
recoursion-plasticity-ai-q4xa7rn.gamma.site
Recoursion, Plasticity, and the AI Field
Labyrinth A deep report for the Labyrinth · 4 May 2026 How the hyperiterative human-AI process illuminates a more general dynamic — one not invented by AI, but made newly visible, rapid, and open to intervention within it. Read the Report
This paper can be read as a genomics paper, but it is best understood as a node within a larger shift: from diagnosis to formulation, from pathology to ecology, from category to context, and from reduction to relation.
"The modern score architecture is real; the disease realism attached to it is far less secure."
Related work on differential susceptibility emphasises that some traits increase responsiveness to both adverse and supportive environments, rather than merely increasing vulnerability to harm. The same inherited sensitivity that elevates risk under adversity may confer advantage under sufficiently supportive conditions.
This is why psychiatry in an affluent rural setting can feel like a different discipline from psychiatry in inner-city London.The clinic may carry the same diagnostic manual, but the ecologies of deterioration differ profoundly.

The formulation model is not merely a clinical courtesy. It is an epistemically superior framework for integrating multi-level causal information. Unlike categorical diagnosis, which forecloses explanatory complexity in favour of administrative legibility, formulation preserves the multiplicity of causes whilst orienting clinical action. It is the framework that psychiatry already possesses — and that it consistently under-deploys in the rush toward biomedical reductionism.
The point is not to glorify suffering, nor to adopt an uncritical neurodiversity romanticism that minimises genuine distress and impairment. It is to ask whether some of the traits now over-represented in psychiatric populations might historically have conferred advantages under different ecological conditions — or might still do so in supportive, non-domesticating environments. That is where the inversion becomes genuinely interesting, and where psychiatry's current institutional arrangements become most visible as contingent social productions rather than neutral medical infrastructure.
"Modern Western institutions may not merely reveal disorders that were always there. They may manufacture deterioration by arranging environments that are hostile to certain kinds of sensitivity, nonlinearity, plasticity, and cognitive style."
If different people occupy different Umwelten — different sign-environments constituted by their particular sensitivities, histories, and relational worlds — then psychiatric categories do not merely describe distress; they also transform it.
It would also require abandoning the fantasy that one can understand psychosis-spectrum, affective-spectrum, or attentional-spectrum phenomena by using diagnosis as both the input to genetic analysis and the output to be explained. If the categories are heterogeneous — and the transdiagnostic genomics strongly suggests they are — then studies that treat them as homogeneous endpoints will continue to produce findings that are statistically impressive but conceptually circular. The field's methodological ambition has far outpaced its conceptual ambition; closing that gap is the most urgent intellectual task in psychiatric genetics today.
"It would also stop speaking as though category assignment had solved the explanatory problem."
E = GΓΔ²

"The real challenge is not whether biology matters. It is whether psychiatry can develop a biology subtle enough to live inside formulation, ecology, and lived worlds rather than pretending to float above them."