The target article persuasively demonstrates the perils of building and testing psychological theories almost exclusively on Western, Educated, Industrialized, Rich, and Democratic (WEIRD) people. We wholeheartedly agree that knowledge based on WEIRD samples is limited at best and misleading at worst. The authors' plea for broadening the knowledge base of psychological science is valid, timely, and of prime importance for the future development of psychological and behavioral sciences.

Whereas the authors' main recommendation for a better science is to sample participants strategically from diverse populations, we recommend an additional research strategy based on a socio-ecological perspective. Just as ecological biologists study animals' behaviors in relation to their natural habitats (e.g., Stutchbury & Morton Reference Stutchbury and Morton2001), socio-ecological psychologists study how natural and social habitats affect human mind and behavior. The first step we recommend is, instead of dismissing the research on WEIRD people, to consider the simple yet critical question of “Why are WEIRD people so weird to begin with?” Once potential explanations are considered, researchers can go on to systematically test these potential causes of human diversity in mind and behavior as the second step.

There already are several recent studies using a socio-ecological perspective that provide important clues as to why WEIRD people are so different from non-WEIRD people, and what factors might cause diversity in mind and behavior (see Oishi & Graham, in press, for a review). Fincher et al. (Reference Fincher, Thornhill, Murray and Schaller2008), for example, wondered why nations, and indeed, regions, vary on individualism versus collectivism (with the former being typical of WEIRD countries). They showed that historically low levels of pathogen prevalence might be one reason – people in nations with low pathogen prevalence do not need to draw sharp ingroup-outgroup distinctions the way people do in nations with high pathogen prevalence. Similarly, Kitayama et al. (Reference Kitayama, Ishii, Imada, Takemura and Ramaswamy2006) wondered about the origins of rugged individualism in the United States, and hypothesized that it is in part due to the history of the frontier spirit. They found that people in Hokkaido, the northern island of Japan with a history of frontier spirit, showed a greater degree of American-style individualism than did mainland Japanese who don't have such a history.

Oishi and colleagues (Oishi et al. Reference Oishi, Lun and Sherman2007; Reference Oishi, Ishii and Lun2009a) likewise wondered why Americans typically define themselves in terms of personality traits, skills, and abilities rather than collective attributes (Markus & Kitayama Reference Markus and Kitayama1991) and show conditional group identification (Cialdini et al. Reference Cialdini, Borden, Thorne, Walker, Freeman and Slone1976). They demonstrated that high residential mobility might be partly responsible for such patterns of self-concepts and group identification (see Oishi, in press, for review). Uskul et al. (Reference Uskul, Kitayama and Nisbett2008) speculated that the degree of economic interdependence might be in part responsible for analytic thinking dominant in the United States. They examined whether herders (who are economically independent) would show a greater degree of analytic tendency than farmers and fishermen (who are more dependent on others in their economic activities) in the same single region of Turkey. Indeed, they found that Turkish herders showed more analytic tendencies than did farmers and fishermen. Likewise, Yamagishi et al. (Reference Yamagishi, Hashimoto and Schug2008) hypothesized that preference for unique choice (Kim & Markus Reference Kim and Markus1999) among Americans is due in part to open as opposed to closed social systems, and they demonstrated that this was indeed the case. Finally, one of the target article authors' own seminal research (see Henrich et al. Reference Henrich, Boyd, Bowles, Camerer, Fehr, Gintis, McElreath, Alvard, Barr, Ensminger, Henrich, Hill, Gil-White, Gurven, Marlowe, Patton and Tracer2005) has importantly shown that market integration and payoffs to cooperation in daily economic activities predicted cross-societal variations in behavioral responses in the Ultimatum Game.

A socio-ecological perspective may help us to understand not only cross-societal variation, but variation within WEIRD populations, too. For example, mean punishment expenditures from the Public Goods Game described in the target article's Figure 4 show high diversity within WEIRD samples: The United States and Australia are at one end and Germany and Denmark are at the other end. The socio-ecological perspective helps us to generate various hypotheses regarding this variation. For instance, the United States and Australia are nations high in residential mobility, whereas Germany and Denmark are nations low in residential mobility. Could these within-WEIRD variations be the result of societal differences in residential mobility (and temporariness of group membership)? If so, would there be a comparable within-nation variation in punishment behaviors between residentially mobile cities (e.g., Atlanta) and residentially stable cities (e.g., Philadelphia), a within-city variation between residentially mobile people and stable people, and even a within-person variation between the times when people are thinking about moving and the times they are thinking about staying? In the area of self-concept and conditional group identification, Oishi and colleagues have found such within-society variations as well as cross-society variations (see Oishi, in press, for review).

We are of course not claiming that all variation between human populations is due to socio-ecological factors. There is no doubt that biological and evolutionary forces also play an important role. However, a socio-ecological perspective does provide a concrete framework for searching for the causes of diversity and universality of mind and behavior.

In conclusion, we agree that psychological knowledge should not be solely based on WEIRD people. We also agree with Henrich et al. that it is important to include large and diverse samples in our science. We recommend two additional steps for researchers. First, ask the simple, yet important question of “Why are WEIRD people so weird?” Second, test whether any potential socio-ecological factors that might make WEIRD people weird account for societal, regional, and individual variations in a broad array of phenomena central to human psychology. These two additional steps are critical because they can convert the research on limited WEIRD samples from a major liability (as the authors suggest) to a major asset from which we can build and develop the type of psychological and behavioral sciences that the authors promote in their target article – the psychological and behavioral sciences that illuminate the causes of universality and diversity in mind and behavior. Instead of dismissing the research based on WEIRD people, we can start a better science from it! We believe that a socio-ecological perspective is particularly helpful to this end.