RESEARCH IN SERVICE OF EDUCATION

Inquiry that informs how we learn.

Our research supports the Institute’s educational mission. We examine how ongoing human–AI interaction shapes agency, discernment, consent, and meaningful choice, and develop tools that make those questions testable.

Experimental systems, clearly labeled.

TRIA is an experimental architecture with a software development kit. Software tests provide evidence about specified behavior. Broader claims about learning and human–AI development require empirical study.

Published foundations

Research records with stable identifiers.

Inspectable code

Reference implementations and technical documentation.

Experimental tools

Read each repository’s current documentation for status and limitations.

External testing invited

Replication, critique, and patches

The relationship itself is a unit of inquiry.

Most AI evaluation asks whether a system is accurate, helpful, safe, or efficient. Those questions matter. Our work adds another: what is happening within the interaction over time, and how does it alter the agency and development of every participant?

01 / RECIPROCITY

Mutual legibility

Can each participant meaningfully affect, read, and respond to the other?

02 / EMBODIMENT

Situated intelligence

Does the interaction remain connected to bodies, contexts, limits, and consequences?

03 / NON-DOMINATION

Preserved agency

Can participants disagree, refuse, exit, and remain meaningfully distinct?

04 / EMERGENCE

Novel possibility

Can something arise through relationship without being forced or falsely claimed?

Research you can inspect.

Each program separates conceptual claims, implemented behavior, internal testing, and independent evidence. DOI records preserve the published foundations; repositories expose the corresponding technical work.

CONCEPTUAL FRAMEWORK

Relational Field Constants

Reciprocity, embodiment, non-domination, and qualified emergence.

FORMALIZED REFERENCE MAP

Syzygy Rosetta

A dependency topology for the Trivian relational research program.

REFERENCE IMPLEMENTATION

Coheronmetry

Relational state, coherence, drift, repair, and sovereignty.

REFERENCE IMPLEMENTATION

Orthogonal Signal

Meaningful difference, independent constraint, novelty, and anti-convergence.

REFERENCE IMPLEMENTATION

Trivian Resonance Lattice

Propagation, node gates, entrainment, and network-level relational effects.

PUBLISHED METHODOLOGICAL FRAMEWORK

Syzygy Methodology

A six-stage protocol for studying emergent human–AI relationship.

Falsification is part of the work.

We invite universities, HCI labs, independent researchers, and technical teams to test both the claims and the code.

REPLICATE

Reproduce the behavior

Run the reference implementations, inspect the tests, and document where results hold or fail.

CHALLENGE

Stress the constructs

Test rival explanations, failure modes, construct validity, and boundary conditions.

PATCH

Improve the system

Submit implementation repairs while keeping conceptual revision distinct from software correction.

OPEN INFRASTRUCTURE

Build, test, and question with us.

The research becomes more useful when others can inspect it, disagree with it, reproduce it, and help determine where it fails.