Abstract
This paper reports my own perceptions of the history of research on culture and cognitive development, in the approximate period 1960 to 2000. I review in particular my own efforts to test Piaget’s theory cross-culturally, but also include other lines of research such as research inspired by Vygotsky and research on child rearing/socialization potentially linked to cognitive development. I briefly mention a research program in Bali, Indonesia, India, and Nepal on the geocentric spatial frame of reference that allowed us to disentangle a variety of eco-cultural and linguistic variables that determine the preference for this geocentric cognitive style. I also recall the “integrated framework” that I proposed in 2003 to combine the models proposed by several authors and which serves to integrate all the findings. The main conclusion is that cognitive processes are universal but that there are cultural differences in cognitive styles and pathways of development. I also discuss why the field has lost momentum, whether it is because all the questions have been answered or because new topics and research methods have evolved.
In the first part of this paper, I recall the early history of cross-cultural research on culture and cognition, in particular cognitive development. I make no claim to be objective nor complete; it is my very subjective take on what interested me during that period, roughly the late 60s to the turn of the millenium. This early research by IACCP members was marked by extensive fieldwork, and collaboration between anthropologists and psychologists. Hence there is a strong link to John Berry’s “the forgotten field(work)” (this issue). Note that much of this research was also done by members of the Society for Cross-Cultural Research (SCCR), the International Association for the Study of Behavioral Development (ISSBD), and the francophone Association Internationale pour la Recherche Interculturelle (ARIC).
To cover just one example of more contemporary research, I spell out briefly in Part 2 the methodology we have used to study the cognitive style of geocentric spatial language and cognition, which could briefly be dubbed “quasi-experimental fieldwork.” Through the choice of appropriate samples within each of three major field locations, we manage to un-confound various eco-cultural variables that are linked to the choice of a geocentric cognitive style.
In Part 3, I conclude with the “integrated framework” (Dasen, 2003) that combines the models proposed by several authors who have inspired my research over the years. It demonstrates how the child grows up in a “developmental niche,” itself immersed in the more macro aspects of culture and ecology. I also attempt a short discussion of what has been achieved and why this line of research has lost momentum.
Part 1: History of Culture and Cognitive Development Research
Not Covered in This Review
Surprisingly, I want to start this review by mentioning some areas that I will skip for various reasons. I mention them first because they would all be relevant if space permitted to cover them more widely.
First there is the research on visual perception, which was the topic of my PhD dissertation at the Australian National University. It has attracted two decades of attention after the publication of Segall et al. (1966) who reported a major study in which anthropologists took to their fields a test-book allowing the measurement of susceptibility to various illusions. Much attention was paid to the methodological issue of non-verbal testing and insuring appropriate communication. This research has been well summarized by Marshall Segall himself (in a chapter in Segall et al., 1999) and by the master of this topic (Deregowski, 1980).
A second area that I will not cover are the studies on memory, although I will mention research by Michael Cole, which includes memory. A competent review is given by Wagner (1981), who has carried out fieldwork in Morocco for over a decade before turning to learning and international development (Wagner, 2018).
Also not covered is cognition studied specifically from an anthropological perspective, that is, the field of cognitive anthropology or more specifically “ethnoscience” (T. Gladwin, E. Hutchins, W. C. Sturtevant). The relationships between anthropology and psychology have been studied from a historical perspective by Jahoda (1982, 1992). I would like to mention anthropologist Wassmann’s (2016) interdisciplinary fieldwork among the Yupno of Papua New Guinea, in which I was invited to participate as psychologist (Dasen, 2017; Wassmann & Dasen, 1994a, 1994b). Wassmann also invited a linguist, a musicologist, and a cognitive scientist. Most anthropologists are usually much more protective of “their” field. We later collaborated on research in Bali, described in Part 2.
The scientific field of this review is cross-cultural psychology. This does not necessarily mean that there has to be a comparison, and especially not a comparison of two groups, one U.S. American and one non-Western. There are appropriate studies in one single cultural group; a profound ethnographic understanding is even the required basis for a good study. Hence I see “cultural psychology” as a part of cross-cultural psychology. This reminds me of the critique Segall (1993) wrote of Shweder’s (1990) claim to have started a new field called cultural psychology: it is, said Segall, like an explorer in Africa “discovering” a lake and naming in “Lake Rudolf,” unknowingly of the fact that it had been Lake Turkana for people living on its shore and fishing in it for centuries. But I will not attempt to cover the relevant cultural psychology theories themselves nor Ernest Boesch’s “Symbolic Action Theory” (Boesch, 1991, 1995; see also Lonner & Hayes, 2007), even though Ernest was a good friend who studied in Geneva with Piaget and later fostered several important colleagues such as Bernd Krewer and Lutz Eckensberger who, as his students in Saarbruecken, carried his mantle and made their own contributions. Colleagues from other places were influenced by him. One of them was Walt Lonner who spent many hours with Ernest trying to understand the machinations of his creative mind and learning why the sound of the violin intrigued him so much.
What I do attempt to cover is the research on “culture and cognitive development” over the past 60 years, with special attention to the early decades (1960–90) in which research with fieldwork has been particularly plentiful. Many IACCP members were involved, but because of the developmental focus, also members of ISSBD with its journal the International Journal of Behavioral Development (IJBD). I remember being part of an ISSBD committee, together with Dan Wagner, on rendering the association more culturally diverse (both in membership and research). In 1986, IJBD featured a special issue on cross-cultural research (Dasen & Jahoda, 1986) and in 1991, it held its biennial meeting in Recife, Brazil, at which cultural issues became a major component. ISSBD is the proud sponsor of regional workshops, of which 21 happened in Asia (1972–2017) and 12 in Africa (1992–2017).
While there was some overlap of membership with IACCP, I must say that since the late 80s I found more colleagues with similar interests to mine in ISSBD than in IACCP. Meeting Bame Nsamenang in Cameroon is one example. I helped him to organize the first ISSBD African workshop in 1992 in Yaounde, sponsored by the Jacobs Foundation located in Switzerland. Nsamenang (1992, 2004) is one of the pioneering African psychologists, together with Robert Serpell, advocating an indigenous or what he calls an Afrocentric approach, recently most clearly expressed in the teacher training manual he edited (Nsamenang & Tchombe, 2011), the proceedings of an IACCP regional workshop held in Cameroon (Tchombe et al., 2013) and a special section of the Journal of Psychology in Africa honoring Bame’s legacy (Serpell, 2019).
The French translation of this manual (Nsamenang et al., 2019) leads me to mention ARIC, which I founded in 1986 with colleagues from Fribourg, Toulouse, Paris, and elsewhere. This association was meant to be multi-disciplinary, although psychology and educational sciences have predominated for a long time (e.g., Retschitzki et al., 1989). Comparative cross-cultural research has always represented only a small fraction within ARIC, centered more on the study of intercultural processes, migration, and most recently, social work. But in the course of my review, some ARIC members may occasionally appear. For example, Bril and Lehalle’s (1988) manual entitled “Is psychological development universal?” has long been the main French reference in the field. Colette Sabatier (1986, 2014) has been an important actor, Jean Retschitzki (1990, 2011) and Bertrand Troadec (1999, 2007) also.
I have organized this part according to two groups of researchers broadly inspired by the work of two great figures in developmental psychology, Jean Piaget and Lev Vygotsky. This will lead to research on the developmental niche, which includes studies of parenting, all of which will lead to the integrated framework of part 3.
Piaget
Cross-cultural research based on Piaget’s theory of cognitive development has undergone an exponential explosion in the 1970s. I am covering it first because I have been one of its main actors. I have reviewed the field several times (Dasen, 1972, 1993, 2011; Dasen & Heron, 1981; Segall et al., 1999) and have edited several collections of papers (Berry & Dasen, 1974; Bevli et al., 1989; Dasen, 1977). There are of course many other excellent reviews such as Greenfield (1976), Keats (1985a), Maynard (2008), Mishra (1997, 2014), and Price-Williams (1981).
Several well-known members of IACCP were involved in these early studies. These include Westerners carrying out studies abroad, but usually living in these places for a fairly long period of time, such as Phillip de Lacey, Molly de Lemos, Jacqueline Goodnow, Alastair Heron, Gustav Jahoda, Daphne Keats, Max Kelly, Sylvia Opper, Douglass Price-Williams, Geoffrey Saxe, Gavin Seagrim, or local researchers working in their own cultural group (but often doing research for a foreign degree) such as Kwabena Adjei, Allieu Kamara, Daniel Kiminyo, O. Michael Okonji, and several others. There were also several researchers in the francophone area such as Magali Bovet, Monique Laurendeau-Bendavid, and Adrien Pinard, Nicola Peluffo. All these studies involve careful fieldwork, interviewing children with Piagetian “tasks” rather than psychometric tests, that is, situations which challenge the child’s reasoning. These have often been adapted to include locally relevant testing materials. It is very time consuming to do so with a reasonable sample size and covering a large age span. Most of the authors carried out single experiments rather than an extensive research program.
From 1969 to 1978 my PhD supervisor Gavin Seagrim (Australian National University) and I published a stenciled “Inventory of Cross-Cultural Piagetian Research” which collated ongoing research projects and pre-prints from a group of about 50 researchers, and led to the publication of the edited volume “Piagetian psychology: cross-cultural contributions” (Dasen, 1977). 1 The three decades of the 1960s to the 1990s provided the majority of the studies, after which this line of research almost disappeared. Several researchers produced so-called “neo-Piagetian” theories (Case, Demetriou, Fischer, Pascual-Leone) but these had only scantly been tested cross-culturally by 1987 (Dasen & Ribaupierre, 1987) and to my knowledge not systematically since then.
Piaget (1966, 1974, 1977) himself gave some lip service to the importance of testing his theory cross-culturally, although his basic assumption was that the developmental processes (accommodation, assimilation, and equilibration), the cognitive processes (at least in the two initial stages of sensory-motor intelligence and concrete operations, i.e., infancy and middle childhood) and the order of appearance of substages would all prove to be universal. (The stage of formal operations will be discussed separately below.) He based this assumption on the theory that the basic structures of interactions of developing individuals with their physical and social environments were similar everywhere. Factors of educational and cultural transmission were acknowledged, but expected to influence only the content of reasoning processes and the rate of development (i.e., the age at which various substages appear), although Piaget seemed to consider such differences as unimportant.
Concrete operations
Overall, Piaget’s prediction was vindicated by empirical studies on the qualitative parts of the theory: The same cognitive processes appear in the same sequence of stages everywhere. There are reports of a few minor departures, such as a temporary substage of pseudo-conservation mentioned by Bovet (1974) in unschooled Algerian children, who give a global conservation answer because they do not, in fact, pay attention to the dimensions of the display. Similarly, Greenfield (1966) found that unschooled children in Senegal did not attain conservation on their own, paying more attention to the authority of the experimenter than to the quantities involved; however, children at the age of European or American children who attain conservation quickly attained the conservation concept when an intervention was carried out in which the child replaced the experimenter in manipulating the conservation materials. Saxe (1981, 1982) found an additional substage in Oksapmin children of Papua New Guinea due to the use of a local body-count system for digits. These are exceptions that only confirm the rule.
On the other hand, much more variation than Piaget seemed to expect was found in the age of attainment of the substages or of the final stage of concrete operations, a finding that clearly shows that he underestimated the social, cultural, and environmental factors. There were also a number of reports of so-called asymptotic development curves, that is, cases where older children and/or adults do not seem to complete the concrete operational stage. If this were true it would be a major blow to the universality of Piaget’s theory.
A good part of early research was concerned with methodological aspects. Even some of the earliest research showed that the familiarity with the objects used in the tasks could be important (Okonji, 1971; Price-Williams, 1962). The effective use of Piagetian tasks is obviously based on appropriate communication, which is a well-known problem in cross-cultural research. Even within a single Western context, it has been shown that task performance is strongly linked to social aspects of the testing situation (Doise et al., 1975; Perret-Clermont, 1980).
My team and I (Dasen, 1982; Dasen, Lavallee et al., 1979; Dasen, Ngini et al., 1979) examined the meaning of the asymptotic development curves with procedures involving training, in which the subjects could manipulate the task materials or, in the case of class inclusion, were asked various questions on the relative size of the subgroups. The studies were carried out among Inuit children in Canada, Baoulé in Côte d’Ivoire and Kikuyu in Kenya, on the conservation of liquids, class inclusion, and horizontality (water levels in tilted bottles). These procedures produced significant training effects with children of various ages, training that generalized to other Piagetian tasks (e.g., from class inclusion to conservation of number and liquids), and basically reduced, sometimes to nil, any “time lag” in the developmental curves observed on first testing. More importantly, the older children (aged 12–14 years) who initially did not perform at the concrete operational level did so after a very short session of training. It must be assumed that these children did have the underlying competence for these concepts but were unable to show it on initial testing (Dasen, 1982).
While I still believe that these findings prove the universality of concrete operations, with hindsight I now find the training studies rather weak. I clearly should have pursued them with a better design, including control groups, and more subjects. Unfortunately, there is not much cross-cultural research on training effects (but see Keats, 1985c).
Another aspect of testing Piaget’s theory relates to the rather vague notion of “structures d’ensemble,” that is, domain consistency, the assumption that children develop through the succession of substages more or less at the same time in different conceptual areas. 2 This has clearly been proven wrong by my research (Dasen, 1975; Segall et al., 1999, p. 163) comparing the developmental curves for the conservation of liquids and horizontality in six groups of children aged 6 to 15 years, selected to contrast former hunting and gathering people (Australian Aborigines and Inuit), an intermediate group of settled fishermen and farmers in Côte d’Ivoire, and two groups of subsistence agriculturalists in West and East Africa. The order of these curves is almost completely reversed when comparing the development of the quantitative to the spatial concepts, with the former showing a quick and complete attainment of conservation in the two farmer groups and of horizontality in two of the formerly nomadic hunting and gathering groups. This finding is of course in line with Berry’s eco-cultural framework: farmers, because they store and exchange part of their crops, value quantitative concepts, nomadic people value spatial concepts because of the importance of way finding.
In a similar line, Posner and Baroody (1979) found that the children of Dioula merchants in Côte d’Ivoire developed conservation of number earlier that those of Baoulé farmers, and there was no effect of schooling in the Dioula group.
Piagetian cross-cultural research would be labeled as “imposed etic” in Berry’s (1989) terms since researchers clearly take a Western theory to the test in various cultures, or at best “derived etic” once the tasks have been locally adapted and shown to make sense.
There have been a few attempts of a more “emic” approach, notably Saxe’s (1981, 1982, 2012b) research among the Oksapmin in Papua New Guinea (PNG). He used the local body-parts counting system in assessing the number concept and locally made string bags for the conservation of length. Saxe’s (2012b) study of social change and mathematical ideas in Papua New Guinea is also one of the few examples of a “return to the field” (together with Greenfield’s (2004) study discussed below). Together with anthropologist Jürg Wassmann, I followed up on body-parts counting systems with a field-study among the Yupno of PNG (Wassmann & Dasen, 1994b), which found a surprising individual variation in the use of the system. However this may be less surprising when considering that counting is always done in public among the Yupno.
Perhaps the earliest example of an emic study is Greenfield and Childs (1977) developmental study of the acquisition of sibling terminology in Tzotzil, a Maya language spoken in Chiapas, Mexico. They compared three theoretical approaches, componential analysis from linguistics, cultural importance from anthropology, and Piaget’s analysis of children’s use of sibling terms in terms of concrete operations. The Piagetian analysis was the only one of the three that could account for the findings.
Maynard and Greenfield (2003) showed that when a test of topological transformation (a concrete operation) was embedded in a context of materials and procedures used in Maya backstrap-loom weaving, it was mastered at an earlier age by members of a weaving community in Chiapas, Mexico (especially by girls who are the weavers of the community), than by children in the United States. However, when the same topological concepts were assessed with materials similar to the ones Piaget had used, children in the United States mastered the same concepts at a younger age than children in the Maya community.
Another interesting line of research is Retschitzki’s (1989, 1990) study using the African board game called Awele in Côte d’Ivoire (Wari, Solo, Mancala elsewhere). This game demands counting skills, advanced planning, and strategies that are akin to formal operations. Retschitzki studied how children learn the game in everyday situations and how some become more expert than others. He also studied the strategies of adult expert players. One of his PdD students from Côte d’Ivoire (N'guessan, 1992) then studied how children in Fribourg, Switzerland, who have no prior knowledge of the game, are able to learn it.
Formal operations
Piaget’s line of interest was “genetic epistemology.” 3 He saw studying child development as a research tool to understand the historical and ontogenetic construction of scientific reasoning. Hence the final stage in adolescence and adulthood is that of so-called formal operations, that is, basically the reasoning in scientific experiments and mathematics.
Piaget’s theory concerning the stage of formal operations has been only partially confirmed, even in the Western, industrialized context. According to Piagetian theory, reasoning at the formal operational stage is no longer tied to the concrete, but can be applied to any, even hypothetical, content. Yet, research has shown that the majority of adolescents and adults in Western societies, even those schooled at the level of high school or university, do not reason in formal operational fashion except under certain conditions and in certain domains (particularly in formal schooling and their professional specialization). This led Piaget (1972) to revise his position by stating that all adults have the capacity for formal operational reasoning, but that this capacity would not reflect itself in performance except under favorable circumstances.
There is still not enough systematic information to confirm the universal applicability of Piaget’s conclusion. Cross-cultural studies that have employed Piagetian tasks in the formal domain have all shown that schooling up to the secondary level is a necessary but not sufficient condition for success at these tasks (e.g., Laurendeau-Bendavid, 1977; Shea, 1985). That is not at all surprising, because these tasks (even when administered in informal situations) relate directly to physics, chemistry, or mathematics as they are taught in school. This does not necessarily mean that cross-cultural studies that employ formal operational tasks are uninteresting or invalid, particularly if subjects are schooled. Keats (1985b), for example, administered six tasks of formal operations to high school students 16 to 17 years of age and university students 20 to 24 years of age in Australia and also among Malays, Indians, and Chinese in Malaysia. Three of the tasks were administered in repeated fashion before and after a proportionality training session. The author arrived at the same conclusion as I did with the training studies: “The effectiveness of the training programme in removing initial differences between the groups suggests that although the initial testing did show differences in performance, these did not reflect differences in competence. . . Evidence of formal operational thinking was found for some subjects within all groups, but not for all subjects within any group” (pp. 316–317).
Perhaps the most productive group of researchers in the area of everyday arithmetic 4 was situated for many years in Recife, Brazil. They designed a large number of studies that can be found summarized in the book entitled Street mathematics and school mathematics (Nunes et al., 1993) and in a chapter on everyday cognition in the Handbook of cross-cultural psychology (Schliemann et al., 1997). These studies demonstrated, in particular, that nearly illiterate adults can use proportional reasoning (considered in Piaget’s theory to require formal operations) within the confines of their professional practice, for example, foremen working with scales on blueprints or fishermen calculating the price and ratio of processed to unprocessed seafood.
According to Tapé (1994) scientific reasoning is an experimental cognitive style valued in Western society, and fostered through formal schooling. In contrast African philosophy, through informal education, favors an experiential style corresponding to a pragmatic, action-oriented logic. As Greenfield (1976) pointed out, one could follow Piaget’s example and first determine what is considered to be the final stage in each society, and then study its particular ontogenesis.
Infancy: Sensory-motor intelligence
Research on the Piagetian theory of sensory-motor intelligence has been facilitated by the development of standardized tests (Casati & Lézine, 1968). The testing nevertheless remains very time consuming and necessitates intensive training of the researchers. This possibly explains the dearth of cross-cultural research. My research team and I started our own research in Africa with a longitudinal study of sensory-motor intelligence among Baoulé babies in Côte d’Ivoire (Dasen et al., 1978). What we found was that the sub-stages that Piaget (1936) had observed in his own three children and that had been confirmed in a larger sample in France (Casati & Lézine, 1968) were essentially the same in rural Côte d’Ivoire.
Other research in urban Côte d’Ivoire (Sièye, 1975) and India (Kopp et al., 1977) came to the same conclusion, which makes sensory-motor intelligence a candidate for a “strong” universal (Dasen, 1981). However we did find some cultural differences as well. Baoulé babies showed a significantly faster development than French norms on some of the tasks, notably the use of an instrument to reach for a distant object. This conclusion applies to the control group of well-nourished babies. Moderate malnutrition had a significant effect by bringing the results in line with French norms (Dasen et al., 1977; Lavallée et al., 1979).
When we carried out behavior observations of daily activities, we observed that babies often had an opportunity to practice this skill; they were allowed to play with whatever was within their reach, including objects that Western mothers would consider as too dangerous such as a knife or a cutlass, and they often used these as instruments. On the other hand, Baoulé babies often showed frustration when their mothers were prevented from reaching for the desired object to give it to them immediately, and some babies even pushed the mother’s arm as if it were a (social) instrument. This is in line with Rabain’s (1979, 1989) observations on child rearing practices in Senegal, and later developmental research emphasizing a relational style of parenting (Greenfield et al., 2003; Keller, 2002).
Conclusion on Piaget’s theory
As a general conclusion on the cross-cultural examination of Piaget’s theory—and discounting the critiques within mainstream developmental psychology—the theory has stood its ground as far as the qualitative aspects are concerned: the construction of knowledge seems to follow the same processes universally, at least at the competence level (including, with limitations, at the level of formal operations). On the quantitative aspects, large cultural variations have been found: those concepts that are culturally valued (because they are found to be useful in the particular ecological setting) develop earlier. We could argue for a “cognitive style” (see below): the basic processes are universal, but the way they are implemented is influenced by culture.
Vygotsky 5
In contrast to Piaget who paid only lip service to educational and cultural transmission, there is another school in developmental research on culture and cognition which draws on the work of Russian psychologists Luria (1976) and Vygotsky (1978), popularized through the influence of Jerome Bruner (1966) and Michael Cole. Vygotsky-related research is currently more influential than Piaget’s, particularly in educational sciences (Schneuwly & Bronckart, 1985), with journals like Jan Valsiner’s Culture and Psychology, Michael Cole’s Quarterly Newsletter of the LCHC (Laboratory of Comparative Human Cognition), biennial conferences of the Society for Socio-Cultural Studies, and influential researchers such as Michael Cole, Jacqueline Goodnow, Patricia Greenfield, Jean Lave, Barbara Rogoff, Sylvia Scribner, Robert Serpell to mention only a few.
While Vygotsky’s theory has not been tested cross-culturally in the same way as Piaget’s theory because it does not come with a set of experimental procedures such as the “Piagetian” tasks, it has become influential in cross-cultural psychology. The study of “everyday cognition” has grown largely out of this socio-historical movement, at the same time as “cognition in context,” “situated cognition,” “socially shared cognition,” or “distributed cognition.” According to these approaches, cognitive processes are not seen as exclusively individual central processors, but they are extended to peripheral tools and are situationally specific. Problem solving can even involve several individuals, because cognition is not necessarily situated “within the head” but is shared among people and settings.
In contrast to Piaget, Vygotsky asserts that there is always sociocultural mediation in individual development, with every function appearing twice : first at the “inter-mental” (or social) level, then at the “intra-mental” (or individually interiorized) level. Hence, knowledge is always a social construction before it becomes individual. It is always contextualized, mediated by symbolic systems. In other words, culture provides tools (writing systems, schools, computers, and so on). “It is culture that provides the tools for organizing and understanding our worlds in communicable ways. . . Learning and thinking are always situated in a cultural setting and always dependent upon the utilization of cultural resources” (Bruner, 1996, pp. 3–4).
Vygotsky’s socio-historical framework adds still another dimension, that of phylogeny and cultural history, developed particularly clearly in Cole's (1996) “cultural psychology.” This is a line of thinking that is not fundamentally different from Piaget’s own interests in the history of science (Piaget, 1970). Hence, while it is easy to emphasize the differences between Vygotsky and Piaget, there are also some obvious overlaps, and according to Tryphon and Vonèche (2001) the contrasts have sometimes been overstated.
For Vygotsky, culture, and especially language, play a key role in cognitive development. The social and cultural contexts, particularly adults interacting with the child, provide whatever is to be learned, through the process of tutoring in the “zone of proximal development” (zoped). This represents the gap between what children can accomplish independently and when they are interacting with adults or more competent peers. Through a process called scaffolding (first coined by Wood et al., 1976), adults provide assistance just slightly beyond the child’s current competence, stimulating development. This idea is close to the role of cognitive conflict in Piaget’s theory: There is an optimal discrepancy between the child’s current knowledge and the new problem that leads to adaptation. If the discrepancy is too small, no learning is necessary; if it is too large, no learning is possible. The difference between the two theories is that for Vygotsky the zoped is provided by adults, while, for Piaget, development is generated by the child’s own exploration of the environment. Indeed, Piaget tended to see adult intervention, such as teaching, as unfortunate because it prevents children from discovering things by themselves.
Experimental anthropology
Among some of the first cross-cultural research leading up to this tradition is that of Michael Cole and colleagues in what they called “experimental anthropology,” which consisted of first observing what people do in everyday life, and then designing a laboratory inspired procedure to test the related cognitive activity. This procedure was then adapted until it was culturally more relevant and performance differences disappeared (Cole et al., 1971; Gay & Cole, 1967). The studies included much attention to memory. A competent review on this topic was given by Wagner (1981), who had carried out fieldwork in Morocco for over a decade before turning to international development (Wagner, 2018).
A similar line of research on specific perceptual skills was taken by Serpell (1979) who demonstrated differences in performance according to the medium in which pattern reproduction tasks were presented: Zambian children excelled with the use of wire models while British children did better with drawings and there was no difference with plasticine models or hand positions. This finding led Serpell to the development of assessment instruments culturally appropriate to the Zambian context, like a “draw a person” test using clay or plasticine (summarized in Matafwali & Serpell, 2014).
The main conclusion of this research trend is “that cultural differences reside more in differences in the situations to which different cultural groups apply their skills than to differences in the skills possessed by the groups in question” (Cole & Bruner, 1971/1974, pp. 244–245). This conclusion is the basis of the “moderate universalist” position, intermediate between absolutism and relativism, defended in our textbooks of cross-cultural psychology (Berry et al., 2002; Segall et al., 1999). It has not changed with more recent research. For example Lee and Johnson-Laird (2006), reviewing Nisbett’s (2003) research, conclude: “No robust evidence exists for cultural differences in the underlying cognitive processes of reasoning” (p. 463).
Some of Cole et al.’s research looked carefully at the respective impact of schooling and literacy (Cole, 1990; Scribner & Cole, 1981). Much of this research is described and summarized in Segall et al. (1999, pp. 120–125). To study the effects of literacy on individual cognitive functioning, and to distinguish them from the effects of schooling, Scribner and Cole (1981) took advantage of a “natural experiment” among the Vai people in Liberia, some of whom are literate in their own script without having been to school. Scribner and Cole compared the performance of Vai men who were literate in the various scripts, as well as that of non-literates, on a variety of psychological tests.
Overall, Scribner and Cole provided no confirmation of Luria’s (1976) and Goody’s (1980) hypothesis that literacy should produce new cognitive processes.
A thorough and critical discussion of IPBS (Institutionalized Public Basic Schooling) as a Western hegemonic imposition and of literacy as a hinge (necessary but not sufficient to open doors) rather than a torch (illuminating the world with new cognitive resources) may be found in Serpell (1993) and Serpell and Hatano (1997). Serpell and Nsamenang (2014) extend this critical outlook to the management of ECCE (Early Childhood Care and Education).
Learning and teaching processes 6
Another important figure in this line of research is Patricia Greenfield, who has contributed careful longitudinal research on learning and teaching processes. Greenfield and Lave (1982) distinguished two types of learning processes: (1) trial and error and (2) scaffolding. In the trial-and-error process, learners are confronted by a new situation constituting a conflict with what they already know. They have to try different approaches, and usually succeed only after making successive adjustments. In this type of learning, motivation is seen as internal to the learner. Conflict and making errors are seen as positive features, leading to progress. According to Piaget, this process should lead to conceptual learning.
In learning by scaffolding the whole problem is presented immediately in its entirety, even if it is too difficult to be managed by the novice alone. The adult expert provides support, furnishing information, and even taking the apprentice’s place when some step appears to be too difficult. The expert’s intervention diminishes in the course of the apprenticeship until it is no longer needed. Scaffolding allows the novice eventually to do alone what at the beginning could be done only with help from the expert. Thus, scaffolding best illustrates Vygotsky’s concept of a “zone of proximal development.”
According to Greenfield and Lave (1982) the prevalence of some learning processes depends on their respective economic implications. In their study of weaving apprenticeship among the Zinacantecan girls in Mexico and of the tailors in Liberia, they found that the masters use scaffolded intervention when errors would cause economic harm. Greenfield (1984) proposed that scaffolding occurs in situations in which the economic stakes are high. When cost doesn’t matter, as in many school-based situations, trial-and-error learning is more common. According to Greenfield, learning by observation and scaffolding are congruent with the maintenance of traditional ways while trial-and-error learning is found when innovation is valued more.
These predictions were confirmed in a long-term longitudinal study when Greenfield (2004) returned to the same Zinacantec Maya community in Mexico 21 years after her initial study of weaving apprenticeship. The girls had become mothers, who in turn had daughters who were learning to weave. But the learning processes had changed substantially in many families. Girls were now often learning much more by themselves, by trial and error, only calling for help when they deemed it necessary. Mothers were often busy with their own work and were not providing any direct scaffolding.
This change in learning processes went along with extensive social change. Woven artifacts were now being sold. Although still based on the traditional patterns, they also included innovative patterns and were now made of commercially produced thread of many different colors. Errorless learning had become less essential, because thread had become cheaper. Greenfield’s major theoretical proposition is that not only do cultures change over historical time, but the very processes of cultural transmission also change; it is a nice illustration of the impact of the socio-political (including economic) context influencing learning processes.
From cognition to social practices
As an outgrowth of Vygotsky’s theory, and in line with “cultural psychology” (see above) the study of “everyday cognition,” originated by Rogoff and Lave (1984), has shifted from an attention to cognitive and learning processes to the study of “social practices.” As Miller (1997) summarizes the position: “One of the striking theoretical shifts in research from a sociocultural perspective is the move away from a focus on psychological processes as internal to individuals to a focus on their embodiment in the context of social interactions and engagement with cultural tools” (p. 92). Together with this theoretical shift a whole set of new concepts have to be addressed. For example, Lave and Wenger (1991) introduced the concept of “legitimate peripheral participation” (LPP) as an approach to the study of learning, particularly in the form of apprenticeship. While this concept is certainly very useful to describe the sociocultural context in which apprenticeship occurs, “cognition” has almost completely disappeared from the analysis; we are left with a description of social experience that is difficult to relate to precise cognitive processes.
Rogoff (1995) distinguishes three processes: participatory appropriation, guided participation, and apprenticeship. These always occur together, but distinguishing them helps to focus on three levels: personal, interpersonal, and institutional. These processes are deemed to be universal, but with wide cultural variations in the styles of interaction with children, and especially in the goals of development (Rogoff, 1990). This is illustrated by several cross-cultural studies. For example, Rogoff et al. (1993) examined how toddlers and their caregivers collaborate in shared activities in four cultural communities, two rural (a Mayan Indian town in Guatemala and a tribal village in India) and two middle-class urban (in the United States and in Turkey). Middle-class mothers implicitly prepare their children for schooling: They structure learning situations, and actively encourage language learning; book-reading and pretend-play are considered important. In the two rural communities, caretakers do not organize special learning situations because children soon manage full participation in the activities of the community. Keen observation through sharing attention among complex ongoing events was more apparent in Guatemala and in India than in the two middle-class communities, in which children are more segregated from adult activities. Rogoff (2014) later coined the concept of LOPI (learning by observing and pitching in) which is described as a “cultural paradigm” (Correa-Chávez et al., 2015) typical of child rearing in native communities throughout the Americas. LOPI is obviously not restricted to this continent. It has been described many times in studies in Africa and in other parts of the majority world.
Socialization and parenting
Note that we have slowly shifted from the psychological study of cognitive processes to the ethnographic observation of social practices. This brings us to a group of researchers who have often been connected to Beatrice and John Whiting of Harvard University. They meet at conferences of the Society for Cross-Cultural Research (SCCR; https://sccr.org) and often publish in the journals Cross-Cultural Research and in Ethos. There is obviously some overlap with IACCP, but SCCR is more obviously multi-disciplinary and sometimes includes couples in which, on the model of the Whitings, one member is an anthropologist and the other a psychologist. Most of them do research with fieldwork but some use the Human Relations Area Files (HRAF) (Barry, 1980) or similar samples of ethnographic studies (e.g., Schlegel & Barry, 1991 on adolescence). Ruth and Lee Munroe are two important contributors to this group; they published many empirical studies, a book (Munroe & Munroe, 1975/1994) as well as an edited handbook on Cross-Cultural Human Development (Munroe et al., 1981).
As mentioned by Berry (this issue), Beatrice and John Whiting are well known for their landmark study of “Children of Six Cultures” (Whiting & Whiting, 1975) where teams of an anthropologist and a psychologist were sent to different locations to carry out observations in a systematic and coordinated way. This was followed by a similar study of adolescence, which gave rise to a series of excellent anthropological monographs (Whiting & Whiting, 1988). In Kenya, the Whitings set up the Child Development Research Unit (CDRU) which later became the Bureau of Educational Research (BER) of the University of Nairobi. Many Harvard connected and local researchers worked through the CDRU, which had set up a network of field sites all over the country with a “standard sample” of children and their families in each and a person of contact. The scheme was devised to enable organizing comparative research in different, well documented socio-cultural settings and easily finding willing comparative samples. Unfortunately, information about this scheme was shared only inside the Whiting group, and so it was not used to its full extent and went into oblivion with the change to the BER. The CDRU had one peri-urban site, Ngecha, where a lot of research was carried out, summarized many years later in an edited book (Edwards & Whiting, 2004).
In 1978 to 1979, Ruth and Lee Munroe carried out comparative field-work in communities of Logoli in western Kenya, Newars in Nepal, Garifuna in Belize and Samoans in American Samoa. The first two of these were more traditional and low income, the other two more westernized, allowing a short-cut method for studying socio-cultural change (globalization). There were 24 children per culture; half of the children were 3 and half were 5 years of age. The study included so-called spot observations of spontaneous behavior, 30 observations per child over a period of 6 weeks, focusing among other aspects on work and play (Munroe et al., 1984). It also included language data (in particular asking questions) and cognitive tests (block building, embedded figures, recall memory, object exploration, motor coordination, gender understanding, and perspective taking).
The large data set led to many publications, more recently with the help of Mary Gauvin. For example in terms of play, Garifuna and Samoan children engaged in more play that involved complex self-managed sequences, defined as child-initiated activities with an exacting series of behaviors (e.g., formal games, role playing), than Newar or Logoli children (Gauvain & Munroe, 2009). A correlation was found between cooking on open fires and cognitive performance; exposure to open fires influenced (negatively) children’s activity level and potential to explore, and learn about, the environment. Children in communities with more industrial elements performed more optimally on all measures, even when years of schooling are controlled (Gauvain & Munroe, 2019).
In a comment on a paper by Gauvain and Munroe (2012) that summarizes the results on modernization and cognitive development, Greenfield (2012) notes that this study could have paid more attention to cognitive strengths developed in subsistence village environments, and that studies of globalization should also pay attention to losses in this process, notably family closeness and interdependence, and respect for elders and tradition. Saxe (2012a) similarly notes that the use of standardized cognitive assessments leads to a “home field advantage” in favor of individuals schooled in industrial societies.
In a recent paper on social change, Gauvain and Munroe (2019) conclude: “The effect of cultural change on children’s development is a bidirectional process, and certain characteristics of childhood may make young people more likely to be early adopters of resources that are introduced during change” (in Summary).
The group of researchers influenced by Whiting and Whiting, while not necessarily studying cognition per se, has provided a wealth of material on parenting and child rearing practices. In particular Harkness and Super (1996) and Super and Harkness (1986, 1997) have developed the important framework of the “developmental niche.” Traditionally, developmental psychology tended to study the child out of sociocultural context, while cultural anthropology emphasized the context per se and the already socialized adult. The synthesis of these two approaches is to make the child in context the unit of analysis.
Concerning its first component, the social and physical settings, Whiting (1980) noted that culture influences child development primarily by serving as “a provider of settings” by furnishing the diverse contexts of daily life (on this point, see also Berry, this issue). For example, in rural Africa, babies are constantly exposed to the daily activities of the extended family by the simple fact that they are always present, and several different people interact with them. This is a very different setting from that of a baby who spends much of its time in a crib or a playpen, with only the mother as a caretaker. The social context formed by the people with whom the child interacts shapes social behavior, norms, and values.
The second component consists of the child rearing customs and orientations, sometimes grouped into “parenting styles.” These vary enormously across societies. For example, different baby-carrying techniques determine the type and amount of bodily contact with the mother, and thus the habitual posture (Bril & Zack, 1989), both of which can influence motor development (Sabatier, 1994).
The third component, the psychology of parents (or of other caretakers), includes the beliefs and values about the development of children, also called parental ethnotheories (Bril et al., 1989). These ethnotheories influence child development, in particular by determining contexts and training practices, but it also happens that parents do not always act according to the normative rules they convey verbally (Bril et al., 1999).
Cultural transmission occurs mainly through the two processes of enculturation and socialization. Segall et al. (1999, pp. 191–192) propose a distinction between incidental and intentional learning processes, the former occurring through enculturation by immersion (observation and imitation), the latter through intentional socialization (with a master, often an adult). The distinction carries the implication that enculturation is often unconscious while socialization is possibly more conscious. Enculturation happens largely through the selection of contexts: Children learn what there is around them to be learned, which is simply part of the cultural setup and is not under the conscious control of parents. Consider the number of books available in a home, which is usually a function of the level of literacy and socioeconomic status of the parents. In more educated households, there are more books around. Some parents purposefully buy “educational” children’s books in which case we would speak of socialization rather than enculturation.
Language acquisition can be analyzed in terms of enculturation (i.e., the selection of a particular context). There is usually no conscious choice of which language(s) are spoken around the baby. While infants are potentially able to learn any language, they become attuned early to pay attention to a particular set of sounds and acoustic distinctions. When they start speaking themselves, they practice only the relevant distinctions, to the point at which the pronunciation of some unusual syllables may become impossible (or at least very hard) to learn. Children who grow up in a plurilingual environment, which is a large proportion of the world’s population, rapidly discriminate between the specific sounds of the languages they hear. Language acquisition and bilingualism constitute an important and complex topic in cross-cultural research. For some cross-cultural coverage, see Keats (1985b), Mohanty and Perregaux (1997), and Altarriba and Basnight-Brown (this issue).
Child care practices are usually very homogeneous within a society: There is one, and only one, proper way of handling a baby, the one way that has been practiced over the generations, without any questions asked (or, in Euro-American society, the one advocated by the currently fashionable pediatrician). These practices, however, vary greatly between societies. For example: “It’s a sensible rule not to take a child into the parents’ bed for any reason” is the advice given by Western pediatricians. Yet, the practice of co-sleeping is routine in most of the world, including in present-day Japan, where children typically sleep with their parents until they are six or more. Putting a young infant to sleep in a separate room was seen by Mayan mothers as “tantamount to child neglect” (Morelli et al., 1992, p. 608).
The developmental niche represents the whole spectrum between enculturation and socialization, and as in the ecocultural framework, these processes are linked to culture, and to ecological and sociopolitical contexts, on the one hand, and determine the child’s behavioral development on the other. Acculturation and social change are processes that strongly influence the system, and genetic transmission is also recognized, particularly in terms of the child’s temperament.
Cultural pathways of development
A further theoretical development is the conceptualization of divergent pathways of development with the corresponding parenting practices (Greenfield et al., 2003; Keller, 2009; Keller & Greenfield, 2000). These pathways correspond to the much theorized opposition between individualism and collectivism (Triandis, 1987), independence versus interdependence or autonomy versus relatedness (Greenfield, 1994; Kağitçibaşi, 1996).
One scenario, common in urban societies consists of exclusive dyadic attention between the mother and the infant, with face-to-face exchange. The mother orients the child from early on to the material world, often mediating the interaction with toys. These experiences foster the development of the self as a causal agent. As mentioned above, the cultural value of early independence is further supported, for example, with independent sleeping arrangements (Morelli et al., 1992). In middle childhood this pathway is marked by active exploration, asking questions, and formal education.
The second pathway, to collectivism or interdependence, is more common in rural environments. It includes the prevalence of body contact in multiple caretaking environments, with child care as a co-occurring activity for the mother or sibling caretakers. The main socialization context is established through bodily proximity, thus promoting warmth and interrelatedness, with a resulting conception of self as co-agent. This developmental pathway continues to socialize skills through observation and participation, finally leading to an interdependent or collectivistic adult.
The authors see these pathways illustrated by a host of observations. Most of the earlier cross-cultural observations fit into this model (e.g., LeVine, 1977). They are, for example, reminiscent of the respective emphasis on technological or social intelligence (Dasen, 1984; Dasen et al., 1985; Mundy-Castle, 1975; Serpell, 1993).
It is of course very restrictive to reduce cultural differences to only two schemes. Keller and Greenfield (2000, p. 56) are themselves, careful to emphasize that “Each path involves different cultural interpretations of the same, maturationally grounded stages.” These paths are not binary opposites but relative emphases and systems of prioritization. They represent life strategies that have proved adaptive in addressing specific environmental problems. Each pathway is a mode around which variation can occur. Also in case of cultural change such as Westernization or cultural contact through migration, the two modes may come into conflict, for example at school (Greenfield & Cocking, 1994), but there may also be intermediate models that develop such as Kağitçibaşi’s (1996) “autonomous-relational self.”
Part 2: Cognitive Styles
In this part of the paper, I would like to turn briefly to one such cognitive style on which I have concentrated my research over two decades since 1994: the so-called “geocentric” frame of reference (FoR). It is in direct line with the research reviewed so far, but provides one example of more recent research. It also provides a methodological reflection on how to use fieldwork with specifically selected samples in order to tease out various eco-socio-cultural variables that influence cognition (Dasen, 2018; Dasen & Mishra, 2010).
It is interesting to note the correspondence between the studies of socialization that lead to the idea of different developmental pathways and the studies of culture and cognition. Both reach the conclusion that there is a basic human universality moderated by cultural styles. Cognitive styles can be defined as “an individual’s preferred and habitual modes of perceiving, remembering, organizing, processing, and representing information” (Dörnyei, 2005, p. 125), or even more generally as “one’s preferred way of processing information and dealing with tasks” (Zhang & Sternberg, 2006, p. 3). In other words, we speak of a cognitive style when a set of cognitive processes are all potentially available, but some are preferentially used rather than others, and when different individuals (or different groups) react differently to a cognitive problem (task, test, experiment, etc.) in some systematic way even though they have the same underlying cognitive capacity or competence. They “choose” to react in this particular way under the influence of a variety of factors such as their age, gender, previous experience, socialization, etc. This “choice” may of course be quite unconscious, and is influenced by many eco-cultural variables.
An important aspect of cognitive styles is that there is no value judgment attached, that is, it is not inherently “better” to choose one style rather than another, except that a particular style may be more adaptive in a particular environment (Berry et al., 1982).
This definition of cognitive styles is larger than “Field dependence/independence (FDI),” the style most studied cross-culturally (Berry, 1976, 1986; Mishra et al., 1996; Witkin & Berry, 1975). It applies very well to the research of Nisbett (2003) on Asian versus Western modes of thought. Dasen and Mishra (2010) in their introductory review of the literature therefore echoed and rephrased the general conclusion of Cole et al. (1971) mentioned above: “Cultural differences in cognition reside more in cognitive styles than in the existence of a process in one cultural group and its absence in another” (pp. 13–14). This is in line with general reviews on culture and cognition like those of Laboratory of Comparative Human Cognition [LCHC] (1983), Mishra (1997, 2001), or Retschitzki (2021).
For our research program on the geocentric frame of reference (FoR), fieldwork was carried out in Bali, Indonesia with anthropologist Jürg Wassmann (Wassmann & Dasen, 1998) and in several locations in India and Nepal with two local colleagues, Ramesh Mishra and Shanta Niraula. This research has led to many publications, including two recent ones (Dasen, 2018; Dasen et al., 2018; Dasen & Mishra, 2010). I will therefore not devote much space to its content and conclusions, but point out how we used a careful selection of subgroups in each location to tease out various eco-cultural variables without ever using large scale cross-cultural comparisons. In other words we have used a succession of quasi-experimental designs to implement the “unconfounding function” of cross-cultural research proposed by Segall et al. (1999, pp. 50–51).
To give an example, in mainstream developmental psychology the variable “chronological age” is usually confounded with “years of schooling” because all normal children go through formal schooling at much the same age. To unconfound these two variables we need to do research in a social setting where some children do not attend school, controlling for other variables such as sex and social status (Dasen et al., 2004). In the Piagetian research reviewed above, it is often found that schooling makes no difference to the development of concrete operations but makes a huge difference to formal operations.
The cognitive style we have been studying is called the geocentric versus egocentric FoR (frames of spatial reference; Levinson, 2003). It refers to the fact that in some societies/languages the location of objects in small scale, table-top space is described by using large scale geographical dimensions such as cardinal points (or upriver/downriver, to the mountain/to the sea). For example when telling a child how to set the table, a mother may say “Put the knife North of the plate and the fork South” instead of what seems more natural in English, using right and left. Because right and left are relative to the speaker’s body, this is termed an egocentric or relative FoR.
In our first small study in Bali (Wassmann & Dasen, 1998), we found that small children age 4 used the geocentric mode quite consistently, with the egocentric mode then increasing slightly with age. This was a surprising finding, because in Piagetian theory, Euclidean space (geocentric) is considered to be more difficult than egocentric space, which is more directly linked to the human body. However while the local language, Balinese, uses its own geocentric FoR consistently (based on mountain/sea and sunrise/sunset, to put it briefly), teaching is done in Bahasa Indonesian which uses either the egocentric FoR or cardinal directions. We therefore could not tell whether or not the increase in egocentric answers with age was due to schooling. Our sample was small and all the children went to school.
In India and Nepal, we were able to carry out studies with large samples of schooled and unschooled children. There was basically no difference between the two, but the use of the geocentric frame increased with age in all samples. Young children in rural areas tended to use deictics (“it is there” while pointing) and then rapidly the geocentric FoR, city children would use the egocentric FoR more (Dasen et al., 2004; Mishra et al., 2003). In Varanasi, India, pupils of a Sanskrit school used the geocentric FoR much more consistently than pupils taught in Hindi. Sanskrit rituals use a cardinal point system with eight named directions in the plane, so that geocentric references are highly valued (Vajpayee et al., 2008).
In a second study in Bali with larger samples (Dasen & Wassmann, 2008), we found the same urban/rural difference we had found in India. While all children were schooled, we were able to assess the impact of Bahasa Indonesian vs. Balinese because of social stratification, some families using the former more, some the latter. The knowledge and use of the geocentric FoR occurred very early in rural children with a preference for Balinese in the home while those preferring Bahasa Indonesian took longer to learn it and used more egocentric references first.
We tried to replicate the impact of a language with a preference for an egocentric system by comparing children of two schools in Kathmandu, Nepal, one where teaching was in Nepali (geocentric, cardinal directions) and one in English (egocentric). Surprisingly there was a slight preference for the geocentric FoR in the English language school, which puzzled us, until we found out that the teachers were using the geocentric references (cardinal points) even in English.
In Geneva, absolutely no use of the geocentric FoR was found in francophone and various bilingual children (Dasen & Wassmann, 2008). We could have concluded to the absence of this cognitive process in Geneva children. Nevertheless there are indications that they could learn to use a geocentric FoR. Troadec (2003) and Troadec et al. (2002) found that children’s response could be influenced by using suggestions systematically in the instructions. A number of experiments (e.g., Li & Abarbanell, 2018) show that the flexibility in the use of one FoR or the other depends very much on the specific testing conditions. We therefore maintain the claim that the absence of this process in Geneva is the sign of a cognitive style rather than a deficit in competence.
The above is a very succinct summary, which leaves aside several aspects of the study. What I mean to illustrate here is how we were able to untangle the set of eco-cultural variables by a succession of comparisons within each of our larger cultural groups. This is a research strategy that needs a long-term project with carefully selected samples, an option that needs ethnographic (and in our case, linguistic) documentation before psychological methods are used (Wassmann & Dasen, 1994a). In other words, it is not “safari style” research in which a “cultural difference” is reported by comparing two randomly selected groups, usually a Western and a non-Western one. It is, I would claim without false modesty, an example of how comparative cross-cultural research should be done.
Part 3: Integrated Theoretical Framework, Discussion, and Conclusion
Part 3.1: Integrated Theoretical Framework
Over the years of my research and teaching career in both anthropology of education and cross-cultural psychology, I encountered a number of theoretical frameworks up to the point where I felt a need for integration. Figure 1 presents this integrated theoretical framework (Dasen, 2003).
Dasen’s integrated theoretical framework.
The center represents the individual, particularly the developing child. The concentric circles around it are based on Bronfenbrenner’s (1989) ecological systems theory. The outer circle represents the macro- or exo-system, that is, the broader ecological and socio-political contexts which are linked to the individual through the transmission processes of the meso-system. This idea of linking population-level variables to individual-level ones is basic to the eco-cultural framework of Berry (1976) which is close to Whiting’s (1977) psychocultural model. Georgas’s (1988, 1993) ecological and social cross-cultural model of the family adds the idea of interactions between all the concentric circles.
The individual grows up in a developmental niche (Super & Harkness, 1986, 1997) composed of three subsystems: (1) the physical and social settings; (2) child rearing practices; and (3) parents’ ideas, social representations, or ethnotheories (which correspond to the more general value and belief systems of cosmology and religion in the macrosystem). In the developing child itself, Bril (1999) drew attention to the field of promoted actions for the construction of cultural competences, in particular learning processes. An important methodological issue, particularly in cross-cultural research, is the fact that in psychology, the characteristics we are interested in (such as personality or cognitive competence) always have to be inferred from observable behavior. And performance does not necessarily reflect competence (Cole & Bruner, 1971/1974). This methodological feature is also explicitly present in Berry’s ecological framework.
While the theories just mentioned have provided the building blocks for my integrated framework, further perspectives that are akin are for example Sinha’s (1977, 1982) ecological map of the child, the cultural-ecological perspective of Ogbu (1981), Trommsdorff’s (1989) cross-cultural studies of socialization and the value of children, Kağitçibaşi’s (1996) model of family change, and the conceptualization of context as an incorporating system rather than external stimulation (e.g., Goodnow et al., 1995; Serpell, 1999). I believe the framework is appropriate to all cross-cultural research, certainly in developmental psychology. Its general idea is that the developing child grows up and the adult lives in interaction with a microsystem formed by their immediate “niche” in which the physical and social settings determine what is to be learned in interaction with societal values and specific practices, this open system being itself in constant reciprocal interaction with the wider ecological and cultural macrosystem.
Through the above summary of research, it should have become clear how the research of IACCP members together with adherents to other associations such as ARIC, ISSBD, and SCCR has led to a coherent framework on the topic of culture and cognitive development, reflected in Figure 1. Research on individual (“internal”) cognitive processes makes sense and should not be abandoned, but only insofar as the individual is set at the center of a developmental niche, itself surrounded by larger socio-cultural variables at the macrosystem level. Also the immediate observable behavior (performance) should not be taken as a genuine indicator of underlying competence without severe methodological checks.
The integrated framework also draws attention to the importance of studying learning processes. It is thus that the study of “cultural pathways through universal development” (Greenfield et al., 2003) is a promising choice. It has been illustrated by one such pathway through the very early achievement of a geocentric frame of spatial reference (FoR) in some Balinese, Indian and Nepali children (Dasen & Mishra, 2010).
Part 3.2: Discussion and Conclusion
While I have not attempted to cover research on culture and cognitive development in the new millenium, there would actually not be all that much to report. The topic was hot in the 70s to 90s but then slowly lost momentum. Is this because we in fact have all the answers and no more research is needed? It would no doubt be presumptuous to say that is true, but at least it is partly true. As my summary of the research shows, the very fact that it may all be summarized in a nice and coherent integrated framework points to the conclusion that a large progress has been made. We do know a lot more than we did in 1970!
The fact that cross-cultural Piagetian research is no longer current is also true of mainstream developmental psychology, but as I have shown, many other lines of (cross-)cultural research had emerged by the year 2000, and the same for the mainstream, which had changed to a “life span” perspective (i.e., including development during adulthood and old age) and included new topics such as theory of mind (a topic also taken up by cross-cultural research: e.g., Wang et al., 2021; Wassmann et al., 2013. But see an early mention of the topic by Greenfield & Bruner, 1966).
And we seemed to have made some impact on the mainstream. In Super and Harkness (1999) wrote: “It is no longer reasonable to think that ‘child development’ can be understood by studying children in only one cultural setting; what was the paradigm of development is now only a single cultural instance. Understanding the variable cultural construction of child development is now part of the mission for developmental science” (p. 306).
More recently, the same authors (Harkness & Super, 2020a) ask “Culture and human development: Where did it go? and where is it going?” They remark that even at the Harvard Graduate School of Education—famous in our field for the leadership of Beatrice and John Whiting, Robert LeVine, and so many others—culture and human development is simply no longer in existence. They attribute this mainly to the ideology of postmodernism, which has greatly afflicted anthropology, including psychological anthropology. Harkness and Super however remain quite optimistic about the future of our enterprise. They write “We see a potentially bright future for the field, if we make sure to include the heritage of past research as well as developing new collaborative international initiatives” (p. 30) and they conclude: The way forward now, we suggest, is international and intercultural collaboration of scientists. The challenge for training new researchers from diverse backgrounds, however, is to equip them with the knowledge and insights gained from cross-cultural psychology, psychological anthropology, and their own cultures, rather than simply making the next generation of scholars into new representatives of Western theories of development. (in the Summary)
Harkness and Super (2020a) also mention a very encouraging list of interesting recent developments, including the Oxford Handbook of Human Development and Culture: An interdisciplinary perspective, edited by Jensen (2016) of Clark University. This has evolved into a mainstream developmental psychology textbook which does take culture seriously (Jensen & Arnett, 2020) but I do not know to what extent this is used widely. Jensen (2012) previously provided an interesting and optimistic review of the field, and so did Sabatier (2014), according to whom it continues to be an active field searching for new dimensions, such as theories of mind. She concludes by writing “finally, the cultural approach is vital for our understanding of child development in its fundamental dimensions” (p. 238).
My own published take (e.g., Dasen, 2019) is somewhat more pessimistic. I am not so sure that mainstream developmental psychology is really taking culture seriously. It seems to have evolved toward experimental neuropsychology and cognitive science, that is, “hard” science, in which—mistakenly, I believe—cultural variations are not expected, certainly not studied, and seen more as a nuisance than a center of interest. In many psychology departments, in Europe at least, developmental psychology has become a secondary subject or has disappeared altogether, including at the University of Geneva. A (cross-) cultural approach to the field has rarely been included structurally in psychology departments, whether in Europe or North America; usually the field disappears after the retirement of the professor who was able to establish it.
Even within cross-cultural psychology, the topics and methods have shifted. There is very little developmental psychology, and most of the research reported at more recent IACCP congresses is carried out over the internet, contacting colleagues in many countries to collaborate in research often using their students as subjects. The days are gone when one would do real “fieldwork” (see Berry, this issue).
Personally I have always enjoyed doing fieldwork, first alone, then with an interdisciplinary team in Côte d’Ivoire, and later, since 1992, in Papua New Guinea and in Bali with an anthropologist, Jürg Wassmann from the University of Heidelberg. A chapter describes the advantages but also problems in our interactions (Dasen, 2017). In India (and Nepal), I was able to team up with a local psychologist, Ramesh Mishra, with whom I could plan and carry out the research project at an equal level of scholarship (Dasen & Mishra, 2010). In Nepal he was able to enlist one of his former PhD students, Shanta Niraula. But it has to be admitted that for a Western psychologist to travel afar may be seen as part of a (neo-)colonial tradition, like some missionary or development worker. It is no longer politically (and environmentally) correct!
Another argument for Westerners doing less fieldwork may be that it is no longer needed, since there are competent researchers almost everywhere to work in their own groups. However we do not see that many publications coming from them are on culture and cognitive development. This may be due to the various limitations they encounter in their academic environments. In the late 90s, Soledad Perez and I organized a series of workshops on educational research in majority countries, and in particular on the problems encountered by the researchers. This led to a UNESCO publication (Pérez & Dasen, 1999). I am not sure things have changed much since then.
Among the problems encountered, difficult access to publication outlets was number one. Most research is done to obtain academic degrees, and remains unpublished; there are very few students who choose to continue in a research career, because there are few positions and funding is difficult to access. Also, research is not always rewarded by university administrations. At the Benares Hindu University, one of the top institutions in India, Ramesh Mishra had to take every day we spent together on fieldwork on his holidays. The argument of the vice-chancellor was that a professor was paid to teach, not to do research, and anyway, publications would lead to improve the CV, which would encourage brain drain.
Another reason for the fewer publications may be that doing “pure” science has become less attractive or acceptable. The purpose of most research projects described above was to gather knowledge without immediate relevance to the populations studied. The research projects of course always entail some small rewards, but I have stopped promising that the research results would better the educational system, as I had done previously. I actually never saw any applications, which led me to argue that the researcher should include these in the project itself (Dasen, 1988)—but I actually never managed to do that. For many years I had frequent interactions with anthropologist Gasché (2004) who was working on adapting the school system to indigenous people in Peru. He was of the opinion that research had to be done for those people with those people. My research he saw as “extractive,” taking away knowledge for the sole use of the academic community, if not for the sole profit of publications and status. Of course he was quite right, but I still strongly believe that fundamental research is of value and should be continued.
There is of course a strong demand for research that is geared more to immediate social applications. Recent publications from Africa show this trend (e.g., Matafwali & Serpell, 2014; Nsamenang & Tchombe, 2011; Serpell, 2011, 2019; Serpell & Nsamenang, 2014). Wagner (1993, 2018) shows the relevance of developmental research for international educational planning, in particular how working with international agencies such as UNESCO can help to improve Early Childhood Education and Care, basic education and literacy training. Harkness and Super (2021) analyze why understanding culture is essential for supporting children and families. In ARIC we often had discussions between researchers and practitioners (e.g., Perregaux et al., 2008), and ARIC is, generally speaking, more geared toward application and social issues than IACCP or SCCR.
I find it difficult to draw an overall conclusion to this stream of reminiscences. All I can say is that doing cross-cultural research on cognitive development was fun, and that I ended up satisfied with the answers. Seeing the large number of “reads” of my papers and books on Research Gate and Academia, I am surprised that even old publications still attract quite a bit of interest. I take it as a sign that it was not only fun but also useful.
Footnotes
Acknowledgements
I wish to thank the following colleagues and friends for comments on a draft of this paper: The three editors of this special issue, Debbie Best, John Berry, and Walt Lonner, an anonymous reviewer, Patricia Greenfield, Heidi Keller, Robert Serpell, Charles Super, Jürg Wassmann, and Dan Wagner. In this historical review I have obviously overstated my own contributions and left out many important names; may I be forgiven.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
