The Control Group Is Out Of Control

A definitive essay on the limits of empiricism. Parapsychology is "the control group for science": Bem's meta-analysis of psi follows almost every proposed rigor-fix (huge N, exact replications, low p, symmetrical funnel plot, clean p-curve) and still "proves" precognition. The lesson: those fixes all target mistaking noise for signal, but none touch confounding or experimenter effects - shown by Wiseman & Schlitz (identical protocol; the believer's runs find psi, the skeptic's don't) and pervasive researcher-allegiance effects. Ends with the "Ouroboros of Scientific Evidence", each authority resting on another, and the lament that science was meant to defeat reliance on biased experts, not join them.

Quality 89 · Excellent. One of the deepest and most-cited essays in the corpus - the "parapsychology as control group" frame and the noise-vs-bias distinction are genuinely original, important, and devastatingly argued. Just below the very top tier reserved for his most culturally pervasive landmarks. Top-Excellent.

Claude’s paradigm shift 70 · Notable shift. A fresh, influential 2014 reframe that changed how many informed readers think about meta-analysis and the replication crisis (rigor fixes address noise, not experimenter/allegiance bias); builds on Ioannidis and Yudkowsky, but the framing propagated widely. Major-ish shift.

Real-world impact 3 · Moderate. The 'parapsychology as the control group for science' reframe and the noise-vs-bias distinction shaped how science-literate and rationalist readers think about meta-analysis and the replication crisis. Real but bounded epistemic influence within a fairly specialized audience, with no material or population-level change — modest RWI.