In this commentary, I argue that to properly assess the significance of the cross-cultural findings reviewed by Henrich et al., one needs to understand better the causes of the variation in performance in experimental tasks across cultures.