How to read Cor.

What Cor is

Cor is a reference work in the lineage of the Allen Brain Atlas, the Human Cell Atlas, and the Human Connectome Project. Those projects are scientific infrastructure — they integrate primary literature into a structured, navigable map of an empirically complex object. They mark their gaps. They take interpretive positions where the evidence forces them to. They ship incrementally. They are not finished theories. They are tools the field uses to do further work.

Cor occupies that genre for the human motivational-emotional architecture.

What Cor is not

Cor is not a finished specification in the engineering sense. There is no machine-readable schema that ships today, no operational eval suite, no audit protocol you can run against a deployed system. The specification — formal, versioned, testable, operational — is the layer being built on top of the atlas. The atlas is what exists now.

Cor is also not a rejection of mainstream psychology or psychiatry. The evidence base draws heavily on findings produced by those fields: Panksepp's affective neuroscience, Bowlby's attachment work, attachment-stress physiology, the post-DSM critique published from inside psychiatry by Insel and others, the replication crisis findings from social psychology. What Cor rejects is one specific interpretive layer: the assumption that a motivational-emotional signal deviating from a population norm constitutes a disorder located in the individual that should be patched at the level of the individual. The reference body for that rejection comes from inside the fields, not from outside them.

What is empirical consensus

The evolutionary conservation of subcortical affective systems across mammals. The attachment system as an evolved domain-specific architecture. Allostatic load as a cumulative cost of chronic stress mismatch. Wanting–liking dissociation in dopaminergic reward circuitry. Phylogenetic depth as a constraint on plasticity. Developmental calibration windows. The empirical existence of mismatch effects in physiological domains — metabolic, circadian, microbiome, light, sleep. Each of these is supported by multiple independent literatures and is not in serious dispute.

What is Cor's synthesis

The integration of those individually-established findings into a single layered architecture (Foundations → Convergences → Mechanisms) and the application of mismatch logic across non-physiological domains — meaning, status, grief, parenting, coalition belonging, reproduction. The synthesis is supported by evidence at the layer of the source findings; the integration choices are Cor's. Where the source fields disagree on an interpretation Cor takes a position on, Cor marks it on the atlas page in the Interpretive Calls section.

What is still unproven

The publicly listed gaps document the questions the atlas does not yet have a defensible answer to. Among them: the degree to which "moral cognition" is a unified domain versus a coalition of more basic mechanisms; the strength of mismatch effects in domains where the evolved baseline is poorly characterized; the empirical status of reproductive-architecture claims in environments where the relevant adaptations have not been directly tested. The gaps page is updated as the atlas grows.

What you can do with Cor now

Read the cases page if you want to see the framework operate on lived experience. Read the atlas page if you want the structural layer — foundations, mechanisms, convergences, demonstrations, challenges, gaps. Read the working papers if you want the alignment-facing argument. Use the evaluation rubrics in the applications section if you are designing an environment, a product, or an AI system and want a structured way to ask whether it serves the architecture or exploits it. Treat Cor as a reference, not a verdict — the way a clinician treats an atlas of human anatomy.