Most People Are Not WEIRD (2010) PDF

Vol 466|1 July 2010 OPINION Most people are not WEIRD To understand human psychology, behavioural scientists must stop doing most of their experiments on Westerners, argue J...

Vol 466|1 July 2010 OPINION Most people are not WEIRD To understand human psychology, behavioural scientists must stop doing most of their experiments on Westerners, argue Joseph Henrich, Steven J. Heine and Ara Norenzayan. M uch research on human behaviour and decides how much of a fixed amount to offer affect the way that experienced investors psychology assumes that everyone a second player, who can then accept or reject make decisions about the stock market6. shares most fundamental cognitive this proposal. If the second player rejects it, We offer four suggestions to help put and affective processes, and that findings neither player gets anything. Participants theories of human behaviour and psychology from one population apply across the board. from industrialized societies tend to divide on a firmer empirical footing. First, editors and A growing body of evidence suggests that this the money equally, and reject low offers. Peo- reviewers should push researchers to support is not the case. ple from non-industrialized societies behave any generalizations with evidence. Second, Experimental findings from several disci- differently, especially in the smallest-scale non- granting agencies, reviewers and editors should plines indicate considerable variation among market societies such as foragers in Africa and give researchers credit for comparing diverse human populations in diverse domains, such horticulturalists in South America, where peo- and inconvenient subject pools. Third, granting as visual perception, analytic reasoning, ple are neither inclined to make equal offers agencies should prioritize cross-disciplinary, fairness, cooperation, memory and the herit- nor to punish those who make low offers4. cross-cultural research. Fourth, researchers ability of IQ1,2. This is in line with must strive to evaluate how their GRACIA LAM what anthropologists have long findings apply to other populations. suggested: that people from West- There are several low-cost ways to ern, educated, industrialized, rich approach this in the short term: one and democratic (WEIRD) socie- is to select a few judiciously chosen ties — and particularly American populations that provide a ‘tough undergraduates — are some of the test’ of universality in some domain, most psychologically unusual peo- such as societies with limited count- ple on Earth1. ing systems for testing theories So the fact that the vast majority about numerical cognition1,2. of studies use WEIRD participants A crucial longer-term goal is presents a challenge to the under- to establish a set of principles that standing of human psychology researchers can use to distinguish and behaviour. A 2008 survey of variable from universal aspects of the top psychology journals found that 96% of Recent developments in evolutionary psychology. Establishing such principles will subjects were from Western industrialized biology, neuroscience and related fields sug- remain difficult until behavioural scientists countries — which house just 12% of the gest that these differences stem from the way develop interdisciplinary, international research world’s population3. Strange, then, that research in which populations have adapted to diverse networks for long-term studies on diverse articles routinely assume that their results are culturally constructed environments. Ama- populations using an array of methods, from broadly representative, rarely adding even a zonian groups, such as the Piraha, whose experimental techniques and ethnography to cautionary footnote on how far their findings languages do not include numerals above brain-imaging and biomarkers. can be generalized. three, are worse at distinguishing large quan- Recognizing the full extent of human diver- The evidence that basic cognitive and tities digitally than groups using extensive sity does not mean giving up on the quest to motivational processes vary across populations counting systems, but are similar in their abil- understand human nature. To the contrary, has become increasingly difficult to ignore. For ity to approximate quantities. This suggests the this recognition illuminates a journey into example, many studies have shown that Ameri- kind of counting system people grow up with human nature that is more exciting, more cans, Canadians and western Europeans rely on influences how they think about integers1. complex, and ultimately more consequential analytical reasoning strategies — which separate than has previously been suspected objects from their contexts and rely on rules to Costly generalizations Joseph Henrich, Steven J. Heine and Ara explain and predict behaviour — substantially Using study participants from one unusual Norenzayan are in the Department of Psychology, more than non-Westerners. Research also indi- population could have important practical University of British Columbia, Vancouver, British cates that Americans use analytical thinking consequences. For example, economists have Columbia V6T 1Z4, Canada. Joseph Henrich is more than, say, Europeans. By contrast, Asians been developing theories of decision-making also in the Department of Economics. tend to reason holistically, for example by con- incorporating insights from psychology and e-mail: [email protected] sidering people’s behaviour in terms of their social science — such as how to set wages 1. Henrich, J., Heine, S. J. & Norenzayan, A. Behav. Brain Sci. situation1. Yet many long-standing theories of — and examining how these might translate doi:10.1017/S0140525X0999152X (2010). 2. Henrich, J., Heine, S. J. & Norenzayan, A. Behav. Brain Sci. how humans perceive, categorize and remember into policy5. Researchers and policy-makers doi:10.1017/S0140525X10000725 (2010). emphasize the centrality of analytical thought. should recognize that populations vary con- 3. Arnett, J. Am. Psychol. 63, 602–614 (2008). It is a similar story with social behaviour siderably in the extent to which they display 4. Henrich, J. et al. Science 327, 1480–1484 (2010). related to fairness and equality. Here, research- certain biases, patterns and preferences in 5. Foote, C. L., Goette, L. & Meier, S. Policymaking Insights from Behavioral Economics (Federal Reserve Bank of Boston, 2009). ers often use one-shot economic experiments economic decisions, such as those related to 6. Ji, L. J., Zhang, Z. Y. & Guo, T. Y. J. Behav. Decis. Making 21, such as the ultimatum game, in which a player optimism1. Such differences can, for example, 399–413 (2008). 29 © 2010 Macmillan Publishers Limited. All rights reserved