Expanding the interpretive power of psychological science by attending to culture

While “culture” takes on many meanings in psychology, we draw upon Kroeber and Kluckholn’s (3) work to highlight two key features. First, culture consists of explicit and implicit historically derived and selected patterns of behavior. That is, culture is context laden and develops through time and experience. While some outcomes of this experience are visible (e.g., behaviors), others are not (e.g., ideas, assumptions, values). Second, culture encompasses both products of action and conditioning elements of further action. The prevailing values and norms shape cultural products (e.g., societal institutions; individual thoughts and behaviors). When cultural products align with prevailing values and norms, they reinforce these values and norms and strengthen their legitimacy. For instance, laws both reflect cultural beliefs about how people should behave and reinforce these beliefs by punishing behavior that deviates from what is considered “good” or “acceptable.” When cultural products do not align with the prevailing cultural values and norms but are not dismissed as “bad” or “unacceptable,” they can challenge the legitimacy of these values and norms and elicit cultural change. For example, Saudi Arabian activists recently challenged and succeeded in overturning a longstanding law banning women from driving. Such challenges to the status quo cause people within a given culture to question the prevailing cultural values and norms and change the patterns of behavior deemed acceptable in that context. Culture therefore is not a static characteristic that differentiates people (e.g., Easterners vs. Westerners), but a set of dynamic processes that both shape and change in response to cultural products (4–7).

Group characteristics (e.g., geography, race, gender) function as proxies for culture inasmuch as they suggest that people who share certain characteristics likely participate in similar cultural processes. However, cultural influences are not uniform (8, 9); variation exists both within and between individuals and groups of individuals in a given cultural context (10–13). The same individual may demonstrate different psychological and behavioral processes in different contexts, depending on both the prevailing norms and expectations and the group characteristics made salient in those contexts (14–17). For example, Asian American women perform worse in math when reminded of their female identity (negatively stereotyped in math) but better when reminded of their Asian American identity (positively stereotyped in math) (18). Such findings suggest that individuals are inextricably linked to the cultural contexts in which they participate. Fully understanding human behavior necessitates understanding the cultural influences on individuals in a given context.

Although culture plays a critical role in shaping cognitions, emotions, and behaviors across contexts, psychologists often overlook cultural influences on the phenomena they study by failing to consider (i) the people from whom data derive [as McGuire critiqued (1)] and (ii) how psychologists’ own cultural experiences shape their assumptions and practices. While certain subfields (e.g., cultural psychology, racial/ethnic minority psychology) place a more explicit emphasis on culture than do others, the field as a whole lacks agreed-upon mechanisms by which to attend to and report on cultural influences. Published papers, for example, rarely discuss which groups are—and are not—represented among subjects (2, 19–24) and which cultural influences are likely to shape subjects’ experiences. Many researchers report basic sample demographics in the methods section but otherwise pay little attention to cultural influences until the discussion section, where culture—if mentioned at all—is described only as a potential moderating variable with little further theorizing about how culture shapes phenomena (25). This lack of attention to culture likely arises from the historical assumption that, unless otherwise shown, psychological research reflects “fundamental” or “universal” truths about human experiences. While many psychologists now reject this notion, assumptions and practices within the field have been slower to change (25–28).

Psychologists also often overlook how culture shapes their approach to research. Although work in related fields such as sociology and education highlights the importance of reporting on researcher positionality or social location (i.e., identities, values, and experiences in relation to the populations or phenomena studied) (29–32), psychologists rarely offer such information. Outside of cultural and racial/ethnic minority psychology (e.g., refs. 20 and 33), there is little discussion of how researchers’ experiences shape their questions, methods, and interpretations. The underlying assumption is that this reflection is unnecessary because the scientific method is “objective.”

When psychologists ignore variation in psychological processes within and between cultural contexts or rely only on culturally homogenous samples, their theories reflect a narrow—and inaccurate—understanding of human behavior. Improving psychological science involves attending to who is represented in research; how culture shapes subjects’ experiences and researchers’ interpretations; and how, where, and why research is conducted. In other words, improving psychological science involves leveraging interpretive power. In the following sections, we illustrate interpretive power, describe how interpretive power manifests in psychological science, and argue that leveraging interpretive power enhances the understanding of diverse psychological phenomena.

The notion of interpretive power originates in the education literature, where it refers to a teacher’s ability to understand diverse cultural ways of knowing (34–36). Teachers with interpretive power consider the cultural contexts in which students participate and how their experiences in these contexts shape their ways of knowing. These teachers assume that if they do not understand a student’s reasoning or engagement, it may be because the student’s cultural experiences differ from their own (36).

Greenfield et al. (37), for example, described an interaction in which a lack of interpretive power undermined a teacher’s ability to understand a student. When a preschool teacher asked students to describe an egg, a Latina student began telling a story about cooking eggs with her grandmother. The teacher ignored her in favor of students who described the physical properties of eggs. As Greenfield et al. discussed, the student was not wrong; she was trying to convey her understanding based on her cultural way of knowing, in which objects are important because of the social relationships to which these objects are tied. Because the teacher did not recognize the student’s contribution as a reflection of her cultural experiences, she assumed the student did not understand the question and inadvertently gave the student negative feedback.

In contrast, Warren and Rosebery (38) described an interaction in which interpretive power enabled a teacher to recognize a cultural difference, allowing her to connect with and enhance learning for a student from a different cultural background. In the first lesson about plant growth, the class planted seeds in soil to observe how stems and leaves grow. In the next lesson, the teacher showed the class a seed that she germinated in a Petri dish of water. An African American student interrupted, asking, “Did you put magic beans in there or something?” While the question was unexpected, the teacher gave the student the benefit of the doubt. She did not immediately know what the student was conveying but considered whether his cultural experiences may have shaped his engagement. As Warren and Rosebery discussed, African American discourse involves expressive argumentation, metaphors, counterfactual reasoning, and language play. The teacher asked the student to explain his thinking. He responded by asking how seeds can germinate without soil, demonstrating that he was trying to reconcile the necessity of soil for growth in the first lesson with the absence of soil in the subsequent lesson. His response showed that he was in fact deeply engaged in learning. Instead of assuming negative intent on the part of the student and potentially undermining motivation, the teacher used the student’s question to explain the apparent contradiction to the class.

In educational contexts, interpretive power allows teachers to understand how students’ diverse cultural ways of knowing shape students’ contributions. Although the term “interpretive power” is rarely used in psychology, interpretive power plays a similar role in researchers’ ability to understand diverse people and psychological processes. All data include variation. While some variations are random noise, others reflect meaningful differences arising from individuals’ cultural experiences. How psychologists make sense of variation depends upon their interpretive power. When researchers do not attend to culture, they dismiss culturally derived variations as errant and misunderstand the people showing these variations. Without interpretive power, psychologists also risk overlooking variation, even in cases when it may be informative precisely because it is nonnormative (e.g., people who show exceptional resilience; ref. 39). Even with otherwise strong methodological and statistical skills, researchers who lack interpretive power miss opportunities to learn from a wide range of psychological processes.

When researchers cultivate interpretive power, their knowledge of cultural influences becomes a tool that guides their empirical approach and interpretation. They expect psychological processes to differ cross-culturally and include diverse perspectives in their work. However, researchers with interpretive power go beyond simply documenting cross-cultural differences; they use their understanding of how culture shapes cognition, motivation, and emotion to build theories that explain why, how, and when psychological processes manifest differently in diverse cultural contexts. Furthermore, these researchers reflect on how culture shapes their own assumptions and empirical decisions. They recognize that their culturally informed experiences may differ from the subjects’, and they use methodologies that capture psychological processes as they occur for the populations being studied.

As an example, throughout the 1980s and 1990s Western psychology largely viewed self-esteem as beneficial for well-being (40). Cross-cultural research, however, demonstrated that East Asian samples showed lower self-esteem and higher self-criticism than Western samples (10). Without interpretive power, Western researchers might conclude that these differences represent an objective reality and assume East Asians feel badly about themselves. However, as Heine et al. (41) argued, this conclusion is misguided because it overlooks cultural differences in how the self is defined and which factors are important to well-being. While self-esteem is beneficial in Western contexts, where cultural norms encourage differentiation and individuation, it is often less so in Eastern contexts, where cultural norms encourage fitting in and maintaining social harmony (4). When researchers fail to recognize such cultural differences, they often use measures that reflect culturally incongruent psychological experiences (e.g., individualistic measures of self-esteem) and thus generate misunderstandings of entire groups of people (41).

While differences in psychological processes are easier to see when the overarching cultural context (e.g., East vs. West) differs, the consequences of overlooking culture are often greater when differences lie between cultural subgroups, particularly when one group is viewed as normative. The emphasis on self-esteem in US education provides one such example. Because research linked self-esteem with positive life outcomes, educators focused on building students’ self-esteem, assuming these efforts would improve performance, particularly among racial minority and low-income students. However, the research did not demonstrate causality or question the extent to which findings generalized beyond predominantly white, middle-class samples. In fact, later work suggested that many efforts to boost self-esteem did not improve academic outcomes for many students and that positive effects of self-esteem interventions were likely limited to white students (42–44). Thus, in working to improve minority and low-income students’ outcomes, teachers who focused on self-esteem often used ineffective, culturally incongruent methods. At the same time, they inadvertently located the cause of educational disparities within racial minority and low-income students themselves (i.e., their lack of self-esteem) rather than in the systemic disadvantages these students face.

In these examples, a lack of interpretive power undermined psychologists’ understanding of psychological phenomena and led to the application of psychological research with little knowledge of how this work would affect the intended beneficiaries. Had researchers attended to culture, however, these problems could have been prevented. When psychologists leverage interpretive power, they can use cultural differences to build theories that explain a greater range of phenomena with greater nuance. For example, they can describe how social structures, norms, and relationships shape how self-esteem is conceptualized and expressed both within and between populations and what effect, if any, self-esteem has on well-being for different individuals in different contexts. These insights improve our understanding of diverse groups and result in more effective interventions. In the following sections, we discuss three characteristics of psychological science that lacks interpretive power and illustrate how attending to culture renders a more accurate and complete understanding of psychological processes. Unless otherwise specified, we focus on psychological science conducted in WEIRD settings by WEIRD researchers and published in the most selective journals, as this research represents a place in which interpretive power is particularly relevant but often lacking.

Research that lacks interpretive power often recruits culturally narrow samples and draws overgeneralized conclusions about psychological processes based on culturally specific findings. In psychological science, the great majority of research is conducted in WEIRD contexts, with limited attention to the fact that the majority of participants are WEIRD (2). While this work provides a wealth of information about WEIRD psychological processes, psychologists know very little about the extent to which these findings generalize. Due to the assumption of homogeneity of psychological processes across people (i.e., the same behaviors reflect the same processes for all people), psychologists often presuppose that processes documented among WEIRD people are normative and thus do not think to explore variations in these processes among non-WEIRD populations or do not believe such investigations are necessary (2, 20, 27, 45).

For example, for decades psychologists studied attributional processes almost exclusively with WEIRD samples. One of the key theories in this area regards the fundamental attribution error (FAE), the tendency to attribute causes of behavior to individuals’ dispositions and to discount situational influences (46–50). The FAE appeared robust in US samples, but rather than examining whether all cultures exhibit FAE, psychologists went on to document the cognitive processes that give rise to this phenomenon (e.g., refs. 47, 49, 51, and 52) among WEIRD samples, thus implicitly supporting the view that the phenomenon is universal. However, psychology’s understanding of the FAE was revolutionized when research included non-WEIRD samples.

Miller (53) demonstrated that the FAE does not occur to the same extent in all cultural contexts: Middle-class American subjects made more dispositional attributions and fewer situational attributions than middle-class Hindu Indian subjects. Miller hypothesized that differences in cultural meaning systems explained FAE variation: While Western cultures promote an individualistic view of the person, Eastern cultures promote a holistic view that considers both the person and situation. Thus, in the West, dispositional attributions are culturally congruent, while in the East, situational attributions are culturally congruent. Furthermore, Miller demonstrated that even within a given cultural context attributional tendencies vary according to individuals’ exposure to different cultural meaning systems. Lower-class Anglo-Indian subjects, who had more exposure to Western cultural norms, made more dispositional attributions than did lower- or middle-class Hindu Indian subjects, who had less exposure to Western cultural norms. These findings challenged decades of research that largely failed to consider how culture shapes attributional processes. A PsycInfo (www.apa.org/pubs/databases/psycinfo/index.aspx) search for the terms “fundamental attribution error,” “correspondence bias,” or “attribution” reveals 5,120 papers published between 1965 and 1984, when Miller’s work was published. These papers provide a deep knowledge, not of a fundamental human process, but of a culturally specific process. Later research (54–60) replicated and extended Miller’s finding, advancing psychology’s understanding of attributions and of how culture shapes meaning making.

As this example demonstrates, decentering WEIRD ways of being by considering non-WEIRD findings as valid and important in their own right can build psychology’s interpretive power. This practice allows researchers to expand theories by documenting variations in how psychological phenomena manifest in different cultural contexts and explaining why these variations occur. However, given the discrepancy in knowledge of WEIRD compared with non-WEIRD groups, cross-cultural comparisons must be executed with caution. Findings that do not hold across cultures—weighted against the larger WEIRD knowledge base—can perpetuate deficit perspectives of non-WEIRD people (61). For example, Hilliard (33) argued that the misuse of culturally biased IQ tests leads to the inaccurate conclusion that African Americans are less intelligent than whites. Hilliard offered guidelines for understanding how and when cultural comparisons are appropriate in psychological science. Merely making cross-cultural comparisons without a deeper understanding of how culture shapes individuals and how culture guides researchers’ assumptions can perpetuate beliefs about cultural superiority rather than improving psychological theories.

Psychological science that lacks interpretive power normalizes WEIRD ways of being and produces inaccurate conclusions about human functioning (see ref. 62). When research focuses on certain populations while overlooking others, it renders an incomplete understanding of psychological processes on a broad scale. Building interpretive power requires recognizing the need to understand diverse cultural experiences in research to build more comprehensive psychological theories.

A second characteristic of psychological research lacking interpretive power is that it overlooks psychological processes occurring in non-WEIRD contexts and thus renders non-WEIRD people and processes invisible. Psychologists often dismiss non-WEIRD data as outliers and consider research with non-WEIRD populations “applied,” “niche,” or “culturally specific.” For example, there are outlets specific to non-WEIRD populations (e.g., the Journal of Black Psychology) but no WEIRD-specific outlets. Compared with general outlets, non-WEIRD outlets are often regarded as less prestigious and have lower scientific impact. Furthermore, when researchers study non-WEIRD groups to which they belong, their work is often criticized as politically motivated “me-search” or “advocacy,” while WEIRD researchers studying WEIRD populations do not face these criticisms. These norms make it difficult to publish non-WEIRD research and ultimately deter psychologists from studying non-WEIRD processes. Indeed, psychologists studying non-WEIRD populations (e.g., racial/ethnic minorities) argue that psychology’s norms and practices preclude inclusion and understanding of non-WEIRD groups (19–23). When top-tier journals do publish research with non-WEIRD samples, reporting norms often undermine the importance of these findings. For instance, many journals request or even require that research using non-WEIRD samples include a WEIRD comparison group, but non-WEIRD samples are rarely, if ever, requested for research reporting findings from WEIRD samples (63–65). This process continues to center theory and praxis around WEIRD processes.

Our own research offers an example of how one non-WEIRD group—Native Americans—remains invisible in social psychology. While 5.4 million US citizens identify as Native American/American Indian/Alaska Native, alone or in combination with another race (66), Native Americans are vastly underrepresented in the psychological literature. Fryberg and Eason (67) found that less than 0.5% of ∼40,000 papers about prejudice, stereotyping, stigma, or intergroup relations mentioned Native Americans, and only 0.2% included Native American subjects. Although Native Americans face disparate outcomes due to prejudice and discrimination, research on these phenomena has paid virtually no attention to Native Americans. Furthermore, much of the literature on prejudice and discrimination focuses on black–white relations, but these findings do not always generalize to groups such as Native Americans, whose histories and relationships with whites differ (68). Failing to include Native Americans in research not only renders Native Americans and their disparate outcomes invisible but also implies that the black/white dichotomy is representative of all types of bias and/or that the forms of bias and discrimination that Native Americans face are not as important to understand as the forms faced by other groups.

Responses to our research with Native Americans similarly suggest that many psychologists are resistant to research that focuses on non-WEIRD groups. Reviewers at top-tier journals, including the Journal of Personality and Social Psychology and Social Psychology Quarterly, have commented, “What can we learn from Natives?”, “How do we know that Native Americans are simply not less intelligent than whites?”, and “Because this research was conducted on a reservation, it doesn’t involve true random assignment.” The first comment suggests that non-WEIRD research does not contribute to psychology’s understanding of human behavior, while the second illustrates a lack of interpretive power, and the third, a failure to appreciate the value of research in non-WEIRD samples for enhancing our understanding of human behavior (33). While all research involves limitations, such resistance suggests that many members of the scientific community do not believe that research with small or hard-to-reach groups is informative.

Rendering non-WEIRD populations invisible in research also perpetuates these groups’ underrepresentation among researchers. Indeed, in 2016, only 1% of associate professors, 1% of undergraduates, and no full professors or graduate student members of the Society for Personality and Social Psychology identified as American Indian or Alaska Native (69). Seeing that non-WEIRD groups are largely excluded from research may imply that psychology does not value non-WEIRD perspectives and may prevent people from these backgrounds—who are crucial to expanding psychological science (22, 64)—from entering or persisting in the field.

The invisibility of non-WEIRD populations also leaves gaps in the field’s understanding of diverse cultural contexts, processes, and people and limits understanding of human functioning as a whole. Rather than building an understanding of psychological processes that includes WEIRD samples as one of many types of samples, psychology has largely built an understanding of WEIRD psychological processes and behaviors under the guise of understanding general human behavior. While our example focuses on Native Americans, many non-WEIRD groups remain underrepresented in research, including transgender people, people living with disabilities or chronic illness, people who have experienced adversity, veterans, working-class and poor people, and many more. All of these populations have something to contribute to psychological research, and all stand to benefit from a better understanding of the issues they face.

When psychologists overlook how culture shapes key psychological processes, they at best develop ineffective interventions for problems in non-WEIRD contexts. At worst, they implement culturally inappropriate treatments that exacerbate negative outcomes among non-WEIRD populations (20, 25, 45). Indeed, psychologists working internationally caution against exporting Western ideas and methods to non-Western contexts (28, 70–75). Watters’ Crazy Like Us: The Globalisation of the American Psyche (76) provides a vivid illustration of the dangers involved. Watters documented how Western countries dominate the study and treatment of mental illness and export Western treatments to contexts in which people’s understanding of and responses to adversity differ. The result is a rise in mental illness among the intended beneficiaries of treatment. Thus, a lack of interpretive power—particularly the failure to question whether WEIRD findings generalize and how cultural factors shape treatment efficacy—harms people psychology intends to help.

On the other hand, research on education and social class illustrates the importance of attending to culture and including non-WEIRD samples to avoid ineffective or detrimental interventions. Stephens et al. (77) found that well-meaning but culturally incongruent messages from universities undermine working-class students’ academic success. Initial studies demonstrated that universities promote independent motives for attending college, and students who endorse these motives perform better academically (i.e., achieve higher grade-point averages), while students who endorse interdependent motives perform less well. Importantly, working-class students, who often struggle in college, endorse interdependent motives more strongly than middle-class students, suggesting that the discrepancy between working-class students’ motives and university motives contributes to social class disparities in performance. Without interpretive power, researchers might conclude that working-class students’ motives are the problem.

However, Stephens et al. leveraged an understanding of how cultural contexts, specifically culturally incongruent contexts, shape academic performance to understand why social class disparities in performance emerge. In one study, students read a university’s welcome letter that included either independent or interdependent values endorsed by the university. When participants read the independent letter, working-class students performed worse than middle-class students on a subsequent academic task. However, when students read the interdependent letter, working-class students performed as well as middle-class students. By focusing on the relationship between individuals and cultural contexts, the research demonstrated that the problem was not with working-class students’ motives (i.e., a deficit perspective) but was with universities’ failure to acknowledge or legitimate these motives.

Recent work has built on these findings, searching for ways to alleviate non-WEIRD students’ anxiety and underperformance in educational contexts (78–81), arguing for culturally grounded (vs. one-size-fits-all) educational interventions (82), and creating interdependent school cultures to improve non-WEIRD students’ outcomes (83). This work provides a starting point for understanding how cultural factors such as race and social class interact with contextual factors to shape students’ outcomes. Future research can further cultivate interpretive power by exploring how multiple sociocultural factors interact to differentiate educational outcomes both within and between diverse populations. Understanding how culture shapes experiences and behavior and making non-WEIRD populations visible in research—in other words, leveraging interpretive power—allows researchers to create more effective interventions.

As the examples above suggest, including diverse groups in psychological research is a necessary but not sufficient condition to improve psychological science. Greater sampling diversity provides opportunities to learn how culture, social structures, interactions, and individual experiences shape psychological processes. However, using this understanding to improve psychological science requires changing scientific practices and assumptions that undermine attention to culture. In the remainder of this paper, we use the culture-cycle framework (Fig. 1) (5) to illustrate how the scientific community can achieve this change. According to this framework, cultural ideas give rise to the norms, assumptions, and practices that shape processes at all levels of culture. Ideas include beliefs about what is right, good, or moral and offer scripts for acceptable behavior. In psychological science, cultural ideas consist of beliefs about which questions, methodologies, and practices represent “good” science. Institutions reflect and foster these cultural ideas. For example, scientific institutions (e.g., professional organizations, journals, funders, universities) provide professional standards rooted in cultural ideas about what constitutes good science that determine who/what gets funded, published, and tenured (84, 85). These standards set the stage for interactions within the scientific community and between scientific and lay communities. Ideas about which subjects and practices are valuable or normative determine how scientists conduct their research and which people feel welcome to participate (22). Finally, individual psychologists’ beliefs, attitudes, and behaviors are shaped by the ideas, practices, and norms handed down through all levels of the culture cycle. “Successful” psychologists are those who conduct research in ways that align with the dominant cultural ideas and practices.

The culture-cycle framework provides a helpful guide for building psychology’s interpretive power by ensuring that all relevant processes, from intraindividual to interpersonal to institutional/societal, are given sufficient attention. These processes mutually reinforce one another to create sustainable cultural change. A culture-cycle approach also explains why psychology often lacks interpretive power. If, for example, psychological science views WEIRD populations as normative, then, over time, all levels of the culture cycle reinforce this idea and convey that studying non-WEIRD populations confers little benefit. Publications, grants, and tenure decisions favor WEIRD research, which is easier and faster to publish, and the notion that WEIRD findings are generalizable goes unquestioned. These norms create uncertainty about why researchers, especially pretenure, would choose to study non-WEIRD populations. To interrupt this process and create change in psychological science, psychologists can replace problematic acultural ideas with ideas that promote attention to and understanding of culture. Psychologists can also make an effort to understand cultural influences in their research, and scientific institutions can implement policies and practices that support and encourage these efforts. While we cannot speak to all the factors that limit psychology’s interpretive power, the next section illustrates how accounting for culture in research questions, design, and analysis/interpretation renders a more complete understanding of human behavior.

When psychologists leverage interpretive power, they expand and improve psychological science. Building interpretive power requires understanding how culture, experience, and context shape both researchers’ and subjects’ perspectives, experiences, and behaviors. It requires understanding that people are products of their cultural environments just as their thoughts, attitudes, and behaviors shape and reinforce these environments (5). Rather than regarding differences as problematic or dismissing them as noise, psychologists with interpretive power view differences as generative and work to understand their causal influences.

By implementing culture-conscious practices throughout the culture cycle, psychological science can challenge WEIRD-centric ideas and norms that create inaccurate understandings of human behavior. As these practices take hold, psychologists as individuals and as a scientific community will cultivate interpretive power. We will no longer be confined to the questions, theories, samples, and methodologies that have impeded our understanding of diverse people and rendered our theories simultaneously incomplete and overreaching. Instead, we will be empowered to conduct research that better captures psychological processes and behaviors as they occur in many different cultural contexts. By consciously and explicitly attending to culture, we will build a science that better understands all human functioning.

We thank Hazel Rose Markus, Douglas Medin, and Arianne Eason for feedback on early drafts of this paper. This research was supported by Raikes Foundation grants (to S.A.F.).