WEIRD Science

Just how far can experimental results in the behavioural sciences, typically carried out on American undergraduates, be generalised? This article reviews a recent paper which discusses this issue in depth.

A fellow MAPPster recently brought to my attention a paper by Joseph Henrich and colleagues of the University of British Columbia, with the provocative title of “The weirdest people in the world?”  This is a “target article” (that is, an article published with the explicit aim of starting a discussion in the pages of a journal) which explores the extent to which the results from typical empirical experiments in psychology can be generalised beyond subjects drawn from what the authors refer to as WEIRD (Western, Educated, Industrialized, Rich, and Democratic) societies.

My initial reaction was to groan at the rather laboured and redundant construction of this acronym (most of the five terms which make it up tend to be associated with each other), and ‑ predictably ‑ the authors of some of the published comments on the target article weighed in with equally terrible examples of their own, such as ODD (Observation- and Description-Described, a “shorthand” for “lacking in ecological validity”), or ‑ from the group of commentators who disagreed the most strongly with the target article ‑ WRONG, constructed in toe-curling fashion from When Researchers Overlook uNderlying Genotypes.  It didn’t help much, either, to discover that the very first examples given by Henrich et al. (2010) of the contrast between other cultures and American undergraduates ‑ the group at the core of the term WEIRD ‑ were of initiation rites of boys within tribes in New Guinea, which appeared to have been selected principally for their ability to induce snickering in the reader.

However, once I got over my cheap indignation at these minor stylistic issues, I found the article to be well thought-out, with much to interest psychologists and other social scientists, as well as occasional philosophers of science (such as me).  The authors’ starting point is to ask, how generalisable are laboratory experiments on a homogeneous group of mostly young, middle-class subjects, drawn from an ethnically skewed population sample even within their own country, to the entire population of the world, as implied by articles whose conclusions include statements starting with “People tend to…”?  The target article starts at the outermost level by comparing American attitudes and responses to experiments such as optical illusions and simple economic simulation games to those of various small-scale societies worldwide (none of whom I had never heard of), then “spirals in” by comparing Western and non-Western large-scale societies, Americans and other Westerners, and typical current American experimental subjects with the U.S. population, past and present.  Their conclusion, in each case, is that what they call WEIRD people are the outliers, from whom it is difficult, and perhaps dangerous, to generalise results to other populations.

The comments ‑ by anthropologists, economists, psychologists, and other scholars ‑ are diverse and rewarding to read, revealing just how much scope for discussion there is on almost any interpretation placed on a phenomenon in the social sciences.  They address, among many other issues, problems with experimental design, the possible effects of culture on brain function and learning processes, the specific limitations imposed by the English language, the prospects for reaching other populations via the Internet in a WIRED (!) world, the influence of the fact that the experimenters are themselves overwhelmingly WEIRD, and whether or not the “differences” observed between populations are, perhaps, simply failures of communication.

One topic which I would have liked to have seen addressed was the question of how important it actually is for results to be widely generalisable.  Although it would be nice to discover some universal truths about how people behave, I suspect that very few that are truly worth knowing have not already been adequately described by Shakespeare or the ancient Greeks; meanwhile, the area of application for many empirical discoveries in the behavioural sciences is likely to remain local to the cultural group where these discoveries are made for a while yet, because of the way that academia and social services are structured (the advent of the Internet notwithstanding).  Science already faces an uphill battle to convince large sections of the world’s population of the truth about evolution and climate change; the dangers of applying the wrong kind of positive psychological interventions to people from a specific culture may not yet be a big problem.  We still know little enough about the WEIRD people.

Overall, the target article and its associated comments represent a good debate on a number of issues which are relevant to psychologists, not least positive psychologists who are hoping to develop interventions.  Although the most spectacular contrasts are, naturally enough, between American undergraduates and members of remote tribes, it is not necessary to go all that far from the laboratory within our own countries to find people and (sub-)cultures which differ radically from our experimental environments, requiring us to review the general applicability of our findings.  This work is an excellent investment of the time taken to read it, and deserves a wider ‑ or should that be WIDER? ‑ audience.

Reference: Henrich, J., Heine, S. J., & Norenzayan, A. (2010).  The weirdest people in the world?  Behavioral and Brain Sciences, 33(2–3), 61–135.  doi:10.1017/S0140525X0999152X  Retrieved June 5, 2012 from

Author’s Bio:

Nick Brown is a MAPP student at UEL, trying to reconcile the need for better science in positive psychology with a looming realisation that trying to measure anything truly interesting with objective data is close to impossible.  He welcomes all comments, rebuttals, and tips – of either the “help” or “PayPal” kind – at

Would you like to comment?

Leave a Reply