Some Unintuitive Properties Of Polygenic Disorders
Read the original on Astral Codex Ten →
Summary
Rebuts E. Fuller Torrey's two best anti-hereditarian arguments about schizophrenia via a simple liability-threshold simulation. (1) Twin concordance is only 15-50% despite ~80% heritability: because the disorder is rare, you need BOTH high genetic AND high environmental risk to cross the threshold; an identical twin shares the genes but gets a fresh random environment, so most don't cross (the sim yields ~20%; a 1970 paper gives ~33%). (2) The Nazi eugenics program killed most schizophrenics but the next-gen rate was normal: the ~163 high-genetic-risk people who DIDN'T develop it (good environment) vastly outnumber the ~20 who did, so removing the 20 barely dents the gene pool. Includes an honest EDIT correcting a 4x->2x weighting error.
Why this score
Quality 68 · Strong. Strong / useful explainer - cleanly dissolves two genuinely counterintuitive arguments that recur in the hereditarianism debate (it's the post ACX-659's eugenics rebuttal cites). Held at mid-Strong rather than higher because it's short, informal (a spreadsheet 'discount-rate simulation' with an acknowledged error), and narrow (one paper, one model) - high pedagogical value, light execution.
Claude’s paradigm shift 42 · Moderate. Moderate - the liability-threshold model is old (the post cites a 1970 paper); this popularizes/applies it to debunk Torrey rather than originating anything.
Real-world impact 2 · Minor. Within-discourse behavioral-genetics explainer; widely referenced for inoculating readers against anti-hereditarian arguments, but no material effect.