Michael Darius Eastwood · Independent AI alignment researcher
Published
The paper of record is Marco Catani (King's College London) and Paolo Mazzarello (University of Pavia), "Leonardo da Vinci: a genius driven to distraction", published in the peer-reviewed neurology journal Brain in 2019 (volume 142, issue 6, pages 1842 to 1846). The authors reviewed Leonardo's documented working patterns across his manuscripts, contemporaneous accounts, and later biographies. They argued that Attention-Deficit Hyperactivity Disorder is the most plausible explanation for his well-attested pattern of starting more projects than he finished, extreme absorption in isolated details, restless mobility across cities and courts, and reportedly disrupted sleep. The paper is careful; it uses "hypothesis" and "the most plausible explanation" rather than "diagnosis". It is a serious argument in a serious journal, and it is the piece to cite when the retrospective-assessment question comes up.
Michael Fitzgerald (Trinity College Dublin) has published a body of retrospective-assessment work on historical figures across autism and ADHD in the psychiatric literature. Simon Baron-Cohen (Cambridge) has framed a related empirical construct, "hypersystemising", as a cognitive pattern producing strong abstract-system pattern-recognition at elevated rates in autistic populations. Both invite the same caveat as Catani and Mazzarello: the argument is about the plausibility of a cognitive profile, not a clinical fact for any named deceased individual. What both bodies of work do jointly is argue that reasoning patterns which look unusual in a modern schoolroom look convergent with the working patterns of some of history's most productive synthesisers.
The alignment argument in Hernandez-Espinosa, Abrahao, Witkowski and Zenil (PNAS Nexus 5(4) pgag076, 2026) is structural: cognitive diversity is a protective factor for alignment research because a monoculture misses the same failure modes in the same direction. The retrospective-assessment literature is an independent piece of evidence for the mechanism the alignment paper names as protective: the cognitive style that produces cross-domain integration has a documented track record of catching what disciplinary specialists miss, precisely because the mind is not filtering through a single discipline's default frames. The claim is structural, not personal.
Not a claim that the present author is Leonardo. Not a claim that ADHD or autism is a superpower. Not a claim that any retrospective assessment of a historical figure is a settled clinical fact. The Polymathic Neurodivergent Profile working paper (Paper C v2.0, 2026) disavows superiority framing on the same page it describes the profile, and this note is consistent with that disavowal. The polymathic method is claimed. The stature of its historical exemplars is not. What is on the record is a mechanism argument tied to peer-reviewed literature; what is not on the record, and should not be, is a personal claim to be counted among the exemplars.
From the book Infinite Architects: Intelligence, Recursion, and the Creation of Everything by Michael Darius Eastwood.