Autism And Intelligence: Much More Than You Wanted To Know

An MMTYWTK resolving the autism-intelligence paradox: genes for autism correlate with HIGH IQ and are positively selected (Polimanti & Gelernter), yet diagnosed autistic people average LOWER IQ (69% below 85). Resolution via three causes: (1) common 'familial' genes (raise IQ, positively selected); (2) rare de novo mutations (lower IQ, selected out); (3) non-genetic insults (obstetric complications, etc., lower IQ). Gaugler et al apportion ~52% familial / 3% de novo / 41% unaccounted. But even familial-only autism (Ronemus) still trends low-IQ, so the paradox isn't fully dissolved by mutations/insults. Scott offers the memorable 'tower-vs-foundation' model (intelligence is a tower needing a mysterious foundation; too-tall-for-the-foundation fails into autism — explaining both de-novo low-functioning cases and the IQ-gene association), and weighs Crespi's perceptual-vs-verbal/rotational-imbalance speculation skeptically. Ends with confidence-calibrated numbered conclusions (1-2 ~100%, the tower-model 25%, Crespi's specifics 3%).

Quality 78 · Excellent. A rigorous, clear resolution of a genuine paradox grounded in robust behavioral/population genetics (familial vs. de novo, positive selection), with the memorable tower-vs-foundation model and honest confidence-calibrated conclusions. Excellent-low.

Claude’s paradigm shift 60 · Notable shift. The familial-vs-de-novo resolution plus the tower-vs-foundation model is a fresh, clarifying synthesis of the autism-IQ literature. Notable.

Real-world impact 2 · Minor. A within-discourse genetics deep-dive; no material-world effect. Within-blog influence.