A Perspective discusses the use of external validation using independent datasets for improving human neuroimaging studies. @dscheinost.bsky.social @matt-rosenblatt.bsky.social www.nature.com/articles/s41...

External validation is a great way to improve rigor in neuroimaging. Although a simple idea is simple—test a predictive model on another dataset—the practice is not. In our Perspective in @natmethods.nature.com, we discuss practical issues and solutions; www.nature.com/articles/s41...

Another years-in-making piece, *just in time* for our symposium at APS tomorrow! Introducing a new IdioNomo Framework for blending idio & nomo models in psych & beyond. @psychonetrics.bsky.social & I are glad to finally share this conceptualization with paths & pitfalls osf.io/preprints/ps...

So glad to finally share work years in the making! What started with fMRI effect size benchmarks ended up showing that conventional study planning detects ~half of expected brain effects, suggesting more distributed processes than the lit shows (+ new methods for mass univ effect/power estimation)