of abstraction — from broad categories such as focused attention or open awareness, to highly specific signatures tied to narrowly described states.

Because experience rarely resolves to a single label, the Archive permits multiple experiential tags to be associated with a given neural state. Archetypes are not assumed; they are discovered as the dataset accumulates.

The Architect is an internal AI agent that processes incoming data and organizes it into structured representations. It analyzes neural signals alongside reported experience, assigns them to existing categories — or creates new ones when necessary — and continuously refines an evolving taxonomy of experiential states.

A separate user-facing agent, the Noetic Interface, makes the Archive queryable in plain language. Both are described in detail on the How It Works page.

As new data is added, the Architect refines the Archive's taxonomy — identifying new archetypes, reorganizing categories, and improving the mapping between neural states and experiential descriptions.

The taxonomy is therefore a living artifact rather than a fixed schema — closer in spirit to a field guide that is rewritten as new species are observed than to a predetermined ontology.

Almost no dataset like this exists today. The Archive is infrastructure that several adjacent fields are quietly waiting for — and its value compounds with every consented session contributed.

• Frontier AI labs — a new class of training and alignment data grounded in real human experiential states.
• Brain–computer interfaces — ground truth for state inference, calibrated against described experience rather than assumed labels.
• Neuroscience research — empirical access to the structure of experience at a scale individual labs cannot assemble.
• Philosophy of mind — measurable correlates for questions that have been almost entirely speculative.

Each new contribution sharpens the taxonomy and broadens what the Archive can answer. That's the moat: the data improves the asset that produces it.