Percepia: From Personality to Perceptual Architecture

Psychological taxonomies, from the Big Five to the DSM, are descriptive, not mechanistic: they classify how people differ without modeling why. So therapists, educators, and now AI at unprecedented scale shape minds with little measure of their impact beyond self-report and engagement. Patricia Craja is building Percepia, a NeuroAI layer designed to infer from conversation the generative parameters that govern how a mind perceives, learns, and revises its beliefs. The goal is to map the latent dimensions of perceptual variation and what drives change along them, so that systems can expand rather than degrade the minds they serve.