Human behavior and thought often exhibit a familiar pattern of withingroup similarity and between group difference. Many of these patternsare attributed to cultural differences. For much of the history of itsinvestigation into behavior and thought, however, cognitive sciencehas been disproportionately focused on uncovering and explaining themore universal features of human minds—or the universal featuresof minds in general.
This entry charts out the ways in which this has changed over recentdecades. It sketches the motivation behind the cultural turn incognitive science, and situates some of its central findings withrespect to the questions that animate it and the debates that it hasinspired. Woven throughout the entry are examples of how the cognitivescience of culture, and especially its elevated concern with differentforms of diversity and variation, continues to influence and beinfluenced by philosophers.
One cluster of philosophical work falls within the traditional subjectmatter of philosophy of science, in this case of the cognitive andsocial sciences. Philosophers have analyzed and assessed the methodsand evidence central to the scientific study of cognition and culture,and have offered conceptual scrutiny, clarification, and synthesis.Research in a second vein sees philosophers themselves contributingmore directly to cognitive scientific projects, (co)constructingtheories, helping build computational models, even gathering empiricaldata. A third kind of work is naturalistic philosophy or philosophy ofnature, wherein philosophers seek to use results from the cognitivescience of culture to inform or transform debates over long-standingphilosophical questions, including questions about the nature ofphilosophy and philosophical methodology itself.
Each of the conjoined expressions in the title of this entry marks outenormous terrain, and neither has a single, strict, canonical, oruncontroversial definition. Philosophically interesting issues andlocal debates relevant to each will be discussed in more detailthroughout the entry, but those discussions will be easier to graspwith some initial, rough but orienting characterizations in hand.
Cognitive science is an interdisciplinary approach to studying minds,mental organization, and intelligent behavior (Thagard 1997 [2019]).The field has developed over the last 70 years, incorporating researchfrom across more traditional disciplines like psychology, philosophy,linguistics, logic, computer science, anthropology, economics,evolutionary theory, and neuroscience (though see Núñezet al. 2019 for doubts about the extent to which these have succeededin uniting into a distinctive field; cf. Stanford 2020). As thesedisciplines converged on the same types of questions from differentdirections, a loose cluster of shared concepts, methodologicalassumptions, experimental paradigms, and theoretical goals emerged.Together, the elements of this cluster have served as a productiveframework for constructing theories and guiding ongoing research aboutbehavior, minds, and mentality (Pylyshyn 1984; Lycan & Prinz 2008;Brook 2009; Samuels, Margolis, & Stich 2012).
At the heart of the framework is the idea that minds and mentalactivity can be understood not as any particular kind of substance,but rather as some form of information processing. The currently mostintuitive and probably still dominant version of the idea is expressedwith the computer analogy, though some have argued that antecedentsand earlier versions of the core insight can be found as far back asAristotle’s doctrine of hylomorphism and its distinction betweenmatter and form (Shields 1991; Nussbaum & Putnam 1992; cf. Nelson1990; Olshewsky 1992). The computer analogy holds that the relationbetween the physical and mental aspects of a person, between bodiesand minds, between the physical and the mental more generally, can beunderstood on the model of the relation between hardware and software(Putnam 1967). On this view, mental states are functional statesidentified by the role they play in the “program” thatchannels information through and guides the control system of aphysical body. Mental processes, then, are understood as computationalprocesses performed on mental representations, symbols that have bothsyntactic and semantic properties, and that are usually but notnecessarily realized in brains. Mental processes and representationscan be modeled using the formal resources of computation theory, withsyntactic relations between the mental representations encodingpatterns of causal relations that hold between the physical statesthat realize them (Fodor 1975; Stich 1983; Rescorla 2015 [2020]). Inthe last several decades, aspects of this orthodoxy have beenchallenged, especially regarding the existence and character of mentalrepresentations (Pitt 2000 [2020]). Several alternatives to andextensions of this “classical computationalist” way ofunderstanding information processing have made inroads into cognitivescience, and have been discussed by philosophers under the headings ofconnectionism (Buckner & Garson 2019), embodied cognition (Clark1997a; Shapiro 2011; R. Wilson & Foglia 2011 [2021]), enactivism(Hutto & Myin 2013; Di Paolo & Thompson 2014), dynamic systemsapproaches (Chemero 2009; M. Anderson 2014), complexity science(Favela 2020), and predictive processing (Hohwy 2013; Clark 2016).
Still, the core idea of mental activity as information processing ofsome sort remains compelling, in part because the differences betweenthe classical computationalist approach and these newer alternativesmay not be very deep (Cao 2020), but also due to the clear researchagenda it suggests. The general aim of cognitive science can be castas the attempt to understand minds and cognition by trying tounderstand the nature and limits of computation and other forms ofinformation processing, and by trying to determine the character andprovenance of the programs our brains, and the brains of non-humananimals, typically run. It also suggests a suite of more specificquestions: How malleable are the different programs? How much of theinformation and processing structure is innately specified? How muchand which elements are acquired via various forms of learning over thecourse of development, and throughout different stages of a lifetime?What is the character of the information processing that underliesdifferent psychological capacities such as vision and visualconsciousness, language acquisition and comprehension, individual andsocial learning, emotion and affect, memory and imagination, skilledbehavior and deliberation, decision-making and moral judgment, and soforth?
These broad questions and the general conceptual framework that iscommon to them give the different disciplines that make up cognitivescience a base camp and a lingua franca with which to pursue differenthypotheses about minds, mental organization, and intelligent behavior.Together they aim to provide a coherent way to bring together andcompare different kinds of theoretical, formal, and empirical work tobear on a core set of questions.
While ‘cognitive science’ refers to anapproachto studying a particular set of phenomena,‘culture’ would appear to refer directly to a type ofphenomenon, rather than a way of investigating it. This much may seemuncontroversial, but little else about culture (or‘culture’) is (Risjord 2012; Lenard 2020). Before itbecame central to the social sciences and humanities, the term had alife of its own outside of academia, where it was primarily used torefer to the training of mental capacities in humans. In this sense,‘culture’ was roughly synonymous with‘education’: a cultured individual was an educated person,and the qualities of educated people, and the things that educatedpeople produced and consumed, were construed as cultural (Jahoda2012). Vernacular usage has retained many of these connotations, andsome of the more technical uses of ‘culture’ remainindebted to these folk origins. Still, most contemporary academicunderstandings of the subject matter have been greatly influenced bythe work of eighteenth-century philosopher Johann Gottfried Herder,whose approach to culture emphasized how human thinking and behaviorare extremely flexible, and steered by environmental variation andsocial learning (Denby 2005).
During the first half of the twentieth century, historical influenceslike these informed numerous conceptions of culture, with differencesbetween them often reflecting the different theoretical work thenotion was being asked to do. By the middle of the century, Kroeberand Kluckhohn were able to identify more than 150 definitions, makingexplicit numerous views about what was required of the concept, andthe demands placed on the it by different fields and theorists(Kroeber & Kluckhorn 1952). A more recent overview (Baldwin et al.2006) argues that definitions of ‘culture’ currently inuse can be placed into one of seven genres based on what they take astheir central theme: structure/pattern, function, process, product,refinement, power/ideology, or group membership. The fact thatcognitive scientists are mostly concerned with patterns, processes,and functions is reflected in this entry, but it is worth noting thatthe other themes tend to attract more attention from adjacent academicdisciplines. For example, political philosophers and criticaltheorists examining postmodernism and postcolonialism often analyzepower relations in terms of ‘dominant cultures’ (Gemignani& Peña 2007), while those doing aesthetics oftenconcentrate on material artifacts that count as ‘culturalproducts’ and part of a ‘cultural heritage’(Hammersley 2019). In sociology and much of cultural anthropology, theemphasis is often on non-material cultural products such as beliefs,values, and norms, and on how these serve to structure human socialarrangements (Sewell 2005). Edward Tylor, the founding father ofcultural anthropology, famously articulated a holistic variation onthis theme, describing culture as
that complex whole which includes knowledge, belief, art, morals, law,custom, and any other capacities acquired by man as a member ofsociety. (Tylor 1871: 1)
The diversity of the theoretical aims and approaches across academicdisciplines interested in culture, along with the inherent complexityof those phenomena identified as cultural, have also inspiredpessimistic conclusions. Some have argued against the possibility of aunified conception of culture, suggesting no single definition couldsatisfy all of the demands placed on the term and still be coherent ortheoretically useful (Byrne, Barnard, et al. 2004; Segall 1984).Others have argued that the lack of such a single univocal definitionwould be no hinderance to scientific research, since no singleidealization of the idea of culture will be equally useful for allexplanatory purposes (Sperber & Claidière 2008). That said,many contend that in order to clarify what culture does and does notdo, refined conceptual tools are needed to think clearly about itsdifferent manifestations and facets, and especially to distinguish itfrom related but importantly distinct phenomena such as ecologicalinheritance (Odling-Smee & Laland 2011) and epigenetics (Jahoda1984; Jablonka & Lamb 2005). In light of this, Ramsey’srecent explication (2013) of a conception of culture that is bothcurrently influential and that has a venerable history (Cavalli-Sforza& Feldman 1981; R. Boyd & Richerson 1985) is well-suited toserve as a touchstone for this entry:
Culture is information transmitted between individuals or groups,where this information flows through and brings about the reproductionof, and a lasting change in, the behavioral trait. (Ramsey 2013:466)
Cognitive science research using alternative conceptions of culturewill be noted below (see especiallySection 2.2).
Information counts as cultural not by virtue of its content but ratherby where it comes from, how it travels, and what it does. Culture istransmitted via behavior, and produces durable, non-ephemeral effectson behavior. Although it is not explicitly mentioned in Ramsey’sexplication,variation in behavior, between individuals butespecially between groups, has been diagnostically important for thoseseeking to integrate culture and cultural explanations into cognitivescience. Many have sought to document patterns of behavioraldiversity, and more recently patterns of psychological diversity, andshow that they cannot be accounted for by genetic or ecologicaldifferences, but rather are the result ofculturaldiversity.
A landmark paper published in 2010 consolidated and raised thevisibility of a cluster of issues surrounding the relationship betweenculture and cognitive science (J. Henrich, Heine, & Norenzayan2010). It laid out the case that the vast majority of research inpsychology used subjects who came from a very specific demographic;most participants in psychological studies had been WEIRD, drawn fromsocieties that are Western, Educated, Industrialized, Rich,Democratic. Other premises of the argument pointed out that not onlywas this a very narrowly circumscribed sample population, it was also(a) an exotic outlier group, especially from an historicalperspective, that was nevertheless (b) typically treated byresearchers as representative of the species as a whole. Results fromstudies done with WEIRD participants (and typicallyby WEIRDcognitive scientists) were commonly extrapolated, taken to be true ofhuman beings in general, and thus to be illuminating universalfeatures of human psychological nature. The negative conclusion wasthat this practice, widespread though it may be, is unjustified, andthat distressingly little was known about the range of variation inhuman cognition, motivation, and behavior, or about the mechanismsthat create and support it. This conclusion also implied that as faraswas known, culture may have a much deeper and moreprofound role in shaping human mentality than cognitive science hadyet appreciated.
The paper was not the first to recognize many of the problems andchallenges it raised, but the acronym it introduced gave arecognizable name to a movement that had been coalescing for some time(Shweder 1990, 2003; Markus & Kitayama 1991; Nisbett & Cohen1996; Rozin 2001; Nisbett 2003; Arnett 2008; also see Broesch et al.2020 for reflections about the entanglement of methodological andethical factors in cross-cultural research, and misgivings that theacronym WEIRD invites misleading binary thinking about culturalvariation). The importance of culture for understanding human mindsand behaviors had long been central to some disciplines nowcontributing to cognitive science, especially anthropology andcross-cultural psychology (Tylor 1871; Malinowksi 1931; Levi-Strauss1958; Triandis & Brislin 1984; Hofstede 1980; Greenfield 2000).Culture had also become integral to the application of biologicalthought to humans and human social behavior (Dawkins 1976, 1982; Tooby& Cosmides 1992; Laland, Odling-Smee, & Feldman 2000;Griffiths 2007) and had been central to the field of culturalevolution from its inception (Cavalli-Sforza & Feldman 1981; R.Boyd & Richerson 1985; even Darwin 1871).
The paper helped consolidate and focus attention on a host ofshortcomings that contemporary cognitive science is still coming toterms with (Apicella, Norenzayan, & Henrich 2020). Many aremethodological. Cognitive scientists concerned about culture and WEIRDsampling bias continue to debate the ideal role of theory informulating hypotheses, interpreting results, and investigating theinteraction between minds and cultural influence in a number of otherways (Muthukrishna & Henrich 2019; Muthukrishna, Bell, et al.2020; Barrett 2020; Rozin 2020; Sear 2020). Others seek to betterintegrate historical evidence into the study of psychology (Turchin2015, 2018; J. Henrich 2020; Muthukrishna, Henrich, & Slingerland2021) and to clarify the best way to study the ontogenetic sources ofvariation in development (Stotz 2010; Heyes 2018a; Amir &McAuliffe 2020). Still others connect the discussion of WEIRD samplingbias to psychology’s recent difficulties with replication(Fidler & Wilcox 2018), arguing that while having moreexperimental participants and increased statistical power is, all elseequal, a virtue, it is not a panacea, especially if those participantsare all WEIRD. Rather, they urge that some questions about behavioraland psychological diversity can be more thoroughly addressed by usingethnographic methods from anthropologyalongsidepsychological experiments. In some cases, important insights abouthuman variation can be won by using these combined methods to study asmall but carefully chosen cultural group (N. Henrich & J. Henrich2007; K. Smith & Apicella 2020; Zefferman & Mathew 2020). Inother cases, however, experimental approaches common to mainstreamcognitive science may be unsuitable for such work, because thedefinition and operationalization of central constructs used toinvestigate phenomena associated with, e.g., intelligence orself-esteem may be difficult to apply, inhibiting rather thanfacilitating genuine cross-cultural insight (Greenfield 1997; Joe2014; J. Henrich 2020).
Appreciation of the blind spots and questionable inferenceshighlighted by the WEIRD challenge has contributed to substantivetheorizing about minds as well. It has informed work focused onestablishing the existence and character of psychological variation,but also made urgent the need to develop a more sophisticated range oftheoretical resources toexplain that variation. Questionshere have often been framed using contemporary versions of thedistinction between innate and acquired traits (Carruthers et al.2005, 2006, 2007). While theses about the presence and significance ofinnate mental characteristics have loomed large in the history ofWestern thought (Samet 2008 [2019]; Markie & Folescu 2021), thedistinction and its attendant concepts are not philosophicallyunproblematic (Griffiths 2002; Mallon & Weinberg 2006; Griffiths& Machery 2008; Gross & Rey 2012). Nevertheless, much researchhas sought to illuminate which aspects of mental structure and content(and even personality, see Gurven et al. 2013; Smaldino, Lukaszewski,et al. 2019) might be part of a genetically shared, (quasi-)universalhuman psychological nature, and which aspects can vary between groups,and how much (Roughley 2021). Further efforts have been directed atdiscovering, of those aspects that are acquired and that do exhibitvariation, which are obtained via individual learning, and which areculturally ‘inherited’, acquired via (a particular typeof) social learning. This framing has in turn brought increasedscrutiny to the psychology of learning itself, and especially thestructure and provenance of those cognitive mechanisms that underliedifferent forms of social transmission. Research on this front will bethe focus ofSection 3.2 andSection 3.3.
Many attempts to establish the influence of culture in shapingbehavioral and psychological diversity take the form of arguments fromexclusion, aimed at ruling out other candidate explanations that mightaccount for the specific pattern of intergroup variation in question.Competing explanations can take numerous forms. For example, until the1950s many instances of behavioral variation were thought to beattributable to genetic differences between human groups (Barkan1992). Research in the life sciences over the last 70 years, however,failed to produce evidence to support such explanations, findinginstead that biological differences between human populations areminor and superficial (Hirschman 2004). The loss of confidence inpurely genetic explanations inspired researchers to continue exploringalternatives.
Alternatively, many social sciences deal in“structuralist” explanations that account for behavior byappeal to factors exogenous to individuals, such as norms, laws,institutions, economic policies, and government types. These wouldexplain patterns of intergroup variation in behavior by appeal todifferences between the structures that organize each group. Sincestructural factors of this sort are passed along socially rather thangenetically, however, they fall comfortably under the broad workingdefinition of ‘culture’ provided above, and so can becounted as a genre of cultural explanation (Haslanger 2016; Davidson& Kelly 2020; Soon 2021).
Most of the alternative explanations that arenot cultural inthe sense at issue here—and so which donot give acentral role to social transmission—attempt to explainintergroup variation by stressing the importance of differences in theexternal physical or ecological environments inhabited by differentgroups, or by appeal to the operation of psychological mechanisms likecausal reasoning or individual trial-and-error learning (Richerson& Boyd 2005). The range of options can be elucidated by examiningevolutionary approaches to human behavior, and comparing the distinctway each approach appeals to such factors in their proprietary modesof explanation (E. Smith 2000). For example, human behavioral ecologyremains largely agnostic, or at least assigns little significance to,the character of the internal psychological mechanisms that might leadto group level behavioral diversity. Instead, behavioral ecologistsappear to simply assume (perhaps unrealistically, see Driscoll 2009)that humans are able to produce behavior that optimizes fitness acrossa wide range of environments, and by implication also assume thatindividual human minds are generally flexible enough to calibratebehavioral strategies to best fit whatever ecological circumstancesthey find themselves in (Grafen 1984). On this view, patterns ofbehavior within a group typically reflect commonalities in theecological circumstances faced by most members of the group, anddifferences in the patterns found between groups reflect, fairlydirectly, differences in the ecological circumstances those groupsinhabit. Such explanations make little mention of mental structure andno mention of culture or cultural inheritance.
By contrast, practitioners of the research program that has becomeknown as Evolutionary Psychology are much more concerned with mentalstructure and evolved psychological mechanisms, and readily discusssome forms of cultural variation (Barkow, Cosmides, & Tooby 1992;also see Hagen 2015). Evolutionary Psychologist’s approach toexplaining human behavioral variation builds on the more generalnotion of phenotypic plasticity. The approach emphasizes thatphenotypic plasticity (of the behavioral phenotype, but also ofmorphological, physiological, and other dimensions of phenotypicexpression) is found not just in humans, but in almost all organisms.Mechanisms that underlie such plasticity allow individuals withsimilar genotypes to express different phenotypes, and thus toflexibly adapt to different environments in a variety of ways. Forinstance, researchers have appealed to this kind of interactionbetween environmental differences, on the one hand, and mechanismsunderlying phenotypic plasticity, on the other, to explain systematicpatterns of variation in the clutch sizes of members of the samespecies of bird found in different geographical regions (Tooby &Cosmides 1992). Evolutionary Psychologists offer explanations with asimilar form for many instances of human variation, such as regionaldifferences in the distribution of values. One such account holds thatthe prevalence in some groups of certain kinds of values andnorms—collectivist or exclusivist—can be explained as anadaptive response evoked by the salience of cues indicating highpathogen threat in the group’s environment. Alternatively, theprevalence in other groups of contrasting values andnorms—individualistic or inclusivist—are explained by theabsence of such cues, and thus an environment that allows theexpression of a different adaptive phenotype (Schaller & Park2011; also see Buchanan & Powell 2016; cf. Gelfand et al. 2011;House et al. 2020).
With respect to the psychological sources of human behavioralplasticity and cultural diversity, Evolutionary Psychologists tend toconstrue human minds as “massively modular” (though theconception of a “module” involved in this kind of claim istypically more liberal than the one made famous by Fodor 1983; seeSamuels 1998, 2000; Sperber 2001; Barrett & Kurzban 2006;Carruthers 2006; Robbins 2009 [2017]). Human minds share a number ofdistinct, genetically inherited dispositions, each shaped by theselective pressures generated by the adaptive problem it evolved tohandle. Each disposition can produce a circumscribed range ofbehaviors to deal with its associated adaptive challenge, but not allthe behaviors made possible by those dispositions are manifested byevery human or every group. Rather, patterns of behavior are shapedand constrained by the character of mental modules, but are alsoexplained in part by appeal to external circumstances, with differentenvironments acting to select orevoke different subsets ofbehaviors from the range already “contained in” and madeavailable by these universally shared psychological dispositions(compare PlatoMeno). Proponents point out that on this view,evoked behaviors can persist even as the environment that originallyevoked them changes or disappears. For example, once a behavioralphenotype is initially expressed it can be effectively locked in by aprocess of developmental canalization (D. Wilson 2009; Witherington& Lickliter 2016). Proponents also stress that the view does notimply that all behaviors and differences between groups are adaptive.For example, the claim that the comparatively higher incidence of teenpregnancies in lower income groups is an adaptive response evoked byconditions of poverty remains a matter of controversy and empiricalinvestigation (Johns, Dickins, & Clegg 2011).
These “evoked culture” explanations associated withEvolutionary Psychology are often contrasted with the“transmitted culture” explanations associated with dualinheritance theory (Norenzayan 2006). While both are designed toaccount for patterns of behavioral variation between groups, theformer do not involve the kind of information conveyed betweenindividuals central to the explication given inSection 1.2. For this reason, the psychological modules from which behaviors areevoked in the explanations given by Evolutionary Psychologists havebeen classified by other theorists as “non-culturalmechanisms” (Mathew & Perreault 2015). Alternatively,transmitted culture explanations point to a different source ofphenotypic plasticity, namely cultural inheritance and thepsychological mechanisms that underpin social transmission. A thirdapproach can be seen as combining elements of both of these (Sperber1996; Morin 2011 [2016]). Advocates of cultural epidemiology, which isoften presented as being distinct from both dual inheritance theoryand Evolutionary Psychology, typically attempt to explain instances ofcultural diversity and stability in terms of how different evolved,domain-specific modules influence the acquisition, transmission,reconstruction, and retention of specific mental representations.
These three forms of explanation are distinct in many ways, but arenot necessarily in competition with each other (Sterelny 2017). Inmany cases advocates of the different approaches are concerned withdifferent kinds of behavior: few Evolutionary Psychologists haveattempted to account for the accumulation of material technology orscientific knowledge. Dual inheritance theorists tend to focus onforms of cooperation, social organization, and culturally complexadaptions while eschewing topics like sexual behavior, matepreferences, and phobias (although there are exceptions; see Dugatkin1996; Faucher & Blanchette 2011; Adriaens & De Block 2006).Neither pay as much attention to the character and dynamics ofnon-adaptive traditions centered on, e.g., recipes, songs, or folktales as cultural epidemiologists typically do (Acerbi, Kendal, &Tehrani 2017). Advocates of the three types of explanation typicallyaccept that each is able to account for some instances of behavioralvariation but not others. Moreover, the boundaries between theseapproaches are vague and overlapping at best, and much good work onevolution and culture combines ideas from more than one. Nevertheless,advocates of different approaches occasionally disagree about the bestway to conceptualize culture (Sperber & Claidière 2008),about the relative importance of social learning and transmittedculture in general (Richerson & Boyd 2008), about whether and howculture should enter into ultimate explanations of behavior(Scott-Phillips, Dickins, & West 2011), and about the bestexplanation of specific cases, such as different features of prosocialreligions (Bulbulia 2004; Norenzayan et al. 2016) and morality(Sripada & Stich 2007; Buchanan & Powell 2018; Brownstein& Kelly 2019).
Philosophers have joined these debates. Many have challengedEvolutionary Psychology’s evoked culture explanations byattempting to show that alternative explanations better capturevarious target phenomena, or by arguing that the kinds of behaviorsthat Evolutionary Psychologists focus on lead them to underestimatethe variation exhibited by our species and the plasticity required toaccount for it (Dupré 2001; Buller 2005; Richardson 2007;Downes 2015). By contrast, philosophers have tended to think more‘with the science’ of cultural epidemiology (Nichols 2002;De Cruz & De Smedt 2015) and dual inheritance theory (Skyrms 2010;Sterelny 2012; Davis & Kelly 2018). Philosophical advocates ofthese approaches see them as more promising in virtue of theconceptual and methodological sophistication they find in them,together with the models they have developed to clarify the kinds ofinteractions that might hold between genes, transmitted culture, andcognition (R. Boyd & Richerson 2005; Gualaderex 2012; also seePaul 2015). As a result, philosophical discussions of culturalepidemiology and dual inheritance theory are often lessstraightforwardly critical. Instead, many aim at drawing outinteresting philosophical implications (Boudry, Blancke, &Pigliucci 2015; Kelly & Hoburg 2017), shoring up conceptualfoundations by offering refinements and amendments (Lewens 2015;Buskell 2019); or emphasizing the need for more elaborated methods toempirically test their hypotheses (Sterelny 2020).
Humans, and some non-human animals, are natural social learners. Acrucial form of social learning is enabled by theirculturalcapacities, typically understood as capacities that allow them toacquire and transmit cultural information. The number, character, andcontribution of the different psychological mechanisms that underliethese cultural capacities remain a matter of much debate withincognitive science, as does the issue of which ones are required toproduce and sustain not just culture but complex cultural traits andcumulative culture. This section describes research thatinvestigates these capacities, the methodological and evidentialreasoning involved, and some of the conceptual issues that itraises.
Important caveat: for the purposes of this section, capacities andmechanisms are classified as “cultural” in virtue of theirfunctionality, what theydo. Broadly speaking, they count ascultural if they support the spread of cultural information. This isimportantly distinct from saying they are cultural in virtue of theirprovenance. Some of the psychological mechanisms that enablethe acquisition and transmission of cultural information—such asthose that underpin mindreading or normative cognition—maythemselves be the products of cultural rather than biologicalevolution, and thus be mostly culturally inherited, e.g., cognitive‘gadgets’ in Heyes’ terminology (Heyes 2018a; Birch& Heyes 2021; also seeSection 3.3). Other mechanisms that enable the acquisition and transmission ofcultural information may be genetically inherited, e.g., cognitive“instincts.” Both would count as contributing to culturalcapacities in the sense at issue here.
Theorists of culture see great significance in a distinction betweenmere culture, on one hand, and cumulative culture, on the other. Thelatter involves not just the spread of information between individualsvia social transmission, but the wherewithal of later generations toretain, refine, add to, and innovate on the cultural repertoire theyinherit from previous generations. Some have claimed that cumulativeculture is distinctly human and evolutionarily unique (Sterelny 2009,Tomasello 2009) while others have suggested that rudimentary instancesof cumulative culture can also be found in some non-human animals likewhales (Whitehead & Rendell 2015), Japanese monkeys (Schofield etal. 2018), and chimpanzees (Marshall-Pescini & Whiten 2008). Thereis little controversy, however, about the claim that the scale andcomplexity of human culture outstrips anything found in other animals(Dean et al. 2014). While some have expressed doubts about whether thecurrent evidence suffices to establish that most human cultural traitsresult from the accumulation of modifications over time (Vaesen &Houkes 2021), most researchers hold that understanding our capacitiesfor cumulative culture, perhaps more so than those that underlieindividual intelligence, creativity, or problem solving, is crucial tounderstanding the distinctive character and success of the humanspecies (Richerson & Boyd 2005; Sterelny 2009, 2012; J. Henrich2016; Derex & Boyd 2015; R. Boyd 2017; cf. Pinker 2010).
There remains disagreement on how to characterize the relevant notionof complexity, however. As Querbes, Vaesen, and Houkes (2014) note,increase in cultural complexity has been characterized in severalways, including as an increase in a trait’s fitness, as anincrease in the number of elements that make up a trait, and as anincrease in the difficulty of transmitting the trait faithfully (seealso Mesoudi & Thornton 2018) or how difficult it is to learn (J.Henrich 2004). These characterizations are not theoretically orempirically equivalent, and each suffers from its own shortcomings.Specifying complexity in terms of fitness, for instance, seems torequire an agreed upon metric for cultural fitness, but no such metrichas been widely accepted (Ramsey & De Block 2017). Relatedly,while it is relatively easy to recognize episodes of technologicalchange that are instances of progress, it is far less clear how toevaluate changes in other kinds of cultural traits, and so it remainsdifficult to know when transformations in, say, a religious orlinguistic system should be classified as improvements or degradations(Miton & Charbonneau 2018).
Attempts to characterize cultural complexity by appeal to number ofelements confront an individuation problem, and it remains unclear howto determine what counts either as a distinct element or a singletrait. A similar individuation problem emerged in the debate overcultural replication. Dawkins famously coined the term‘meme’ to refer to discrete bits of culture that play arole in cultural evolution similar to that played by genes inbiological evolution, arguing that memes are independent replicatorswhose propagation serves their own interests (Dawkins 1976).Subsequent research revealed, however, that it is very difficult todetermine how a bit of cultural information might qualify as singlememe (Chvaja 2020), and many scholars are skeptical that culturalvariants are usefully thought of as being discrete or primarilyserving their own interests (Hull 1982, 1988; Heyes & Plotkin1989; R. Boyd & Richerson 2000; Claidière &André 2012; De Block & Ramsey 2016). Consequently, memeticshas largely disappeared from academic discussion (though see Sterelny2006 and Dennett 2017 for exceptions).
What has not entirely waned, however, are debates over the nature ofcultural acquisition, whether and when the spread of culture isusefully thought of in terms of replication, and if not why not. Someproponents of replication hold that most stable cultural traditionsare primarily the result of imitation, and that imitation drives akind of faithful, high fidelity copying of cultural variants thatshould qualify as replication (Tomasello 1999). Dual inheritancetheorists argue that the type of high fidelity copying suggested bythe term ‘replication’ may or may not be rare, but it isnot required for complex cultural adaptations to evolve. Rather,cumulative cultural evolution can be sustained by low fidelityimitation because the influence of many teachers and models cancompensate for noise and copying errors (J. Henrich & Boyd 2002).Cultural epidemiologists also reject the need for replication, arguinginstead that many traditions are mainly the result of ‘guidedvariation’. This is a phenomenon whereby learners do not simplycopy what they are exposed to, but rather creatively reconstructrepresentations from what they observe, slightly but significantlyaltering the cultural variants they acquire in the process ofacquiring them. In these cases, the recipient of the cultural signalis not a passive vessel, but actively filters or even transforms therepresentations that they adopt. Moreover, cultural epidemiologistshold that these processes of reconstructive learning aresystematically influenced by stable features of human cognitivearchitecture, which can be represented as ‘attractors’ inthe abstract possibility space of cultural evolutionary models(Sperber 1996; Claidière & Sperber 2007; Claidière,Scott-Phillips, & Sperber 2014; Lewens 2015; Morin 2011 [2016];Buskell 2017; Scott-Phillips 2017). For example, a proponent of‘pure’ replication might hold that the pronunciation ofwords remains relatively stable because individuals tend to copy withhigh fidelity the pronunciation of their peers and parents. Aproponent of the cultural epidemiology approach could offer adifferent explanation, claiming rather that the pronunciation of wordsevolves in the direction of better fit with the mechanisms ofverbalization, comprehension, and retention, and then stabilizesaround these attractors (for similarly competing accounts of some ofthe common features of religion, see Boyer 2001; Atran 2002;Norenzayan et al. 2016).
As just noted, this debate about cultural transmission and thecomplexity and character of cultural variants is entangled withdebates about different kinds of human universals, how common or rarethey are, and how they should be explained (Brown 1991, 2004; Singh etal 2021; also see J. Henrich, Blasi, et al. forthcoming). For example,if features of a shared human psychological nature are implicated inreconstructive social learning, and those features, in turn, helpstabilize certain types of group level behavioral regularities byfunneling social transmission towards cultural attractors, then thosetypes of group level behavioral regularities are likely to be commonto many cultures (Sperber 1996; De Cruz 2006). If, on the contrary,high-fidelity copying is the rule in social learning, then smallmodifications in initial conditions, minor copying errors, and slightperturbations in the flow of cultural information have a better chanceof being propagated, and to develop into more and more dramaticdifferences between cultures (Acerbi & Mesoudi 2015; but seeTchernichovski, Eisenberg-Edidin, & Jarvis 2021 for how imitationcan sustain high ingroup variability).
To a large extent, many of these are empirical issues that turn on therelative importance of the respective transmission processes indifferent episodes of social learning, as well as the relative effectof each process on cultural stability. However, there may beconceptual issues involved as well (Lewens 2015), concerning whetheror not the type of cognitive reconstruction of cultural variantsposited to occur in guided variation is necessarily at odds withreplication, or is perhaps a process that sometimes produceshigh-fidelity copies, but other times does not (Boudry 2018). Suchlocal discussions are part of the larger debate over how to bestconceptualize, study, and understand the significance of cumulativeculture, cultural complexity, and the psychology and populationstructure (e.g., Derex & Mesoudi 2020) required to create andsustain it.
While social learning is a central concern to those focused onculture, other forms of learning are also important for understandingthe range of interactions between cognition, culture and itsaccumulation, and perhaps the evolution of cultural capacitiesthemselves. For instance, while social learning appears to be suitedto contribute to the rapidspread of new behaviors andbeliefs in a population, individual learning may be better suited tocontribute to theinitial generation of new behaviors, novelbeliefs, and to the production ofinnovations on existingcultural variants. Theorists have carved out different positions onhow the distinction between individual and social learning should beconceived, and how these two facets of cultural capacities mightinteract.
Cognitive scientists have traditionally been interested in learning,broadly defined as the acquisition by individuals of new information,knowledge, and skills. There is also an established history ofresearch on related cognitive phenomena such as attention, perceptionand memory, much of which focuses on identifying functions and causalmechanisms, and on developing methods to optimize learning and memory(De Houwer & Hughes 2020). Historically speaking, the distinctionbetween social and individual learning does not seem to have beengiven much theoretical significance. Those participating in cognitivescience’s recent cultural turn accord it greater importance.
Efforts to articulate a precise and theoretically useful version ofthat distinction have run into complications, however, and have notalways yielded intuitive results. For example, an episode of learningcan be plausibly classified as social if a conspecific facilitates theacquisition of new information (Heyes 1994, 2012). Individuallearning, on the other hand, is often construed as acquiring newinformation directly from the environment. But these specificationsare not mutually exclusive, and together they entail that manyepisodes of human learning qualify asboth individual andsocial. The environments that our species inhabits are oftenthemselves engineered, and are thus to a large extent shaped by otherhumans, often in ways specifically designed to facilitate learning(Sterelny 2012; J. Henrich 2016; Lewens 2017). As many have noted,this is an important form of (epistemic) niche construction andecological inheritance. It bears important similarities to behaviorsfound in a range of other species whose members live in environmentsthat are deeply shaped by other members (and sometimes previousgenerations) of their own species (Laland, Odling-Smee, & Feldman2000; Kendal, Tehrani, & Odling-Smee 2011; Aaby & Ramseyforthcoming). In response to such difficulties, others have urged thatindividual learning should be distinguished from social learning inthat only episodes of the latter involve social transmission,understood as happening when “the behaviour of an individual[comes] to resemble the behaviour of others through socialinteraction” (Heyes 2021: 228; also see R. Boyd, Richerson,& Henrich 2011).
Aside from the question of how specific episodes of learning should beclassified, there seems to be something close to consensus aboutfunctionality. Broadly construed, the primary adaptive function ofindividual learning is to generate information that is useful fordealing with (often novel features of) one’s environment. Theadaptive value of social learning, on the other hand, comes fromavoiding the costs of individual learning, allowing an individual toacquire useful information from others without having to go throughthe trouble of figuring it out on their own. This consensus, however,gives rise to a puzzle whose solution remains a matter of controversy(Rogers 1988). What has become known as Rogers’ paradox centerson the question of how capacities for social learning could haveevolved. The difficulty arises from the thought that pure sociallearners would be unable to help increase the mean fitness of thepopulations they invade or in which they live, because they would beunable to contribute any innovations or useful new information. Theywould just imitate others, copying what is already there. From thisperspective, social learners are free-riders, exploiting the knowledgeand skills created by the epistemic work of individual learners whileproviding nothing novel or beneficial in return (Laland 2004).
Nevertheless, social learning is undeniably found in a range ofspecies, with humans exhibiting especially sophisticated forms ofsocial transmission. Some proposed solutions to Rogers’ paradoxhave emphasized that capacities for social learning can be adaptive(and thus evolve and spread) if social learning is selective.Individuals can use mixed strategies, strategically switching betweenindividual and social learning in different circumstances (Enquist,Eriksson, & Ghirlanda 2007). They can also be selective about whothey learn from, when, and in what domains of knowledge (J. Henrich& McElreath 2003; Buttelman et al. 2013; Koenig & Sabbagh2013). Others hold that capacities for social learning likely evolvedfrom capacities for individual learning, in a process of descent withmodification that was driven by gene-culture coevolutionary processesand the kind of epistemic niche construction mentioned above (Sterelny2021: chapter 1; Kendal, Boogert, et al. 2018; also see Heyes 2018aand Taylor, et al 2021).
Some researchers have expressed surprise that social learning isthought to be so crucial for understanding what is distinctive abouthumans, especially in light of its apparent ubiquity in the animal(and even botanical) world (Hill 2010). Similar concerns have led sometheorists to distinguish what they call ‘culturallearning’, holding it to be a form of social learning thatessentially involves fairly sophisticated forms of cognition likeperspective-taking or intersubjectivity (Tomasello, Kruger, &Ratner 1993, also see Hoppit & Laland 2013; Legare & Nielsen2015; J. Henrich 2016). The term “social transmission”remains useful, however, as it is neutral on this score. It is alsointuitively inclusive of psychological capacities for learning butalso for teaching, both of which support the flow of culturalinformation between individuals and generations (also see Heyes &Moore forthcoming).
Some of the most fruitful research into the psychology of socialtransmission has been comparative, investigating the similarities inand differences between the cultural capacities of humans and thosefound a wide range of other animals (Andrews 2020b: chapter 8). Muchof this research is structured by efforts to isolate and understand“difference makers” (Buskell forthcoming; Charbonneau2020), those ecological, social, and especially psychological factorsthat are distinctive of humans and responsible for our unique culturalsophistication. Here too a central debate involves the issue ofcumulative culture and the kind of psychological machineryrequired not just to socially acquire and transmit information, butfor it to accumulate over time, enabling each generation to innovateand build on the advances inherited from previous generations (Dean etal. 2014; Mesoudi & Thorton 2018; cf. Querbes, Vaesen, &Houkes 2014; Haidle & Schlaudt 2020; Vale et al. 2021). It hasbeen known for some time that other animals have the wherewithal tosustain group-specific behavioral traditions (Buxton 1948; Imanishi1957; Whiten, Goodall, et al. 1999; Laland & Galef 2009; foroverviews of more recent work see Whitehead & Rendell 2015; Aplin2019; Whiten 2019). However, it remains unclear whether any of thosetraditions accrue complexity and adaptive functionality in the sameway their human counterparts do, or what additional culturalcapacities would be necessary or sufficient to enable groups ofnon-human animals to collectively achieve this “racheteffect” of accumulation (Tomasello 2009, 2014; Tennie , Call,& Tomasello 2009).
Much of this comparative work is also guided by hypotheses about thekinds of evolutionary pressures that selected for uniquely humancultural capacities and drove the dramatic expansion of relative brainsize in the human lineage. For instance, one framing (Baimel et al.2021; also see Muthukrishna, Doebeli, et al. 2018) pits this family of“cultural brain” hypotheses that focus on the adaptivechallenges associated with social learning and handling culturalinformation against “Machiavellian intelligence”hypotheses (Byrne & Whiten 1988 and Whiten & Byrne 1997).These latter hypotheses see key selective pressures as stemming notfrom culture but from more competitive social dynamics that arise fromantagonistic scheming among conspecifics (cf. Downes 2008 [2021];Lewens 2008 [2020]). Another view holds that many of our particularcultural capacities are the product of cultural rather than biologicalevolution, cognitive gadgets that are assembled and passed along aspart of a cultural inheritance rather than a genetic one (Heyes2018a). On this account, what is unique about human brains is thatthey are innately equipped with a set of cognitive mechanisms forassociative learning that are hypertrophied compared to other species,but are not genetically specialized for any particular domain orcontent. These general learning mechanisms, together with some numberof attentional biases, are often used to acquire all manner ofculturally inherited, domain-specific packages of skills, e.g., toplay chess, to drive a car, to read and write, etc. These may alsoinclude skills that allow individuals to imitate others more adeptly,thus allowing them to learn to become better and more selective sociallearners (also see Dennett 2017).
As this and above discussions suggest, much effort has been directedat distinguishing different grades of social learning and the kinds ofcognitive sophistication required to engage in each. The disagreementsbetween the participants seem to be mainly about the number and natureof genetically transmitted learning mechanisms, and how canalized thedevelopment of different types of social learning is. Of the differentmechanisms, imitation, understood as the ability of an individual toobserve and then copy another’s body movements, has attractedthe most attention (Zentall 2006; Whiten et al. 2009). A key debateconcerns whether imitation in particular is underpinned by geneticallyinherited, domain specific psychological mechanisms, or whether it issubserved by domain general associative learning mechanisms that arethemselves further sculpted by cultural influences (Heyes 2016a, 2021;cf. Fessler & Machery 2012). A related debate concerns thesignificance ofover-imitation, the tendency of humans tocopyall of the steps in the arc of a model’s behavior,even those that are superfluous to the perceived goal (Lyons et al.2011; Keupp, Behne, & Rakoczy 2013; Clay & Tennie 2018; Hoehlet al. 2019). Researchers disagree about whether over-imitation is amistake, produced by a generally adaptive capacity for social learningmerely overreaching and exhibiting predictable false positives incertain circumstances, or whether it is a key to the human capacity tosupport not just culture but cumulative culture. An influentialhypothesis about how it might do this is by driving the reproductionand spread of cultural traditions that are so complex that theadaptive benefits they provide are causally opaque to andunappreciated by their users (J. Henrich 2016).
Another debate is over the importance of imitation relative to otherforms of social learning, both to the transmission and accumulation ofculture (Caldwell & Millen 2009; Singh et al. 2021) and to themore creative forms of acquisition of interest to culturalepidemiologists (Sperber 1996; Boyer 2001; Nichols 2004; Morin 2011[2016]). This issue has attracted the attention of philosophers inpart because it intersects with those conceptual issues about fidelityand replication mentioned inSection 3.1. Thus, empirical findings in the cognitive science of imitation standto shed light on conceptual questions about if and when the culturalvariants that spread via social transmission can serve asunits ofselection in cultural evolution, in something analogous to theway that genes serve as units of selection in genetic evolution, invirtue of the fact that they can be replicated with high levels offidelity (R. Boyd & Richerson 2000; J. Henrich & Boyd 2002;Claidière, Scott-Phillips, & Sperber 2014; Acerbi &Mesoudi 2015; Boudry 2018; Charbonneau 2020; Charbonneau & Bourrat2021; also see Lloyd 2005 [2020]; Wilkins & Bourrat 2018[2020]).
A related line of research explores the ways in which imitation andsocial learning are selective, guided by strategies and heuristicsthat dispose individuals to acquire some the behaviors, skills, andideas they are exposed to rather than others (Laland 2004; Chudek,Brosseau-Liard, et al. 2013; Koenig & Sabbagh 2013; Heyes 2016b;Kendal, Boogert, et al. 2018). Many different strategies have beenexplored, and those have been taxonomized in a variety of ways (see J.Henrich & McElreath 2007). Some of the most discussed types ofstrategies include aconformity heuristic that inducesindividuals to copy behaviors and ideas that are already common intheir community (Muthukrishna, Morgan, & Henrich 2016);sensitivities tostatus andsuccess that make onemore likely to copy prestigious, dominant, successful, or otherwisehigh-ranking individuals (J. Henrich & Gil-White 2001; Cheng,Tracy, et al. 2013; Cheng & Tracy 2014; Maner 2017, Jiménez& Mesoudi 2019, 2021), and sensitivities to the similarities ordifferences between oneself and those one might learn from, sometimesdiscussed in terms ofmy-side bias (Mercier & Sperber2017) ortribal social instincts (Richerson & Boyd 2001;Richerson & Henrich 2012). Though they are often called“biases” and lead to what can appear to be reasoningerrors when evaluated in isolated episodes of learning, modeling worksuggests that when such heuristics are prevalent in a population theycan collectively produce results that are beneficial, both for theindividuals whose social learning they are guiding and for thecultural repertoire of the group those individuals belong to(Richerson & Boyd 2005; see also Skyrms 2014). Recentphilosophical work has taken up the question of whether or not thesekinds of selective social learning heuristics can be assimilated totraditional perspectives on rationality and epistemic virtue (N. Levy& Alfano 2020; Fadda 2021). Other philosophical research hascritically examined the explanatory work done by appeal to suchheuristics, and looked at how these biases relate to different formsof social influence (Padalia 2014; Chellappoo 2021).
When social learning is important for the members of a species, it isnot unreasonable to expect there to be complementary psychologicalmechanisms that guide and facilitateteaching. In humans,researchers have investigated the propensity to engage in what hasbeen called ‘natural pedagogy’ (Csibra & Gergely 2006;Gergely, Egyed, & Király 2007, but see Little, Carver,& Legare 2016). For example, some of this work looks atcomplementary patterns of attention and eye contact that emerge inpedagogical settings, and the psychological mechanisms that support acomplex interaction between the ostensive signals that teachersprovide for their students and the referential expectations thatlearners have of their teachers (Csibra & Gergely 2009; cf. Heyes2018b). Other researchers who focus on human mentalizing capacitiesapproach their subject matter from a broadly pedagogy-centeredperspective, developing a general reinterpretation of thosecapacities’ basic functions. On this mindshaping view, themechanisms that underpin human folk psychological practices are notdesigned to merely passively interpret others’ behavior; rather,they primarily evolved to help (co)construct and regulate the minds ofothers. They thus guide many pedagogical tasks involved in activelyhelping to structure and influence the minds of learners (Mameli 2001;McGeer 2007, 2015; Zawidzki 2013). Others have argued that humanpedagogy is both active and deeply facilitated not just by internalpsychological mechanisms, but by meticulously structured learningenvironments—the cumulative epistemic niches—that are acrucial part of the non-biological inheritance that one generationuses to educate the next. These include engineered physicalenvironments (Clark 1997a), but also norms that govern relationshipsbetween apprentices and their mentors (Sterelny 2012) which can scaleup into larger informal and formal institutions of education (Gillett2018). Others have explored ways in which the psychological machineryof learning and teaching contributes to the generation, intentional orotherwise, ofinnovations in cultural variants, and thus tothe production of variation on which cultural evolution operates(Claidière, Scott-Phillips, & Sperber 2014; Legare &Nielsen 2015; Mesoudi 2016; Laland 2017; Leibo et al. 2019; Gopnik2020; Miu et al. 2020; Rule, Tenenbaum, & Piantadosi 2020).
While the psychological mechanisms underlying social transmission arewidely thought to play a foundational role, other cognitive abilitieshave been put forward as additional candidates for being keydifference makers that allow human minds to produce and sustain uniqueforms of complex cumulative culture. Though the exact character oftheir respective contributions remains unclear, some prominentcandidates include: novel and more powerful forms of memory (Donald1991, 2007; Carruthers 2015), especially for particular sequences ofactions (Ghirlanda, Lind, & Enquist 2017; Milne, Wilson, &Christiansen 2018; Petkov & Cate 2020); imagination and mentaltime travel (Suddendorf & Corballis 2007; Dor 2015, 2017; Fuentes2017); analogical thinking (Brand, Mesoudi, & Hewlett 2021);metacognition (Heyes 2016b; Dunstone & Caldwell 2018; also seeHeyes, Bang, et al. 2020); mindreading (Carruthers 2013; Heyes &Frith 2014; Moore 2021; cf. McGeer 2007; Zawidzki 2013); socialtolerance brought about by self-domestication (J. Henrich 2016; Hare2017; Wrangham 2019); shared intentionality (Tomasello, Carpenter, etal. 2005; Tomasello & Moll 2010); norm psychology (Chudek &Henrich 2011; R. Boyd 2017; Davis & Kelly 2018; cf. Andrews 2020a;Fitzpatrick 2020; also see Kelly & Setman 2020); resolve (Ainslie2021; Setman & Kelly 2021); natural language (Ross 2007; Tamariz& Kirby 2016; Dennett 2017: chapter 9; Sterelny 2020); narrative(Scalise Sugiyama 2001, 2017; B. Boyd 2009; Sterelny 2012); reasoning(Pinker 2010; Mercier & Sperber 2011, 2017); and various tool andtechnological proficiencies (Vaesen 2012; A. Schulz forthcoming; Birch2021; cf. Osiurak & Reynaud 2020).
Faced with this proliferation of options, several theorists haveespoused “no magic bullet” positions, holding that thereis unlikely to be any single, crucial psychological advance that willturn out to bethe pivotal cognitive ingredient needed forcumulative culture. Rather, they argue, the gap between human andnon-human capacities for culture will likely flow from small changesand minor increases in the functionality of several psychologicalmechanisms that are shared between species, along with newinteractions that emerge between those and with other ecological andsocial factors (Sterelny 2012; J. Henrich 2016; Laland & Seed2021; also see Ross 2019).
As illustrated throughout the previous sections, philosophers have notbeen mere consumers of research on the cognitive science of culture,but also active contributors to various projects. This section shiftsgears, highlighting several areas in which philosophers have worked toclarify the implications of this research for more conceptual debatesand to draw out its consequences for discussions that havetraditionally been seen as philosophical.
Indeed, philosophers are more and more frequently weaving insights andevidence about culture and cognition into discussions across a numberof areas of philosophical inquiry. This entry does not pretend toprovide an exhaustive overview of such work. Thus, it is worthexplicitly noting that scholars in nearly every philosophical subfieldhave used findings from the cognitive science of culture and culturaldiversity to enrich their understanding of their subject matter andsupport their arguments. For example, in the philosophy of religionand philosophy of language, recent insights into the cognitivefoundations and dynamics of culture have been used to argue that humancapacities for religion and language are not as universal and/orinnate as many philosophers have taken them to be (Cloud 2014; Davis2020). In aesthetics, philosophers have used work on culture andcognition to explore more specific questions about phenomena that werealready taken to exhibit significant culturally variation, such asartistic form and aesthetic taste (Jan 2007; Verpooten & Dewitte2017). In what follows, the focus will be on philosophicalsubdisciplines were the influence of work on the cognitive science ofculture and cultural diversity seems to have been most pervasive:philosophy of mind, philosophy of psychiatry, ethics, epistemology,and philosophical methodology.
The cognitive science of culture has become relevant to one of themost provocative and productive ideas put forward in the philosophy ofmind in recent decades, namely the extended mind thesis (Clark &Chalmers 1998). This is the claim that minds (human minds, anyway,though perhaps others as well; see Japyassú & Laland 2017)can and frequently do extend beyond brains and out into the world. Thethesis can be understood as a particular expression of a broader setof ideas that also includes embodied, embedded, and enactivistapproaches to cognition, and which are often discussed in terms ofexternalism about the mind (Rowlands, Lau, & Deutsch 2020). Thecommon thread running through positions characterized as externalistis the thought that the boundary between a biological organism and itsenvironment is less psychologically important than it maypretheoretically seem. Rather, insists the externalist, manyinteresting and important mental phenomena span that physical border,and what is external to an individual’s skull and skin can bedeterminative and even constitutive of their mind and its contents (R.Wilson 2004).
This core idea has been developed in several ways. Initial versions ofexternalism focused on semantics and mentalcontent, holdingthat what some intentional mental states are about is at least in partdetermined by factors outside the body, including social arrangements,cultural patterns of expertise and deference, and norms that structurethe division of linguistic labor (McGinn 1977; Burge 1979; Fodor 1987,1994; cf. Putnam 1975a). Later and more contentious forms ofexternalism focused on thevehicles of content, holding thatsome cognitive processes and mental states are themselves locatedoutside the head and body. According to this more “active”version of externalism, when a part of the environment or piece oftechnology becomes sufficiently integrated with an individual’sother mental processes, that bit of the world external to the skullis, strictly speaking, just as much a part of her mind as what’sgoing on in her brain. Her cognition extends into the world beyond herhead.
The extended mind thesis is controversial (Rupert 2004; Gertler 2007,2012; Adams & Aizawa 2008) and portions of the debates it hassparked are informed by an array of views about how to understand therelationship between cognition and different aspects of culture. Forinstance, Sterelny (2003, 2012) develops the conceptual resources ofniche construction theory to show how humans actively engineer thecultural niches in which they live, learn, and raise children. Heaccepts that the accumulated structure and adaptive information foundin human environments are key factors for explaining humanintelligence and behavior, but he stops short of fully embracing theextended mind thesis. He holds that human cognitive functioning isgreatly enhanced and amplified—scaffolded—bysocial, epistemic, and technological resources that are culturallytransmitted and located outside the head. But he argues that genuinelyextended mentality is rare, and should be understood as anextreme, limiting case of the more common and theoretically importantphenomena of cognitive scaffolding (Sterelny 2010).
Others have explored more specific ideas about how the hallmarks ofextended cognition are related to human culture and culturalcognition. Many have examined aspects and implications of our abilityto smoothly ‘couple’ our minds with our technology (Clark2007; Palermos 2014; Carter & Palermos 2016) as well as with ourphysically and socially constructed environments (Hutchins 2008, 2011;Davidson & Kelly 2020). Others have explored the idea that culturewas part of the design problem that selected for minds that arecultural—porous, technologically extendable, deeply sociallypermeable—in exactly these ways (Tomasello 1999; R. Boyd,Richerson, & Henrich 2011; Dennett 2017; Kelly & Hoburg 2017;Muthukrishna, Doebeli, et al. 2018). Some of this work focuses onmaterial culture and technology (Jeffares 2010; Malafouris 2013;Sterelny 2018), including how artifacts might scaffold or extendspecific psychological faculties like memory (Clark 2005; Heersminkforthcoming), mathematical cognition (Menary 2015), or economicreasoning (Clark 1997b). Other work focuses on sociality and socialorganization (Gallagher 2013; Sterelny 2016, 2021; Carter, Clark,& Palermos 2018b; Gallagher, Mastrogiorgio, & Petracca 2019).Contributors to this strand of research have also explored the ways inwhich investigations into the nature of culture and extended cognitioncan dovetail with and help inform investigations into the nature ofdistributed cognition, collective mental states, and group agency(Hutchins 1995; Tollefsen 2006; Donald 2007; Huebner 2008; Smaldino& Richerson 2013; Jablonka 2017). This line of thought has eveninspired hypotheses about the evolution of human brain size, linkingincreases in collective intelligence, the division of cognitive labor,and the externalization of knowledge with the decrease in the size ofindividual human brains that appears to have occurred around 3000years ago (DeSilva et al. 2021).
While the experience and recognition of cognitive dysfunction is by nomeans unique to any one culture, academic psychiatry was long focusedon and practiced mainly by WEIRD individuals. Nonetheless, thebeginning of Western ethnopsychiatry and cross-culturalpsychology—the study of the cultural aspects of mental healthand illness (Jovanovski 1995)—predates the recent cultural turnin cognitive science. Early pioneers were iconoclasts of Westernpsychiatry, rejecting its ethnocentrism and setting out to discoverwhat they could learn from non-Western approaches to mental suffering(Ellenberger 1959; Torrey 1972). Their research agendas were driven byquestions about the kinds of practices other cultures had developed totreat individuals with behavioral or mental problems, and how cultureis crucial to understand how individuals give meaning to theirsymptoms (Good 1993). They were also concerned with how culturesdiffered in their conceptions of such problems, especially in how andwhere they drew the distinction between normal and abnormal mentalconditions (Kleinman 1988; Fabrega 1991). What attracted the mostattention, however, were inquiries into the nature and significance ofso-called culture bound syndromes, those ways of being mentally unwellthat are confined to, or are only recognized as diseases within, aparticular culture. Well-studied examples of non-Western culture boundsyndromes includeamok (an episode of rage that often ends inkilling, thought to be typical for the Indonesian Archipelago, seeGriffiths 1997 for illuminating philosophical discussion)koro (a genital-shrinkage anxiety, most common in China andSouthwest Asia, Crozier 2012), andlatah (a conditionassociated with a disordered startle response that can lead toabnormal and extreme behavior, found in southeast Asia, Simons 1996).Other candidates that might fit the description (or that may have fitit at one time) include old hag (Ness 1978), fugue (Hacking 1998),multiple personality disorder (Hacking 1995a), anorexia nervosa (Banks1991), and the Truman Show delusion (Gold & Gold 2012, 2014).
The significance of this range of syndromes lies not in theirstrangeness, but in the striking evidence they provide about the waysin which culture shapes mental illness. Indeed, it suggests thatculture has a profound influence on several key dimensions ofpsychological dysfunction, affecting not just how it is diagnosed andtreated, but also how it is expressed and assigned socialsignificance. For example, schizophrenics from different culturesexperience different kinds of auditory hallucinations, which they andthe members of their community interpret in different ways (Luhrmannet al. 2015). The very existence of culture bound syndromes alsosuggests the possibility that the forms of mental disorder common inWestern societies may be no exceptions to the rule. Certain eatingdisorders, for instance, may be culturally transmitted, spreadingcontagiously through social learning and media consumption(Vandereycken 2011). Evidence also suggests that while some crossculturally shared elements of PTSD flow from an innate human responseto danger, others symptoms exhibit variation. For example, thedepressive symptoms associated with PTSD in American war veterans maybe the result of sufferers understanding themselves to have violatedtheir own culture’s moral norms (Zefferman & Mathew 2021;also see Munch-Jurisic forthcoming).
The mounting evidence that mental disorder is so dramatically shapedby culture seemed to be unsettling. Mainstream psychiatry had longaccepted the general claim that cultural factors can play a role inmental health and the etiology of disorders, by creating conditionsthat trigger unhealthy levels of anxiety or stress, or by stabilizingother circumstances that elicit more extreme and pathological forms ofdiscontent with civilization (Freud 1930). Nonetheless, culturalvariation in symptoms was given relatively little attention comparedto what were thought to be the core, pan-cultural indicators ofuniversal forms of dysfunction. Indeed, until recently, theDiagnostic and Statistical Manual of Mental Disorders (DSM)treated disorders commonly found in WEIRD cultures, along with the waythey presented in symptoms, as (quasi-) universal, while classifyingthose only found in other cultures as ‘culture-boundsyndromes’ and treating them as exotic outliers. Hence oneextreme and slightly tongue-in-cheek description of the perspectiveadopted by the DSM: “American mental illness is universal, andother cultures have specific conditions” (Murphy 2015).
This entrenched orthodoxy has drawn more critical scrutiny as culturalapproaches to psychiatry have gained more traction. For instance, aninfluential ‘network theory’ of mental disorder shows thatnetwork models can provide important new insights into the patterns ofcausal interactions among symptoms, and how these patterns serve tostructure (certain) mental disorders (Borsboom, Cramer, & Kalis2019). A virtue of these models is that they allow forstraightforwardly cultural factors to be involved in the relevantpatterns of causal interaction (also see Kendler, Zachar, & Craver2011). These models are thus able to capture, for example, the kindsof ‘labelling effects’ that have been posited to accountfor the way that naming and talking about certain deviant forms ofbehavior can stigmatize them, feeding back and influencing thepsychological states of those who are so labelled, often producingfurther negative impacts on their mental health (see Hacking 1995b;Tekin 2014; also see Mallon 2016 for more general discussion of‘looping effects’ and their interaction with cognition;cf. Goffman 1961 for a sociological perspective on similarphenomena).
Further evidence of the unsettling of the orthodoxy is that the latestedition of the DSM explicitly dropped the terminology of‘culture bound syndromes’, replacing it with‘cultural concepts of disease’. The rationale for thismodification was that it “acknowledges thatall formsof distress are locally shaped, including the DSM disorders”(American Psychiatric Association 2013, italics in original).Nevertheless, there is reason to think the changes may be merelycosmetic and peripheral rather than indicative of a deeper structuralreorientation of Western psychiatric institutions. Philosophicalcritics have argued that at its heart the DSM remains committed to abiomedical model of disease, still conceptualizing mental disorder viaan analogy with somatic disease. This conception remains inadequate,they argue, in no small part because it is unable to capture the deepinfluence of culture on minds and mental health (Tekin 2011; Tekin,Flanagan, & Graham 2017).
Nevertheless, philosophers of psychiatry continue to reflect on howthe growing appreciation of cultural diversity should inform issues inthe conceptual foundations of the field (Washington 2016; Radden 2019;Murphy ;2010 [2020]; also see Adriaens & De Block 2011). Some haveaddressed questions about whether psychopathologies and categories ofmental disorder are best understood as natural kinds, or whether somealternative—socially-constructed kinds, pragmatic kinds,conventional kinds, interactive kinds, normative kinds—providesa better model (Tsou 2007; Cooper 2004, 2007; Khalidi 2010; Kendler,Zachar, & Craver 2011; Tekin 2016; Stegenga 2018). Given thesubject matter, these debates have immediate practical and ethicalimplications as well. For example, according to proponents of thenormative model, a culture’s conception and categories of mentaldisorder are mainly a reflection of what it values rather than anyobjective mental or physiological facts. This would raise familiarchallenges concerning relativism and disagreements about mentaldisorder between cultures: behaviors and states of mind that areunusual (or even normal) but considered normatively acceptable in oneculture may be pathologized in another, and there would be no furtherfact of the matter that would make correct or incorrect either way ofseparating genuinely pathological and disordered behavior frombehavior that is merely unusual or socially unacceptable behavior(Feit 2020; but see Wakefield 2006). Others have explored differentethical aspects of psychiatric diversity and cognitive dysfunction.For example, the widespread exportation of Western practice andthought about mental illness to the rest of the world may haveresulted in a global transformation of human mental life (Watters2010), leading some philosophers to call for a postcolonial psychiatry(Kamens 2020).
Empirical approaches to moral psychology have risen dramatically ininfluence and visibility in the last few decades (Nichols 2004; Prinz2007; Sinnott-Armstrong 2008a,b,c; Doris and the Moral PsychologyGroup 2010; Doris, Stich, & Walmsley 2020; Doris, Stich, et al.2020; Alfano, Loeb, & Plakias 2018; Kelly & Setman 2020), but,as in other areas of cognitive science, practitioners continue towrestle with the WEIRD challenge. Moral psychologists concerned withcultural and cognitive variation have added to the case thatWesterners are outliers, and along many of the dimensions that appeardirectly relevant to moral theory (Graham et al. 2013, 2018).Comparatively speaking, WEIRD people appear to be moreindividualistic, guilt-ridden, impersonally prosocial, and acceptingof impartial fairness norms, but less nepotistic, less loyal tofriends, and less likely to exhibit ingroup favoritism (J. Schulz etal. 2019; J. Henrich 2020; also see Davis & Kelly forthcoming fora cultural perspective on the psychology of ingroup/outgroup biases).Their moral judgments also appear to be more sensitive to theintentions and mental states of those whose behavior they areevaluating (Barrett, Bolyanatz, et al. 2016; McNamara et al. 2019;Curtin et al. 2020; Barrett & Saxe 2021; also see Willard &McNamara 2019). This has led some philosophers to argue that theexplicit moral theories produced by Western moral philosophers,especially those who rely mainly on their own intuitions, may beworrisomely parochial rather than expressions of panhuman ortranscendent moral truth (Flanagan 2016; Van Norden 2017; see alsosection 4.5 below). Others have explored the ways in which research on thebiological and cultural origins of human moral psychology can helpinform our understanding of the nature of morality itself (Joyce 2007;Machery & Mallon 2010; Stanford 2018). This perspective has ledsome to argue that the category of morality, understood as a specialsubdomain of normativity in general, is itself culturally parochial,an historically recent creation of Western philosophers (Machery 2018;also see Stich 2018; Davis & Kelly 2018).
Others have drawn on empirically-oriented work on culture andcognition to inform more specific debates within moral theory, broadlyconstrued. A wealth of empirical evidence has demonstrated culturaldifferences in behaviors, norms, and judgments concerned withreciprocity, fairness, and altruism (Sober & Wilson 1998; J.Henrich, Boyd, et al. 2004, 2005; Mathew, Boyd, & Van Veelen 2013;Baumard, André, & Sperber 2013; Sterelny 2016).Naturalistic philosophers, in turn, have attempted to clarify thesignificance of the revealed variation, spelling out what it mightmean for metaethical theorizing (Machery, Kelly, & Stich 2005;Doris & Plakias 2008) or how it might reflect on the range andmoral status of different forms of altruism and cooperation (Binmore2006; Zollman 2008; Clavien & Chapuisat 2013; Sterelny, Joyce, etal. 2013, Sterelny 2021; Curry 2016; Curry et al. forthcoming;Piccinini & Schulz 2019; Doris, Stich, & Walmsley 2020).Several philosophers have argued that there is non-trivial culturalvariation in the most basic components of morality, and as such thatthere are unlikely to be many moral universals (Prinz 2008; A. Levy& Levy 2020; cf. Mikhail 2007, 2011).
Philosophers have long been interested in the genealogy of morals(Nietzsche 1887), and the subsequent rise of the fields of cognitivescience, Evolutionary Psychology, and empirical moral psychology hasbrought with it a surge of interest in evolutionary debunkingarguments of morality. These have sparked a great deal ofphilosophical debate, but it has been slow to seriously integratedcultural evolution or the contemporary cognitive science of cultureinto the discussion (Kelly 2017). Critics have pointed out thatdebunkers have rarely paid close attention to the details ofscientific accounts of the evolution of morality, even though some ofthe premises of their debunking arguments seem to rely on suchdetails, including details about the cultural roots of many componentsof our moral minds (Cline 2015; Braddock 2021). Others have similarlytraced out how empirical inaccuracies in the premises of debunkingarguments undermine their conclusions (Machery & Mallon 2010;Sterelny & Fraser 2017; A. Levy & Levy 2020).
Unsurprisingly, moral realists who take a genealogical perspective seethings differently than the debunkers (Huemer 2016). Many philosophershave been attracted to the idea that human cultures are in the midstof a significant historical trend in the direction of moral progress,as the last several centuries have witnessed a putative triumph of theopen society, a proliferation of broadly inclusive norms and valuesacross the globe (Jamieson 2002; Singer 1981 [2011]; Buchanan &Powell 2018; Gaus 2021; also see Pinker 2011, 2018). This wide andrelatively rapid spread of inclusivity is difficult forevolutionarily-inspired moral skeptics and anti-realists to accountfor, the realist argues, because the extension of moral concern beyondmembers of a cultural ingroup seems to be maladaptive from anevolutionary perspective. Thus, realism seems better able to accountfor the trend, explaining why inclusivist norms and values spread soquickly, and of why it only happened so recently.
Critics have responded by arguing that thereare viablealternative accounts for the trend, explanations that appeal to theinterplay between cultural capacities and cultural dynamics ratherthan to the discovery and tracking of mind independent moral facts(Hopster 2020, also see Cofnas 2020). They also point out that it hasbeen known since the establishment of the field of cultural evolutionthat cultural evolutionary processes can and not infrequently do leadto biologically non-adaptive outcomes (R. Boyd & Richerson 1985),and that explanations that can account for the spread of packages ofnorms and values, but do not require a commitment to moral realism,have since grown in sophistication and empirical support (Richerson,Baldini, et al. 2016). Other critics have argued further that whetheror not the realist’s preferred explanation of the historicaltrend is, indeed, the best explanation depends not only in theempirical details about the evolution and character of our moral andcultural capacities, but also on the details of history and how thespread of inclusivity actually occurred. They conclude the realistargument also requires deeper empirical engagement on a number offronts to be convincing (Hassan 2019).
The improved understanding of the impact of culture on cognition, andespecially of the importance of social learning and the extent anddepth of cognitive diversity (J. Henrich, Blasi, et al. forthcoming),has also attracted the attention of epistemologists (Machery, Stich,et al. 2017; Mizumoto et al. 2018; also see Figueroa & Harding2003, especially Wylie 2003; Feest & Sturm 2011). Westernepistemology has traditionally focused on questions about what it isfor individuals to have knowledge, but the last several decades haveseen a growing interest in socially-oriented approaches to epistemicphenomena and to knowers themselves. These social approaches, in turn,are better positioned to capture and account for the ways in whichaspects of culture are epistemologically significant (Carter, Clark etal. 2018a; Grasswick 2006 [2018]; Goldman & O’Connor 2019[2021]).
For example, this social turn in epistemology has sparked researchinto cognitive mechanisms for epistemic vigilance, especially thosethat monitor and regulate communication (Sperber, Clément, etal. 2010; Mercier 2020). It has also directed attention towards manydifferent aspects of epistemic norms, those informal rules of conductthat regulate how people handle information, assess evidence ofvarious kinds, make public assertions (Kneer 2021), etc. Epistemicnorms also underpin social trust, prescribing how one ought to assigncredence to testimony given by people who bear different socialidentities, belong to different groups, and are associated withdifferent institutions and forms of expertise (Lackey 2008; Goldberg2010; Littlejohn & Turri 2014; Henderson 2020). An especiallyfruitful strand of this research has examined variation in such norms,and in the ways in which they can be unfair to members of marginalizedcultural groups, and thus contribute to a myriad of forms of epistemicinjustice (Fricker 2007; Medina 2013; Dotson 2014). Other research atthe intersection of social epistemology and political philosophy(Edenberg & Hannon 2021) has looked at the way certainconfigurations of epistemic norms, together with the operation ofselective learning heuristics like the my-side bias and tribal socialinstincts mentioned inSection 3.3, can create filter bubbles and echo chambers (E. Anderson 2021; Nguyen2020), contribute to the spread of misinformation (O’Connor& Weatherall 2019), and sustain extreme forms of polarization andpartisanship (Rini 2017, also see Raymond, Kelly, & Hennesforthcoming). Work on the character of scientific inquiry has alsolooked at the role of epistemic norms in the production of knowledgeby scientific communities (Kuhn 1962; Barnes et al. 1996; Vaesen2021), including norms that specify what counts as acceptable evidence(Longino 1990) and norms that prescribe that empirical evidence beaccorded a unique significance in the construction and evaluation ofscientific argumentation (Strevens 2020).
Another strand in this research examines familiar tensions betweenindividual and collective levels of analysis as they emerge inepistemological contexts. For example, psychological tendencies thatlook like biases or failures of rationality from the point of view ofindividual reasoning may be the very same tendencies that, when widelyshared, allow for the accumulation over time of group beneficialcultural traits and adaptive wherewithal. The tendency of humans toover-imitate, for example, may lead them to occasionally expend extratime and energy mimicking unnecessary steps in a sequence ofbehaviors. Considered individually, episodes of copying “toomuch” can look like a mistake, and the underlying propensitylike a source of irrationality and inefficiency. From a collectiveperspective, however, such a tendency, distributed across members ofthe population, may be a key factor in the group’s ability tocreate, sustain, and faithfully transmit cumulative culture andknowledge over the span of generations (Hoehl et al. 2019, also see J.Henrich 2016). Similarly, others have argued that what look likereasoning mistakes when examined in isolation are driven by heuristicslike a my-side bias that make important contributions to agroup’s ability reap the benefits of dividing cognitive labor(Mercier & Sperber 2011, 2017; also see Peters forthcoming onconfirmation bias). A similar perspective has been used to recastquestions about epistemic accountability, suggesting that while many‘biases’ may facilitate collective reasoning and the grouplevel production of knowledge, individuals who employ those biasesfail to qualify as epistemically virtuous. Those individuals, that is,do not deserve any epistemic credit for the knowledge their biaseshelp produce (N. Levy & Alfano 2020; also see Palermos 2016 andAlfano 2021; cf. Rini 2017 for an argument that inverts theevaluation, holding that in some contexts partisanship in acceptanceof testimony can be individually virtuous but lead to epistemicoutcomes—like the widespread circulation of fake news—thatare clearly vicious at collective levels).
Contemporary philosophy that is considered Western is recognized assuch in virtue of tracing its origins back to a particulargeographical region, centered especially in Athens and Rome, but alsoin virtue of a shared body of topics and methods. Appreciation ofhuman cultural and cognitive diversity can fuel the suspicion thatmany topics worthy of philosophical consideration have been given lessattention than they deserve by the Western tradition. This, in turn,can be used to support the case that more Western philosophers shouldbranch out, exploring comparative projects and studying othertraditions (Van Norden 2017, also see Coseru 2009 [2017]; Outlaw 2010[2017]; Vargas 2007; Kasulis 2019; Wong 2001 [2020]; Møllgaard2021).
Within the Western tradition, arguments from cultural and cognitivediversity have come to loom large in debates about philosophicalmethodology. Some have pointed out that many traditions use approachesthat seem alien to contemporary Western philosophy, such as meditationor syncretism (McLeod 2016, also see Struck 2016 for a case thatdivination was taken seriously by ancient Greek philosophers, andJones 2001 for an argument that philosophical methods are universal).Most discussions in this area, however, have centered on intuitions,and their character, distribution across populations, and proper rolein philosophical theorizing (also see Pust 2012 [2019])
Since the early 2000s, experimental philosophers have gathered theirown data and used it to support a variety of projects (Sytsma &Livengood 2016; Knobe & Nichols 2017). One of the most prominenthas tried to develop arguments showing how diversity—in culture,in cognition, and especially in philosophical intuition—presentschallenges to philosophers who rely too heavily on appeals to theirown intuitions, or solely on the intuitions of a select few (Eglerforthcoming). Intuitions about philosophical issues may vary in waysthat raise doubts about the significance of projects that fail to takeaccount of such variation. As an early manifesto put it:
Since it’s in some sense an accident that I had the culturalupbringing that I did, I am forced to wonder whether my intuitions aresuperior at tracking the nature of the world, the mind, and the good.(Knobe & Nichols 2008, 11; also see DiPaolo & Simpson 2016;Andow 2016)
Conclusions reached mainly on the basis of ‘close to home’intuitions may be relative and limited in scope, the argument goes, asphilosophers in other cultures and with other intuitions mayrationally arrive at different or even contrasting conclusions.
The empirical situation as revealed by experimental philosophy,however, is far from simple or straightforward. Initial effortspresented evidence of diversity in the intuitive judgments people fromdifferent groups made about the moral, epistemic, and semanticfeatures of a range of scenarios. For example, one of the firstexperimental philosophy studies found that East Asian participantsexhibited markedly different responses than Westerner participants tothe kinds of Gettier cases commonly used by analytic epistemologists(Weinberg, Nichols, & Stich 2001). Similarly, the firstcross-cultural study of semantic intuitions about reference, whichused scenarios based on Kripke’s famous Gödel-case, againfound evidence of interesting differences between East Asian andWestern participants (Machery, Mallon, et al. 2004).
Some, but not all, of these results have replicated (Cova, Strickland,et al. 2021). While later studies seemed to confirm that semanticintuitions differ cross-culturally (Beebe & Undercoffer 2016; vanDongen et al. 2021), subsequent research has indicated there may becross-cultural convergence on Gettier-cases and other epistemicintuitions (Kim & Yuan 2015; Machery, Stich, et al. 2017; Yuan& Kim forthcoming; but see Waterman et al., 2018; Sękowski etal. forthcoming). Empirical research on intuitions about free will hasrevealed another messy picture, with some work suggesting that manyyoung Americans have compatibilist intuitions (Nahmias et al. 2006),while other studies find evidence of convergence, with participantsfrom different cultures embracing incompatibilism and rejectingdeterminism (Sarkissian et al. 2010). Other research suggests thatpeople in Asia tend to make judgments about free will thatsystematically differ from the judgments in the rest of the world(Hannikainen et al. 2019). Investigation into aesthetic intuitionsfound that participants across cultures broadly agree that aestheticjudgments are subjective and not universally valid, though there isevidence of patterns of differences in detail (Cova, Garcia, &Liao 2015).
Philosophers continue to debate the implications of this complicatedand evolving empirical picture. Some have explored how such evidencemay be put to use by experimental philosophers engaged in positivesubstantive projects of explication and conceptual engineering(Shepherd & Justus 2015; Schupbach 2017; Nado 2021). Even whenintuitions about philosophical cases do not significantly vary,systematic investigation of them may still yield important evidence inthe analysis of concepts (Stich & Tobia 2016, also see Knobe 2016and Machery 2017).
The empirical picture has been interpreted by advocates of moretraditional (Western) methods to support those methods. Some havegiven a close analysis of the nature and use of intuitions, anddefended the perhaps surprisingly view that even if they do exhibitpatterns of cross-cultural variation, it would be irrelevant tophilosophical methodology because intuitions play only a minor role inphilosophical argumentation (Cappelen 2012, also see Ichikawa, Maitra,& Weatherson 2012). Others have accepted that intuitions areimportant to philosophical methodology, but argued that the empiricalrecord does not reflect enough variation, or the right kind ofvariation, to be philosophically troubling (Knobe 2019; cf. Stich& Machery forthcoming). Some have been most impressed with thediscovered uniformity, arguing that it deflates experimentalphilosophy’s diversity-based methodological challenge (Ichikawa2012).
Another line of response to the challenge focuses on the details inthe patterns of variation, arguing that the revealed diversity isoften superficial or philosophically toothless (Sosa 2009), or thatdisagreement suggested by the presence of conflicting intuitions wouldusually disappear if the individuals from different cultures coulddiscuss cases or provide justifications for their intuitions (Deutsch2010). Debates about one such response, which has become known as theexpertise defense, consider in detail the possibility that someinstances of variation are innocuous. Some people’s intuitions,specifically those people with special expertise or training, are moresensitive to philosophical issues. Thus, not only should differencesbetween lay people and experts to be expected, but, continues theargument, the intuitions of the experts deserve more evidential weightin philosophizing (Devitt 2011; Williamson 2011; also see Horvath& Koch 2021). This defense, claim its critics, relies on whatappear to be questionable empirical assumptions about experts andtheir intuitions (Weinberg, Gonnerman, et al. 2010; Schwitzgebel &Cushman 2015; Wiegmann, Horvath, & Meyer 2020; Horvath &Wiegmann forthcoming). One response that is particularly intriguingfor those interested in cognition and cultural variation is that theintuitions of experts—trained philosophers—seem to varyquite dramatically according to their linguistic background (Vaesen,Peterson, & Van Bezooijen 2013). These debates continue. No one,however, has argued against the idea that reflection on culturaldiversity in worldviews, values, conceptualizations, cognition, orintuition can be philosophically valuable in ways that go far beyondtheir relevance to debates about methodology.
How to cite this entry. Preview the PDF version of this entry at theFriends of the SEP Society. Look up topics and thinkers related to this entry at the Internet Philosophy Ontology Project (InPhO). Enhanced bibliography for this entryatPhilPapers, with links to its database.
Africana Philosophy |altruism: empirical approaches |cognition: embodied |cognitive science |comparative philosophy: Chinese and Western |connectionism |culture |epistemology: social |evolution: cultural |experimental moral philosophy |experimental philosophy |externalism about the mind |feminist philosophy, interventions: social epistemology |human nature |innateness: historical controversies |intuition |Japanese Philosophy |Latin American Philosophy |mental disorder |mental representation |mind: computational theory of |mind: in Indian Buddhist Philosophy |mind: modularity of |moral psychology: empirical approaches |natural selection: units and levels of |normative cognition, psychology of |psychiatry, philosophy of |psychology: evolutionary |rationalism vs. empiricism |replication and reproduction |reproducibility, scientific
View this site from another server:
The Stanford Encyclopedia of Philosophy iscopyright © 2023 byThe Metaphysics Research Lab, Department of Philosophy, Stanford University
Library of Congress Catalog Data: ISSN 1095-5054