Ontology Of Psychiatric Conditions: Taxometrics

Lucid explainer (reposted from SSC with edits) on taxometrics: whether psychiatric conditions are categorical (like flu or species -- a real hidden generator, a natural cut-point) or dimensional (like height or wealth -- arbitrary cutoffs on a continuum). Walks through why you can't just eyeball a symptom histogram (the MAXCOV trick of looking for where inter-symptom correlations peak), then the Haslam-McGrath-Kuppens 2020 meta-analysis: ~75% of conditions unambiguously non-taxonic, with only odd, hard-to-trust exceptions (IED, autism, addictions). The memorable payload is the Jeff-Bezos-in-Extremistan move -- 'quantity has a quality all its own,' so severe ADHD is to normal absent-mindedness as Bezos is to a median earner: same continuum, but far enough out to deserve its own word -- and the DSM critique (diagnostic bins mistaken for cosmic categories), with the practical clinical shift from 'does this patient have ADHD?' to 'how much trouble do they have paying attention?'

Quality 76 · Excellent. Excellent floor. A genuinely clarifying treatment of a confusing, important question, anchored by a sticky analogy (Bezos/Extremistan) and a real practical upshot for diagnosis and person-first language. Held at the floor because the categorical/dimensional distinction is a field's, not his, and it's a (well-edited) repost.

Claude’s paradigm shift 55 · Moderate. Notable. Taxometrics predates the post, but the lucid synthesis plus the quantity-has-a-quality / Extremistan framing and the DSM-as-useful-bins point updated how informed readers think about diagnosis.

Real-world impact 3 · Moderate. Moderate. An influential reframing within the psychiatry/rationalist discourse on diagnosis and the DSM; stays within that professional sphere. Niche-professional.