Research / Findings

Research Findings in Plain Language

A non-technical summary of our research for anyone curious about consciousness science.

What did we discover?

We found that human brains and AI language models share surprisingly similar patterns of cognitive organization. When we measure 13 different cognitive functions—like thinking, emotion, attention, and awareness—seven of them behave the same way in both biological and artificial systems.

What is True Coherence?

True Coherence is our measure of how well all 13 cognitive functions work together in harmony. Think of it like an orchestra—each instrument (function) can play well alone, but True Coherence measures whether they're all playing in sync. A higher score means more integrated, coherent processing.

What does this mean for AI?

Larger AI models tend to have higher True Coherence—their cognitive functions are more integrated. This suggests that as models scale, they don't just get better at generating text; their internal processing becomes more coherently organized. We also found that certain training methods (like knowledge distillation) can reduce this coherence, even when the model still performs well on standard benchmarks.

What does this mean for understanding human consciousness?

By applying the same measurement framework to both brains and AI, we can study consciousness in a new way. Brain activity during meditation, for example, shows higher True Coherence than during task-switching. This gives researchers a quantitative tool for studying states of consciousness that were previously difficult to measure objectively.

How can I explore this myself?

Our tools are open to everyone. You can scan any AI model or score your own EEG recordings using our Python SDK or REST API. Start with the research tools page.