Bard and colleagues make a compelling case that developmental psychologists should increase the field’s cultural scope – via international collaborations, shared databases, and leveraging global scholars and samples for theory development and testing. We too are passionate advocates of the value of global collaborations yet are also keenly aware of the steep trade-offs in terms of efficient resource allocation. Consortiums of laboratories and international collaborators can develop and test far fewer theoretical hypotheses than a similar number of laboratories independently pursuing their own research questions using local convenience samples.

The need for greater cultural diversity throughout psychology may simply be an inconvenient truth – effortful and expensive to achieve, but necessary. Still, the expected value of building and testing theories using diverse samples and contributors is ultimately a function of how much cross-cultural variability in the focal phenomena truly exists. Bard and colleagues draw on the landmark “WEIRDest people in the world” article (Henrich, Heine, & Norenzayan, Reference Henrich, Heine and Norenzayan2010). However, the WEIRD case, like much of psychology, was built on small studies filtered by bias toward statistically significant results. Our meta-analysis of cross-cultural laboratory experiments (Cyrus-Lai et al., Reference Cyrus-Lai, Jha, Schaerer, van Aert, van Assen and Uhlmann2025) reveals that between-subject designs, the most prevalent design type in this literature, had only 7% power to detect interactions reported between experimental condition and national group. The less common within-subjects designs, in which each participant was exposed to all conditions, scored higher in terms of observed power, yet were still underpowered (48% power), suggesting that the area of cultural differences research may be prone to false positives.

The strongest evidence against the assumption of massive cross cultural variability stems from crowd-sourced replication initiatives that repeat the same studies across laboratories in dozens of countries (e.g., Klein et al., Reference Klein, Ratliff, Vianello, Adams, Bahník, Bernstein, Bocian, Brandt, Brooks, Brumbaugh, Cemalcilar, Chandler, Cheong, Davis, Devos, Eisner, Frankowska, Furrow, Galliani and Nosek2014). Pre-registered replications of findings in flagship journals in social psychology reveal less cross-cultural heterogeneity than one would expect, based on the theoretical claims of cultural psychologists. Replicable effects tend to vary across sites in magnitude but not sign, while non-replicable effects showed no heterogeneity (Olsson-Collentine, Wicherts, & van Assen, Reference Olsson-Collentine, Wicherts and van Assen2020).

Replication projects specifically targeted at cross-cultural findings have also failed to support the thesis that Western populations constitute statistical outliers. In Many Labs 2 (Klein et al., Reference Klein, Vianello, Hasselman and Nosek2018), most previously observed or theoretically expected cross-national differences did not reappear. Schimmelpfennig et al. (Reference Schimmelpfennig, Spicer, White, Gervais, Norenzayan, Heine, Henrich and Muthukrishna2024a, Reference Schimmelpfennig, Spicer, White, Gervais, Norenzayan, Heine, Henrich and Muthukrishna2024b) argue that Many Labs 2’s international samples were insufficiently non-WEIRD (e.g., disproportionately composed of college students and therefore more educated than average). However, Henrich and colleagues (Reference Henrich, Heine and Norenzayan2010) cited numerous cross-cultural comparisons involving college students as well as reviews of such evidence that supported their thesis. Thus, comparisons of college students across countries remain relevant evidence when debating the extent and pervasiveness of cultural differences. Further, samples from non-WEIRD populations were also typically larger in replication projects than in the original studies. Thus far, no convincing evidence exists that any of the WEIRD dimensions consistently moderate the results of crowd replication projects.

Several more recent replication projects converge with the results of Many Labs 2. Cao and colleagues (Reference Cao, Carstensen, Gao and Frank2024) attempted to replicate 12 foundational findings from cross-cultural psychology using participants from the U.S. and China (a non-Western, non-individualistic, and non-democratic society). Effects sizes for only 5 of the 12 expected cultural differences in effects were statistically significant in the same direction as in the original studies. Five tasks revealed no cultural differences, and two tasks were associated with statistically significant reversals of the expected East-West difference. Tierney and colleagues (Reference Tierney, Hardy, Ebersole, Viganola, Clemente, Gordon, Hoogeveen, Haaf, Dreber, Johannesson, Pfeiffer, Huang, Vaughn, DeMarree, Igou, Chapman, Gantman, Vanaman, Wylie, Storbeck, Andreychik, McPhetres and Uhlmann2021) attempted to replicate four findings on American work morality, finding strong similarities in judgments and behaviors among participants from the U.S., Australia, the U.K., Canada, Ireland, and India. Delios and colleagues (Reference Delios, Clemente, Wu, Tan, Wang, Gordon, Viganola, Chen, Dreber, Johannesson, Pfeiffer and Uhlmann2022) also demonstrate that findings in the field of international strategic management that are reproducible in the original setting, typically generalize to alternative time periods and geographies – results directly contrary to that field’s assumption of sensitivity to context.

Henrich and colleagues (Reference Henrich, Heine and Norenzayan2010) present cultural differences in the Müller-Lyer illusion (Segall, Campbell, & Herskovits, Reference Segall, Campbell and Herskovits1966), as a cautionary tale regarding assumptions of universality. However, Amir and Firestone (Reference Amir and Firestone2025) review evidence that such differences were not observed by other researchers and that the influential Segall et al. (Reference Segall, Campbell and Herskovits1966) findings were compromised by questionable research practices. No doubt many cultural differences, including in visual perception (Kroupin et al., Reference Kroupin, Davis, Lopes, Konkle and Muthukrishna2025), exist, but small-sample, pre-2012 studies on the topic are not necessarily a reliable guide.

The above discussed literature provides evidence of bias in the published literature of cross-cultural psychology and that replication projects showed low reliability of cross-cultural differences. Hence, the replication crisis in cross-cultural psychology cautions against extending the assumption of cultural differences to other fields. Highly impactful findings (Isager et al., Reference Isager, van Aert, Bahník, Brandt, DeSoto, Giner-Sorolla, Krueger, Perugini, Ropovik, van ’t Veer, Vranka and Lakens2023) and grand theories that make universalist claims such as shared intentionality and attachment theory, as critiqued by Bard and colleagues, should indeed be subject to input from cultural experts and empirical generalizability tests including Many Labs collaborations. However, we suggest that most developmental psychologists can continue conducting localized science using convenience samples and collaborators while adopting epistemic humility regarding how far their effects generalize. Like Bard and colleagues, we also stress the importance of “constraints on generality” statements (Simons, Shoda, & Lindsay, Reference Simons, Shoda and Lindsay2017). We believe this is a more efficient strategy to advance science than focusing on cross-cultural collaborations. As they gain impact and importance, a subset of localized projects can then be followed up by international collaborations. A key element of these small-, medium-, and large-scale projects should be open science (e.g., preregistering studies, open data, and publicly sharing research materials) to reduce bias as much as possible.

If the presumption of universality is an intellectual bias, so too is the assumption of cultural differences. Should efforts to increase cultural diversity in developmental psychology yield modest returns in terms of uncovering differences, as already seen in social psychology, such efforts might then be reserved for only the most important discoveries and theories. If the results thus far are any indicator, we may discover as many similarities that unite us as differences to study and celebrate.