29.12.02 · psychology / cross-cultural-indigenous

Cross-cultural cognition: Nisbett's geography of thought, East Asian holistic vs Western analytic reasoning

shipped3 tiersLean: none

Anchor (Master): Markus and Kitayama 1991 Psych. Rev. 98:224; Nisbett, Peng, Choi, Norenzayan 2001 Science; Masuda and Nisbett 2001; Ji, Nisbett, Zhang 2004; Choi, Nisbett, Norenzayan 2000; Peng and Nisbett 1999; Norenzayan et al. 2007; Henrich, Heine, Norenzayan 2010 Nature 466:29; Henrich 2020 The WEIRDest People in the World

Intuition Beginner

Do people from different cultures actually think differently — not just speak different languages or hold different beliefs, but perceive, categorise, and reason in fundamentally different ways? Richard Nisbett and his collaborators at Michigan spent more than twenty years testing this question. Their answer, summarised in Nisbett's 2003 book The Geography of Thought, is that the average cognitive style of East Asian populations (Chinese, Japanese, Korean) differs systematically from the average cognitive style of Western populations (American, European) across attention, categorisation, causal explanation, and reasoning.

The pattern they reported runs as follows. East Asian participants tend to think holistically — attending to relationships and context, grouping objects by family resemblance, explaining events by situational factors, and preferring dialectical resolutions of contradiction. Western participants tend to think analytically — attending to focal objects and their attributes, grouping objects by shared rules, explaining events by individual dispositions, and preferring formal-logical resolutions of contradiction. The differences are averages: there is enormous variation within each population, and the same person can switch frames depending on context.

Nisbett's cultural-history hypothesis links these averages to philosophical traditions. Ancient Greek philosophy (Aristotle, Plato) emphasised categories, rules, and linear logic; ancient Chinese philosophy (Confucius, Laozi) emphasised harmony, relationships, and dialectical balance. The claim is that these traditions shaped cognitive tendencies that persist today. The contemporary debate is whether the differences are as robust as the early reports suggested, and how far they generalise beyond the small samples of the original studies.

Visual Beginner

The picture lays out the cross-cultural cognition program as a horizontal axis from 1991 to the present. Each milestone is marked with its contributor and the central experimental or theoretical move. The arc opens with Markus-Kitayama's 1991 framework for the independent and interdependent self, becomes a systematic cognitive-style program with Nisbett et al.'s 2001 Science synthesis, accumulates empirical paradigms through the 2000s, and is reframed by the WEIRD critique in 2010.

Two strands run in parallel. The self-construal strand runs through Markus-Kitayama and the social-psychological literature on independent versus interdependent self. The cognitive-style strand runs through Nisbett's program and its experimental paradigms. Their intersection is where the strongest claims about cross-cultural cognition live.

Worked example Beginner

Setting. Masuda and Nisbett (2001) showed short videos of underwater scenes to thirty-six Japanese and thirty-six American students at Kyoto University and the University of Michigan. Each scene contained focal fish (large, bright, fast-moving) and background objects (rocks, seaweed, smaller animals). An eye-tracker recorded where each participant looked, moment by moment, across roughly $90$ seconds of footage.

Step 1. Where the eyes went. American participants looked first, and longest, at the focal fish. Their gaze locked onto the biggest, brightest, most central object in the scene. Japanese participants looked at the background first — the rocks, the seaweed, the smaller animals — and only then moved to the focal fish. The average ratio of background-looking time to focal-object-looking time was approximately $1.4$ times higher for Japanese than for Americans.

Step 2. What was remembered. After the videos, participants wrote what they had seen. Japanese descriptions began with the background ("there was a pond, with rocks and seaweed...") roughly $80$ per cent of the time; American descriptions began with the focal fish ("there was a big fish, what looked like a trout...") roughly $80$ per cent of the time. Japanese participants reported roughly twice as many background-object relationships as Americans.

Step 3. Recognition with changed background. Later, participants were shown the focal fish again, sometimes against the original background and sometimes against a new background. Americans recognised the focal fish at roughly the same rate in both conditions ( $M \approx 0.83$ vs $0.81$ ). Japanese recognition dropped when the background changed ( $M \approx 0.76$ vs $0.61$ ): they had encoded the fish together with its context, and removing the context impaired recognition.

What this tells us: the East Asian holistic style and the Western analytic style are not abstractions but measurable patterns in where the eyes go and what the memory retains. East Asian participants attend to the relationships between focal objects and their context; Western participants attend to the focal object detached from its context. The same physical scene produces systematically different perceptual records.

Check your understanding Beginner

Exercise (easy, multiple choice).

In the Masuda-Nisbett 2001 eye-tracking study, which of the following best describes the recognition performance of Japanese participants when a focal fish was shown against a new background?

A. Japanese recognised the fish at the same rate as against the original background. B. Japanese recognition of the fish dropped when the background was changed. C. Japanese recognised the fish better than Americans in both conditions. D. Japanese failed to recognise the fish in either condition.

Hint

Recall that Japanese participants were reported to have encoded the focal fish together with its surrounding context, while American participants encoded the fish detached from context.

Answer

B. Japanese recognition of the fish dropped when the background was changed. Japanese participants had encoded the focal fish together with its context; changing the background disrupted retrieval. American participants, who had encoded the fish detached from context, recognised it at roughly the same rate regardless of background. This asymmetry is the load-bearing empirical finding for the claim that East Asian attention is more context-sensitive than Western attention.

Exercise (easy, true-false).

The cross-cultural differences reported by Nisbett and colleagues are interpreted as genetic differences between East Asian and Western populations.

Hint

Consider the bicultural-priming studies, in which the same Hong Kong Chinese participant is shown to switch between holistic and analytic styles depending on which cultural frame has been activated.

Answer

False. The evidence is environmental and cultural, not genetic. Bicultural-priming studies (Hong, Chiu, Kung 1997 onwards) show that the same bicultural individual — for example a Hong Kong Chinese student exposed to both Cantonese and Western media — can be shifted toward holistic style by Chinese cultural primes and toward analytic style by Western cultural primes within a single experimental session. A genetic account cannot accommodate such frame-switching; a cultural account does so naturally. The cross-cultural differences are therefore best understood as learned cognitive styles shaped by cultural environment, not as fixed biological attributes.

Exercise (easy, numeric).

In the Masuda-Nisbett study, the average ratio of background-looking time to focal-object-looking time was approximately $1.4$ times higher for Japanese than for Americans. If American participants spent $40$ seconds looking at focal objects, approximately how many seconds did Japanese participants spend looking at the background in addition to the American background-looking time, assuming Americans spent $30$ seconds on background? State the answer as a single number of seconds.

Hint

The Japanese background-looking time is $1.4$ times the American background-looking time. The additional seconds are the difference.

Answer

$12$ seconds. Japanese background-looking time is $1.4 \times 30 = 42$ seconds. The additional time beyond the American $30$ seconds is $42 - 30 = 12$ seconds. The exact magnitude varies across published reports of the paradigm; the point of the exercise is that the gap is large enough to be visible in raw looking-time totals, not only in derived ratios.

Formal definition Intermediate+

The cross-cultural cognition program formalises cognitive style as a stable, measurable pattern in how a participant deploys attention, categorisation, causal attribution, and reasoning. Two styles are contrasted. The framework draws on Witkin's field-dependence/independence (1962), Berry's ecocultural model (1970), and Markus-Kitayama's self-construal framework (1991), and is given its programmatic statement in Nisbett, Peng, Choi, and Norenzayan (2001) ^{[NisbettPengChoiNorenzayan2001]}.

Definition (Independent vs interdependent self-construal; Markus-Kitayama 1991). Let $S$ denote a participant's self-construal — the working representation of the self in relation to others. An independent self-construal, normative in Western populations, represents the self as separate, autonomous, stable across contexts, and defined by internal traits (abilities, attitudes, preferences). An interdependent self-construal, normative in East Asian populations, represents the self as relational, context-sensitive, and defined by social roles and ongoing relationships. The two construals are endpoints of a continuum rather than a dichotomy; within-population variation is substantial, and the same individual can be primed toward either construal ^{[MarkusKitayama1991]}.

Definition (Holistic vs analytic cognitive style; Nisbett et al. 2001). Let $T$ denote a cognitive task drawn from one of four domains — attention, categorisation, causal attribution, or reasoning. A holistic style, associated with East Asian samples on average, is characterised by: (i) attention to the field and to the relationships between focal objects and their context; (ii) categorisation by family resemblance and thematic relations; (iii) causal attribution to situational and contextual factors; (iv) dialectical reasoning that seeks a middle path between apparently contradictory propositions. An analytic style, associated with Western samples on average, is characterised by: (i) attention to focal objects and their attributes detached from context; (ii) categorisation by shared rules and necessary features; (iii) causal attribution to dispositional properties of individual actors; (iv) formal-logical reasoning that resolves contradictions by determining which side is correct.

Definition (Cultural-history hypothesis). The cultural-history hypothesis attributes the persistent average difference in cognitive style between East Asian and Western populations to the long-term influence of their respective philosophical traditions. Ancient Greek philosophy (Aristotle, Plato, the Stoics) emphasised categories, rules, linear logic, and the abstraction of objects from their context. Ancient Chinese philosophy (Confucius, Laozi, Mencius) emphasised harmony, relationships, dialectical balance, and the embedding of objects in their field. The hypothesis is that these traditions, mediated through language, education, and social practice, shaped the cognitive tendencies that contemporary cross-cultural research measures.

Definition (WEIRD sample; Henrich-Heine-Norenzayan 2010). A WEIRD sample is one drawn from a population that is Western, Educated, Industrialised, Rich, and Democratic. Henrich et al. (2010) ^{[HenrichHeineNorenzayan2010]} showed that WEIRD samples — and especially American undergraduates — are statistical outliers on a wide range of psychological measures including visual perception, fairness, cooperation, self-concept, and analytic-holistic style. The critique implies that findings based on WEIRD samples cannot be assumed to generalise to the human population; it does not entail that WEIRD findings are wrong, only that they require cross-cultural replication before being treated as universal.

Counterexamples to common slips

All East Asians think the same way. No. The reported differences are averages across populations. Within-culture variation is substantial — typically much larger than the between-culture mean difference — and many individual East Asian participants score more analytically than the Western mean, and vice versa.
All Westerners are analytic. No. The same caveats apply. Sub-populations within Western countries (working-class, rural, religiously orthodox) often score less analytically than the standard university sample; urbanisation, formal schooling, and Protestant cultural inheritance all shift the mean.
The differences are genetic. No. The decisive evidence is bicultural-priming: the same individual (e.g., a Hong Kong Chinese bicultural) shifts between holistic and analytic styles within a single session in response to cultural primes. A genetic account cannot accommodate such frame-switching; a cultural account does so naturally. Cross-cultural adoption studies and second-generation immigrant data show intermediate patterns, also inconsistent with a fixed genetic channel.
The WEIRD critique invalidates cross-cultural research. No. The critique strengthens the research by demanding more diverse sampling, multi-site collaboration, pre-registration, and measurement-invariance testing. It does not negate the specific findings on holistic-analytic style, which are themselves comparisons across non-WEIRD and WEIRD samples; it sharpens the demand for replication.
Nisbett proved East Asians are superior. No. Holistic and analytic styles are equally useful in different contexts. Holistic attention is advantageous for change detection in complex scenes; analytic attention is advantageous for isolating variables in formal reasoning. The framework is descriptive of average patterns, not evaluative of worth.
The Geography of Thought is settled. No. The framework remains in active debate. Several specific priming effects have failed to replicate; the magnitude of between-culture differences varies across paradigms; and some findings are sensitive to the choice of stimuli, language of testing, and experimenter effects. The broad pattern replicates; specific effect sizes vary.
Markus-Kitayama is the only framework. No. Triandis's individualism-collectivism, Hofstede's cultural dimensions, Schwartz's value circumplex, and the ecocultural model of Berry all address overlapping territory from different angles. Markus-Kitayama is the most-cited for the self-construal construct specifically; the others contribute distinct dimensions.
Cultural differences are fixed. No. Second-generation immigrants show intermediate patterns; bicultural individuals frame-switch within minutes; multi-decade cohort comparisons show secular drift. Globalisation, urbanisation, and education appear to be reducing some traditional differences while preserving others.
Indigenous psychologies are anti-science. No. Indigenous psychologies (Filipino Sikolohiyang Pilipino of Enriquez, Indian Indian Psychology of Sinha and Rao, Chinese indigenous frameworks of Hwang and Yang, Maori frameworks) complement cross-cultural comparison by building local psychologies from within cultural traditions rather than importing Western constructs. They are not a refusal of method but a call for methodological pluralism.
The fundamental attribution error does not exist in East Asia. It exists but is attenuated. Choi, Nisbett, and Norenzayan (1999) and subsequent replications show that East Asian participants make dispositional attributions, just less than Western participants; the cross-cultural difference is in magnitude, not in presence-versus-absence.
These findings justify stereotypes. No. They describe population averages, not individual destiny. Stereotype application requires the additional, invalid step of treating an average difference as a deterministic property of each individual member of the population. The empirical literature explicitly warns against this conflation.

Key result: the East Asian holistic vs Western analytic cognitive styles Intermediate+

Result (The Nisbett et al. 2001 cross-cultural cognition program). Across the four cognitive domains of attention, categorisation, causal attribution, and reasoning, and across multiple operationalisations in each domain, East Asian samples (Chinese, Japanese, Korean) on average show a more holistic style and Western samples (American, European) on average show a more analytic style. The differences are consistent with the Markus-Kitayama (1991) framework for independent versus interdependent self-construal and with the cultural-history hypothesis that links the styles to the ancient Greek and ancient Chinese philosophical traditions.

Defence. The program rests on five experimental paradigms, each replicated at least once.

Attention. Masuda and Nisbett (2001) ^{[MasudaNisbett2001]} used eye-tracking on underwater scenes; Japanese participants attended more to background context, Americans to focal fish, with the recognition asymmetry under changed backgrounds described in the worked example. Kitayama, Duffy, Kawamura, and Larsen (2003) replicated the framing with the framed-line task: participants drew a line in a square either of the same absolute length as a model line (an analytic, context-independent judgement) or of the same length relative to the surrounding square (a holistic, context-dependent judgement). Japanese performed better on the relative task; Americans on the absolute task.

Categorisation. Ji, Zhang, and Nisbett (2004) ^{[JiNisbettZhang2004]} presented triads of words (e.g., panda, monkey, banana) and asked participants to choose the two that go together. East Asian participants preferred thematic groupings (monkey + banana, on the basis that monkeys eat bananas); Western participants preferred taxonomic groupings (panda + monkey, on the basis that both are animals). The same pattern emerged for object triads and for verbal categories.

Causal attribution. Choi, Nisbett, and Norenzayan (1999) ^{[ChoiNisbettNorenzayan1999]} had participants read an essay written under instructed constraint and rate the author's true attitude. American participants rated the attitude as more genuine to the essay (the standard fundamental-attribution-error pattern); East Korean participants weighted the situational constraint more heavily. The fundamental attribution error is present in both populations but attenuated in the East Asian sample.

Reasoning. Peng and Nisbett (1999) ^{[PengNisbett1999]} presented pairs of apparently contradictory propositions (e.g., about fuel economy versus safety in automobiles). Chinese participants preferred a dialectical resolution that acknowledged partial truth in both sides; American participants preferred to identify which side was correct. When given a series of conditional syllogisms, American participants showed stronger competence with formal-logical validity; Chinese participants showed stronger tolerance for apparent contradiction.

Bicultural frame-switching. Hong, Morris, Chiu, and Benet-Martinez (2000) primed Hong Kong Chinese biculturals with either Chinese or American cultural images and observed a shift in performance on attribution tasks toward the corresponding mono-cultural baseline. The result is decisive against a purely genetic account and demonstrates the cultural-environmental mediation of cognitive style.

Caveat (the WEIRD qualification). The studies above compare samples drawn largely from East Asian university populations and American / European university populations. Henrich, Heine, and Norenzayan (2010) ^{[HenrichHeineNorenzayan2010]} showed that such samples are outliers on many psychological measures. The within-population variation exceeds the between-population mean difference in nearly every paradigm; small-sample and single-site studies are vulnerable to effect-size inflation; some priming effects have been harder to replicate in the post-2011 era. The defensible claim is therefore that the broad pattern of holistic-analytic style differences replicates within the populations studied, and that generalisation to all humans — or to non-urban, non-university, non-industrialised populations — requires broader sampling. The 2010s multi-site collaborations (Mosteller et al.; the Many Labs replications) found the broad pattern at reduced effect sizes; specific paradigms remain contested.

Bridge. The cross-cultural cognition program is the foundational reason that "human cognition" cannot be inferred from a single cultural population, and this is exactly the picture that identifies the analytic style with the focal-object attention measured by the Masuda-Nisbett eye-tracking paradigm and the holistic style with the context-sensitive attention measured by the Kitayama-Duffy framed-line task. The central insight is that cognitive style is a measurable average over a population, subject to within-group variation and to bicultural frame-switching, and the bridge is between the self-construal construct of Markus-Kitayama 1991 ^{[MarkusKitayama1991]} and the cognitive-style constructs of Nisbett et al. 2001, with the cultural-history hypothesis supplying the mediating account. The framework builds toward 29.05.04 working-memory research, where the central-executive component of the Baddeley-Hitch model determines how attention is allocated between focal and contextual information, and the holistic-analytic contrast is precisely the population-level signature of how that allocation differs; the pattern appears again in 29.13.04 intelligence-testing history, where the cultural loading of the Army Alpha/Beta and the hereditarian misreading of group IQ differences are the methodological warning that the WEIRD critique sharpens, and the chapter-closing synthesis in 29.12.01 places the present unit within the broader cross-cultural survey that anchors the chapter.

Exercises Intermediate+

Exercise 1 (easy, multiple choice).

Markus and Kitayama's (1991) distinction between independent and interdependent self-construals is best described as:

A. A dichotomy in which every individual is either fully independent or fully interdependent. B. A continuum along which populations show different average positions, with substantial within-population variation. C. A genetic trait fixed at birth. D. A measure of intelligence, with the independent construal indicating higher cognitive capacity.

Hint

Recall that the framework was formulated to account for population-level differences in cognition, emotion, and motivation, and explicitly allows within-population variation and context-dependent frame-switching.

Answer

B. A continuum along which populations show different average positions, with substantial within-population variation. The Markus-Kitayama framework treats independent and interdependent self-construals as endpoints of a continuum. Populations differ in their average position on the continuum — Western samples average more independent, East Asian samples average more interdependent — but within-population variation is substantial and the same individual can shift toward either construal in response to context or experimental primes. The framework is descriptive of population averages, not a dichotomy at the individual level.

Exercise 2 (easy, true-false).

The WEIRD critique of Henrich, Heine, and Norenzayan (2010) holds that findings from Western, Educated, Industrialised, Rich, and Democratic samples should never be used in psychological research.

Hint

The critique is about generalisation, not about validity within the sampled population. Re-read the Definition of WEIRD sample above.

Answer

False. The WEIRD critique holds that findings from WEIRD samples cannot be generalised to the human population without cross-cultural replication, because WEIRD samples — and especially American undergraduates — are statistical outliers on many psychological dimensions. The critique does not hold that WEIRD findings are wrong within their own population; it holds that they are insufficient evidence for universal claims. The cross-cultural cognition program is itself a partial answer to the WEIRD critique: it compares WEIRD to non-WEIRD samples directly, rather than treating the WEIRD sample as the universal baseline.

Exercise 3 (medium, numeric).

A cross-cultural study reports that on a focal-object attention task, the American sample mean is $\overset{ˉ}{X}_{A} = 100$ with standard deviation $s_{A} = 15$ , and the East Asian sample mean is $\overset{ˉ}{X}_{E} = 88$ with standard deviation $s_{E} = 15$ . Compute Cohen's $d$ for the between-culture mean difference, using the pooled standard deviation. Report the answer rounded to two decimal places.

Hint

With equal standard deviations, the pooled estimate is $s_{p} = 15$ . Cohen's $d$ is the difference in means divided by the pooled standard deviation: $d = (\overset{ˉ}{X}_{A} - \overset{ˉ}{X}_{E}) / s_{p}$ .

Answer

$d = 0.80$ . With $s_{A} = s_{E} = 15$ , the pooled standard deviation is $s_{p} = 15$ . Then

d = \frac{X ˉ _{A} - X ˉ _{E}}{s _{p}} = \frac{100 - 88}{15} = \frac{12}{15} = 0.80.

A Cohen's $d$ of $0.80$ is conventionally interpreted as a large effect. Many of the original Nisbett-program paradigms reported effect sizes in this range; the post-2011 multi-site replications typically found smaller effects in the $d = 0.2$ to $0.4$ range, which is the central reason the modern literature treats the broad pattern as robust but the specific magnitudes as contested.

Exercise 4 (medium, short answer).

In Ji, Zhang, and Nisbett's (2004) categorisation paradigm, participants were given triads such as (panda, monkey, banana) and asked to pick the two that go together. Explain the typical East Asian and Western response patterns, and state why this counts as evidence for the holistic-analytic distinction rather than merely a difference in vocabulary or world knowledge.

Hint

Consider what kind of relation each pairing rests on: panda + monkey share a taxonomic category (both animals); monkey + banana share a thematic relation (monkeys eat bananas). Taxonomic categorisation is rule-based; thematic categorisation is relation-based.

Answer

East Asian participants on average prefer thematic or family-resemblance groupings — in the example, monkey + banana, on the basis that monkeys eat bananas (a thematic relation) and that pandas and monkeys share the broader family resemblance of mammals without being grouped by a sharp rule. Western participants on average prefer taxonomic or rule-based groupings — in the example, panda + monkey, on the basis that both are animals (a shared category with necessary features).

The result counts as evidence for the holistic-analytic distinction rather than a vocabulary difference because the triads are constructed to be familiar to both populations (pre-testing establishes this), the instructions are held constant across groups, and the same pattern emerges for non-verbal object triads where vocabulary is not in play. The relevant dimension is which kind of relation the participant treats as criterial: a context-embedded relation (thematic, holistic) or a decontextualised rule (taxonomic, analytic). Within-population variation is substantial, and the same participant can shift toward the alternative pattern when primed; but the average difference between East Asian and Western samples replicates across stimulus sets and across laboratories.

Exercise 5 (medium, short answer).

Explain in plain English why the bicultural-priming finding (Hong, Morris, Chiu, Benet-Martinez 2000) is decisive evidence against a strong genetic account of holistic-analytic cognitive-style differences.

Hint

The same individual participant is tested in two conditions separated by a brief priming manipulation. What can change between the two conditions, and what cannot?

Answer

Bicultural-priming studies recruit participants who have internalised both East Asian and Western cultural frames — typically Hong Kong Chinese students who have grown up with both Cantonese and English media. In a single experimental session, the participant is randomly assigned to be primed with either Chinese cultural images (e.g., a Chinese dragon, pictures of Hong Kong landmarks) or Western cultural images (e.g., the Statue of Liberty, pictures of American landmarks), and then completes a cognitive-style task such as the Ji-Nisbett categorisation triads or a causal-attribution scenario.

The finding is that the same individual shifts toward the holistic pattern under Chinese primes and toward the analytic pattern under Western primes, with effect sizes on the same order as the cross-cultural mean difference.

A genetic account cannot accommodate this pattern. The participant's genome is fixed across the two conditions of the session. What changes between conditions is the activated cultural frame — an environmental manipulation lasting seconds. The within-individual frame-switch therefore identifies the cultural environment, not the genetic endowment, as the proximate cause of the cognitive-style shift. This is the strongest single piece of evidence against a strong genetic interpretation of the cross-cultural cognition findings, and it is the reason the literature treats the differences as learned cognitive styles shaped by cultural environment.

Exercise 6 (hard, short answer).

The Nisbett program has been criticised for over-simplifying the philosophical traditions it invokes — treating "ancient Greek philosophy" and "ancient Chinese philosophy" as monolithic, and ignoring counter-traditions within each. State two specific historical complications that the cultural-history hypothesis must accommodate, and explain how each either weakens or refines the hypothesis.

Hint

Consider: (i) the place of dialectical reasoning in Western thought (Heraclitus, Hegel, Marx); (ii) the place of formal logic in non-Western thought (the Indian Nyaya school, the Chinese Moist Later Mohist canon); (iii) the fact that substantial within-region variation exists in both traditions; (iv) the fact that the modern East Asian populations studied are not the inheritors of a direct philosophical transmission but of mediated, modernised, language-reformed traditions.

Answer

Two of the most pointed complications:

First, the existence of a formal-logical tradition outside ancient Greece. The Indian Nyaya school (from Gautama's Nyaya-sutras, c. 2nd century BCE) developed a sophisticated five-step syllogism with explicit rules of inference; the Chinese Later Mohist canon (c. 4th century BCE) developed a logic of names and objects with parallel attention to validity. If formal logic were a uniquely Greek invention, the cultural-history hypothesis would predict that South Asian and East Asian populations lacking Greek contact would show no formal-logical tradition. The empirical record contradicts this prediction. The hypothesis must therefore be refined: what matters for cognitive style is not the existence of a formal-logical tradition but the cultural dominance of one orientation over the other. In Greece the formal-logical strand (Aristotle) became canonical; in China the dialectical-relational strand (Confucius, Laozi) became canonical and the Later Mohist tradition was marginalised.

Second, the mediated transmission of tradition. Modern East Asian populations do not inherit Confucian philosophy through direct unbroken transmission. The Chinese literary tradition was reshaped repeatedly — most dramatically by the May Fourth movement of 1919 and the Cultural Revolution of 1966-76 — and by the language reforms of the twentieth century that standardised Mandarin. Japanese and Korean cultures absorbed Confucian influence through mediated transmission and through local syntheses (Shinto-Buddhist-Confucian in Japan; Shaman-Buddhist-Confucian in Korea). The cultural-history hypothesis must therefore claim that what is transmitted is not a fixed philosophical doctrine but a deeper orientation — toward context, relationship, and dialectical balance — that survives the surface-level restructuring of the philosophical tradition. This refinement weakens the strong version of the hypothesis (direct philosophical causation) but preserves a weaker, more defensible version (long-term cultural orientation shaping cognitive style).

The two complications together refine rather than refute the cultural-history hypothesis. The original Nisbett formulation over-simplified; the contemporary literature treats the hypothesis as a candidate explanation requiring the kind of historical detail that Lloyd (1990, Demystifying Mentalities) and Lloyd and Sivin (2002, The Way and the Word) supply.

Exercise 7 (hard, symbolic).

Suppose the holistic-analytic cognitive style is modelled as a latent variable $θ \in R$ , with population distributions $θ \sim N (μ_{A}, σ^{2})$ in a Western population $A$ and $θ \sim N (μ_{E}, σ^{2})$ in an East Asian population $E$ , where higher $θ$ corresponds to more analytic style. A participant is classified as analytic if $θ > c$ for a threshold $c$ . Write an expression for the difference in the proportion classified as analytic between populations $A$ and $E$ in terms of the standard normal cumulative distribution function $Φ$ , and derive the condition on $(μ_{A} - μ_{E}) / σ$ under which the between-population difference in classification rate is less than $0.1$ when $c = μ_{A}$ .

Hint

If $θ \sim N (μ, σ^{2})$ , then $P (θ > c) = 1 - Φ ((c - μ) / σ) = Φ ((μ - c) / σ)$ . With $c = μ_{A}$ , evaluate $P (θ > μ_{A})$ for each population.

Answer

The condition is $δ < 0.253$ . The derivation proceeds in three steps.

Step 1. Under the model, the proportion classified analytic in population $A$ is

p_{A} = P (θ_{A} > c) = 1 - Φ (\frac{c - μ _{A}}{σ}),

and in population $E$ is

p_{E} = P (θ_{E} > c) = 1 - Φ (\frac{c - μ _{E}}{σ}) .

Step 2. The difference is

p_{A} - p_{E} = Φ (\frac{c - μ _{E}}{σ}) - Φ (\frac{c - μ _{A}}{σ}) .

With $c = μ_{A}$ :

p_{A} = 1 - Φ (0) = 0.5, p_{E} = 1 - Φ (\frac{μ _{A} - μ _{E}}{σ}) = 1 - Φ (δ) .

Step 3. Combining and solving. The condition $p_{A} - p_{E} < 0.1$ becomes $Φ (δ) - 0.5 < 0.1$ , that is, $Φ (δ) < 0.6$ . The $60$ th percentile of the standard normal is approximately $δ^{*} \approx 0.253$ . The between-population classification difference is therefore less than $0.1$ whenever

δ = \frac{μ _{A} - μ _{E}}{σ} < 0.253.

Interpretation: even when the between-population mean difference in the latent cognitive-style variable is a quarter of a standard deviation, the proportion classified analytic in the two populations differs by less than ten percentage points (with the threshold at the Western mean). The model formalises why within-population variation dominates between-population difference: the classification rates overlap heavily. At $δ = 0.8$ (a large effect by Cohen's conventions, of the magnitude originally reported in some Nisbett-program paradigms), the classification difference is $Φ (0.8) - 0.5 \approx 0.288$ , a substantial but far from deterministic gap.

Exercise 8 (hard, short answer).

A critic of the cross-cultural cognition program argues: "Most of the published effect sizes in the original Nisbett-program studies were in the $d = 0.5$ to $d = 0.8$ range, while the post-2011 multi-site replications have typically reported $d = 0.2$ to $d = 0.4$ . The cross-cultural cognition program is therefore refuted." State precisely what the multi-site replication results do and do not show, and identify the strongest defensible conclusion the program can sustain.

Hint

Consider: (i) the difference between an effect being smaller than originally reported and an effect being absent; (ii) the heterogeneity of replication results across paradigms (eye-tracking versus priming versus reasoning); (iii) the role of measurement-invariance failures and sample differences in moderating effect sizes; (iv) what counts as a "defensible conclusion" in a field where within-population variation exceeds between-population mean differences.

Answer

The multi-site replication results show three things. First, the original effect-size estimates from small-sample single-site studies were inflated by the standard small-study biases (publication bias, $p$ -hacking on outcome selection, effect-size selection from multiple comparisons). The reduction from $d = 0.5$ to $d = 0.8$ down to $d = 0.2$ to $0.4$ is broadly consistent with the reduction seen across psychological subfields after 2011. Second, the effect sizes vary heterogeneously across paradigms: the Masuda-Nisbett eye-tracking result has replicated relatively well in pre-registered multi-site studies, while specific cultural-priming effects have shown the steepest declines. Third, the within-population variation in cognitive style remains substantially larger than the between-population mean difference, so that even at $d = 0.3$ the population overlap (measured by the probabilistic superiority statistic) is on the order of $0.55$ to $0.60$ .

The replication results do not show that the holistic-analytic distinction is absent, that there is no cross-cultural variation in cognitive style, or that the Nisbett program's central constructs are empty. The reduction is in magnitude, not in presence.

The strongest defensible conclusion is therefore the following. The broad pattern — that East Asian samples on average score more holistically and Western samples on average more analytically, across multiple operationalisations in attention, categorisation, attribution, and reasoning — replicates at modest effect sizes. The original strong version of the program (large, robust, near-universal differences) is not supported. The moderate version (real population-mean differences in cognitive style, with substantial within-population variation, susceptible to context and priming, requiring diverse sampling and measurement-invariance testing before extrapolation) is supported. The program is a useful descriptive framework for population-level cognitive-style research, but it cannot license per-individual predictions, and it must be qualified by the WEIRD critique at every extrapolation to non-sampled populations.

Interpretive debates and developments Master

Result 1 (Markus and Kitayama 1991: the foundational self-construal framework). Hazel Markus and Shinobu Kitayama's Psychological Review paper ^{[MarkusKitayama1991]} introduced the distinction between the independent self-construal (normative in Western populations: the self is autonomous, stable, defined by internal traits) and the interdependent self-construal (normative in East Asian populations: the self is relational, context-sensitive, defined by social roles). The framework synthesised a large body of cross-cultural research on cognition, emotion, and motivation under a single construct and provided the conceptual scaffolding for the cross-cultural cognition program that followed. The paper is among the most-cited in the history of psychology. Its enduring contribution is the move from describing cross-cultural differences as anomalies to be explained away, to treating them as data that constrain theories of the self.

Result 2 (Nisbett, Peng, Choi, Norenzayan 2001: the programmatic synthesis). The 2001 Psychological Review paper ^{[NisbettPengChoiNorenzayan2001]} consolidated a decade of work at Michigan into a single programmatic statement: East Asian populations on average exhibit a holistic cognitive style, Western populations on average exhibit an analytic cognitive style, and the differences manifest across the four domains of attention, categorisation, causal attribution, and reasoning. The paper formalised the cultural-history hypothesis as the mediating account, naming ancient Greek philosophy (Aristotle, Plato, the Ionian tradition) and ancient Chinese philosophy (Confucius, Laozi, Mencius) as the long-term sources of the divergence. The paper is the load-bearing reference of the present unit.

Result 3 (Masuda and Nisbett 2001: attention to context). The eye-tracking paradigm ^{[MasudaNisbett2001]} showed that Japanese and American participants allocate visual attention differently across complex scenes, with Japanese attending more to background context and Americans to focal objects. The result is the most robust single finding of the program, having replicated in multiple laboratories with the basic pattern intact, and is the paradigmatic demonstration that the holistic-analytic contrast operates at the level of perception rather than only at the level of explicit reasoning.

Result 4 (Kitayama, Duffy, Kawamura, Larsen 2003: the framed-line task). The framed-line task gave participants a square containing a line and asked them to reproduce the line either in absolute length (analytic, context-independent) or in length relative to the surrounding square (holistic, context-dependent). Japanese performed better on the relative task; Americans on the absolute task. The result complements the Masuda-Nisbett eye-tracking finding by isolating the perceptual judgement from object recognition.

Result 5 (Ji, Nisbett, Zhang 2004: categorisation by family resemblance vs rules). The triad task ^{[JiNisbettZhang2004]} showed that East Asian participants on average prefer thematic or family-resemblance groupings, while Western participants on average prefer taxonomic or rule-based groupings. The result is one of the cleanest operationalisations of the holistic-analytic distinction, because the same triads are interpreted by both populations, the criterion differences are explicit, and the instructions are held constant.

Result 6 (Choi, Nisbett, Norenzayan 1999: the attenuation of the fundamental attribution error). The cross-cultural extension of the fundamental-attribution-error paradigm ^{[ChoiNisbettNorenzayan1999]} showed that the dispositional bias is present in both East Asian and Western samples but is attenuated in the East Asian sample, where situational-contextual attributions receive more weight. The result refines rather than refutes the original Ross (1977) finding: the fundamental attribution error is a real cognitive bias, but its magnitude is culturally moderated.

Result 7 (Peng and Nisbett 1999: dialectical vs formal-logical reasoning). The reasoning paradigm ^{[PengNisbett1999]} showed that Chinese participants on average prefer dialectical resolutions of apparent contradiction (acknowledging partial truth in both sides), while American participants on average prefer formal-logical resolutions (determining which side is correct). The result is the most contested of the program, because the content of the contradictions has been shown to influence the cross-cultural difference, and because the operationalisation of "dialectical" is harder to validate than the operationalisations of attention or categorisation.

Result 8 (Hong, Morris, Chiu, Benet-Martinez 2000: bicultural frame-switching). The bicultural-priming paradigm showed that Hong Kong Chinese biculturals shift toward holistic style under Chinese primes and toward analytic style under Western primes within a single session. The result is decisive against a strong genetic account of the holistic-analytic difference: the same genome is present in both conditions, and only the activated cultural frame differs. The result also licenses the dynamic conception of culture, in which cultural style is a learnable frame rather than a fixed attribute.

Result 9 (Henrich, Heine, Norenzayan 2010: the WEIRD critique). The Nature paper ^{[HenrichHeineNorenzayan2010]} and the extended Behavioral and Brain Sciences target article showed that WEIRD samples — and especially American undergraduates — are statistical outliers on many psychological dimensions, including visual perception, fairness, cooperation, self-concept, and analytic-holistic style. The critique is not a refutation of cross-cultural research; it is the demand that the field treat WEIRD samples as one data point among many rather than as the universal baseline. The 2020 Henrich monograph The WEIRDest People in the World extends the critique into a multi-millennial account of how Western populations became psychologically peculiar through institutional and demographic changes (the Catholic Church's marriage-and-family-kinship policies, the rise of markets, the spread of literacy, the Reformation).

Result 10 (Nisbett 2003 and the modern debate: The Geography of Thought). Nisbett's 2003 monograph ^{[Nisbett2003]} synthesised the Michigan program for a general audience and became the most-cited public face of the cross-cultural cognition program. The book has been criticised for over-simplifying both philosophical traditions (Lloyd 1990, Demystifying Mentalities; Lloyd and Sivin 2002, The Way and the Word), for overstating effect sizes on the basis of small-sample single-site studies, and for underplaying the heterogeneity within East Asian and within Western populations. The contemporary literature (Norenzayan et al. 2007; the 2010s multi-site replications) treats the broad pattern as robust at modest effect sizes, with specific paradigms remaining contested and with the WEIRD critique as a permanent methodological qualification.

Result 11 (Indigenous psychology: Kim, Berry, Greenfield, Enriquez, Sinha, Hwang). The indigenous psychology movement (Kim and Berry 1993, Indigenous Psychologies) develops psychology from within non-Western cultural traditions rather than importing Western constructs. Filipino Sikolohiyang Pilipino (Enriquez 1992) recovers kapwa (shared inner self) and pakikiramdam (empathic sensing); Indian Indian Psychology (Sinha 1997; Rao 2014) recovers constructs from Yoga, Vedanta, and Buddhism; Chinese indigenous psychology (Hwang 2012; Yang 1997) recovers ren, yuan, mianzi, and guanxi. These frameworks are not refusals of method; they are calls for methodological pluralism, demanding that the constructs under study be validated within the cultural tradition under investigation rather than imposed from without. The movement complements the cross-cultural program by supplying local constructs that the cross-cultural comparison then tests.

Synthesis. The foundational reason that the cross-cultural cognition program has reshaped cognitive psychology over the past three decades is that it operationalised the holistic-analytic distinction across four cognitive domains and measured it on diverse samples; the central insight is that cognitive style is a population-level average over substantial within-population variation, with the same individual capable of frame-switching within minutes under cultural priming, and this is exactly the picture that identifies the East Asian average with holistic attention-to-context and the Western average with analytic attention-to-focal-objects. Putting these together with the bicultural frame-switching evidence, the WEIRD critique, and the measurement-invariance constraints, the bridge is between the Markus-Kitayama self-construal framework of 1991 and the Nisbett et al. cognitive-style program of 2001, with the cultural-history hypothesis supplying the mediating account and the indigenous-psychology movement supplying the local-construct validation that the cross-cultural program presupposes. The pattern recurs across every domain in which a population-level psychological average has been mistaken for an individual-level destiny, and the pattern generalises to the broader principle that within-group variation is the load-bearing statistical fact in any cross-cultural comparison: the broad holistic-analytic pattern replicates at modest effect sizes; specific paradigms remain in active debate; and the WEIRD critique sharpens the demand for diverse sampling without negating the cross-cultural program.

Full argument set Master

Proposition 1 (Frame-switchability entails that the proximate cause of holistic-analytic cognitive-style differences is environmental, not genetic). If a single individual participant can be induced to exhibit a holistic cognitive style in one experimental condition and an analytic cognitive style in another condition within the same session, then the proximate cause of the cross-population difference cannot be a fixed genetic attribute, even if genetic variation contributes to within-population variation in baseline cognitive style.

Proof. Let $θ_{i}$ denote the cognitive-style value of participant $i$ on a holistic-to-analytic scale. Let $G_{i}$ denote the participant's genome, which is fixed across the session. Let $P_{t} \in {H, W}$ denote the cultural prime administered at trial $t$ (Chinese holistic prime or Western analytic prime). The bicultural-priming finding (Hong, Morris, Chiu, Benet-Martinez 2000) is that, for a bicultural participant $i$ ,

E [θ_{i} ∣ P_{t} = H] < E [θ_{i} ∣ P_{t} = W],

with the gap substantive in magnitude and statistically reliable.

Suppose, for contradiction, that the proximate cause of $θ$ is the fixed genome: $θ_{i} = f (G_{i})$ for some function $f$ independent of $P_{t}$ . Then $E [θ_{i} ∣ P_{t} = H] = E [θ_{i} ∣ P_{t} = W] = f (G_{i})$ , since the conditioning event carries no information about $G_{i}$ and $f$ does not depend on $P_{t}$ . This contradicts the bicultural-priming finding. Therefore the genome cannot be the proximate cause of the cross-priming difference.

The conclusion is not that genetics is irrelevant to cognitive style. Within-population heritability of cognitive-style dispositions may be non-zero, and gene-environment interaction may modulate the priming response. The conclusion is the narrower claim: the proximate cause of the between-population difference in cognitive style is the cultural environment, including the activated cultural frame, not a fixed genetic difference between populations. This is the formal content of the cultural-environmental interpretation of the cross-cultural cognition findings. $□$

Proposition 2 (Within-population variation exceeds between-population mean difference for the holistic-analytic cognitive style). Let $θ \sim N (μ_{A}, σ^{2})$ denote the cognitive-style distribution in a Western population $A$ and $θ \sim N (μ_{E}, σ^{2})$ the distribution in an East Asian population $E$ , with shared standard deviation $σ$ and Cohen's $d = ∣ μ_{A} - μ_{E} ∣/ σ < 1$ . Then the proportion of East Asian participants scoring above the Western mean exceeds $Φ (- d)$ , and for $d \leq 0.5$ this proportion exceeds $0.30$ .

Proof. Without loss of generality assume $μ_{A} > μ_{E}$ (Western mean more analytic). The proportion of East Asian participants with $θ$ above the Western mean $μ_{A}$ is

P (θ_{E} > μ_{A}) = P (\frac{θ _{E} - μ _{E}}{σ} > \frac{μ _{A} - μ _{E}}{σ}) = 1 - Φ (d) = Φ (- d) .

For $d = 0.5$ , $Φ (- 0.5) \approx 0.309$ , so that approximately $31%$ of East Asian participants score above the Western mean. For $d = 0.3$ (the typical post-2011 multi-site replication magnitude), $Φ (- 0.3) \approx 0.382$ , so that approximately $38%$ of East Asian participants score above the Western mean. The corresponding proportions for $d = 0.7$ and $d = 0.8$ are approximately $0.242$ and $0.212$ respectively — still far from zero.

The implication is that the holistic-analytic population difference is a difference in distribution means, not a difference in the ranges of cognitive styles present in the two populations. The two distributions overlap heavily. Any individual-level prediction based on population membership is therefore weak: the probabilistic superiority (the probability that a randomly sampled Westerner scores higher than a randomly sampled East Asianer) is $Φ (d / 2)$ , which for $d = 0.5$ is approximately $0.638$ . The cross-cultural cognition findings describe population-level regularities; they do not license per-indual classification.

This is the formal content of the within-population-variation caveat that runs through every cross-cultural cognition study. The broad pattern replicates; the population overlap is large; per-individual inference from population membership is invalid. $□$

Connections Master

Cross-cultural and indigenous psychology survey 29.12.01. The present unit is the chapter-closing depth companion to the cross-cultural survey. Where 29.12.01 introduces the full sweep of cross-cultural and indigenous psychology — WEIRD critique, individualism-collectivism, cultural-dimension models, Markus-Kitayama self-construal, acculturation, indigenous psychologies, decolonial frameworks — the present unit narrows to the cognitive-style program specifically, developing the Nisbett synthesis at the depth that the survey can only sketch. The survey is the prerequisite; the present unit is the deeper cut.
Working memory and the Baddeley-Hitch model 29.05.04. The central-executive component of working memory is the cognitive-system locus at which attention is allocated between focal-object information and contextual information. The holistic-analytic contrast is, at the cognitive-system level, a population-level signature of how that central-executive allocation differs. The Kitayama-Duffy framed-line task and the Masuda-Nisbett eye-tracking paradigm both measure, in different modalities, how central-executive attention is divided across the visual field; the population difference in that allocation is the cognitive-system content of the present unit's findings.
Intelligence-testing history and the cultural-loading critique 29.13.04. The Yerkes Army Alpha/Beta programme of 1917-21, the Binet-Terman hereditarian misreading, and the Jensen-Herrnstein-Murray heritability controversy of the post-1969 period are the methodological cautionary tales that the WEIRD critique sharpens into a systematic methodological demand. The cultural loading of the Army instruments is exactly the kind of construct-validity failure that cross-cultural cognition research must avoid; the hereditarian misreading of group differences is exactly the within-versus-between conflation that the present unit's Proposition 1 and Proposition 2 formalise. The two units converge on the methodological lesson that population averages and individual destinies require distinct evidence.
Cognition and intelligence survey 29.05.01. The cognitive-frame companion to the present unit. Where the present unit treats holistic-analytic style as a population-level construct, 29.05.01 treats cognition and intelligence as individual-differences constructs. The two units converge on the construct of fluid intelligence ( $Gf$ ), which is what holistic-analytic differences in attention-to-context modulate at the cognitive level: the focal-object attention characteristic of analytic style is what the standard $Gf$ tasks (Raven's Matrices, series completion) reward, while the context-sensitivity characteristic of holistic style is what they fail to capture.

Historical & philosophical context Master

The lineage of cross-cultural cognition runs from discredited precursors through methodological reforms to the contemporary program. Lucien Lévy-Bruhl's 1910 Les fonctions mentales dans les sociétés inférieures ^{[LevyBruhl1910]} distinguished a "primitive mentality" governed by participation-mystique from Western rational thought; the framework is discredited and is mentioned here only to mark the historical low ground from which the modern program had to climb. Lev Vygotsky's 1934 Myshlenie i rech ^{[Vygotsky1934]} founded cultural-historical psychology, treating higher cognitive functions as products of social and cultural mediation rather than as biological fixed attributes; the Vygotskian inheritance is the conceptual scaffolding on which the modern cross-cultural program rests, even where it is not cited explicitly.

The mid-century psychometric tradition supplied the first formal operationalisations. Herman Witkin's work on field-dependence and field-independence (Witkin et al. 1962 ^[Witkin1962]) showed that individuals differ reliably in the extent to which their perception of an embedded figure is dominated by the surrounding field; the construct was individual-difference rather than cross-cultural in its original formulation, but John Berry's ecocultural model (Berry 1970 ^[Berry1970]) extended it to cross-cultural comparison, predicting that field-dependence would be higher in agricultural, sedentary, tightly-structured societies and lower in nomadic, hunting-and-gathering societies. The Berry model is the first cross-cultural cognitive-style framework to operationalise the ecology-cognition link.

The programmatic modern synthesis was the joint work of Hazel Markus and Shinobu Kitayama (1991) and Richard Nisbett, Kaiping Peng, Incheol Choi, and Ara Norenzayan (2001). Markus and Kitayama's Psychological Review paper ^{[MarkusKitayama1991]} introduced the independent-interdependent self-construal distinction that supplies the social-psychological construct under which the cognitive-style differences are organised. Nisbett, Peng, Choi, and Norenzayan's 2001 Psychological Review paper ^{[NisbettPengChoiNorenzayan2001]} consolidated a decade of Michigan experimental work into the holistic-analytic cognitive-style program, with the cultural-history hypothesis as the mediating account. Masuda and Nisbett (2001) ^{[MasudaNisbett2001]} supplied the eye-tracking paradigm that is the program's most-cited single demonstration; Ji, Nisbett, Zhang (2004) ^{[JiNisbettZhang2004]} the categorisation triads; Choi, Nisbett, Norenzayan (1999) ^{[ChoiNisbettNorenzayan1999]} the cross-cultural extension of the fundamental attribution error; Peng and Nisbett (1999) ^{[PengNisbett1999]} the dialectical-versus-formal-logical reasoning paradigm. Nisbett's 2003 monograph The Geography of Thought ^{[Nisbett2003]} brought the program to a general audience.

The methodological reform of the program was driven by the WEIRD critique. Joseph Henrich, Steven Heine, and Ara Norenzayan's 2010 Nature summary and the extended Behavioral and Brain Sciences target article ^{[HenrichHeineNorenzayan2010]} showed that WEIRD samples are statistical outliers on many psychological dimensions, including the holistic-analytic style. Henrich's 2020 monograph The WEIRDest People in the World ^{[Henrich2020]} extended the critique into a multi-millennial account of how Western populations became psychologically peculiar through the institutional and demographic changes set in motion by the Catholic Church's marriage-and-family-kinship policies, the rise of markets, and the spread of literacy. The contemporary program is the synthesis of the Markus-Kitayama-Nisbett theoretical framework with the WEIRD-aware methodological reforms of the post-2011 era.

Bibliography Master

Berry, John W. "Ecological and Cultural Factors in Spatial Perceptual Development." Canadian Journal of Behavioural Science 2, no. 3 (1970): 196–206.
Choi, Incheol, Richard E. Nisbett, and Ara Norenzayan. "Causal Attribution across Cultures: Variation and Universality." Psychological Review 106, no. 4 (1999): 747–763.
Enriquez, Virgilio G. From Colonial to Liberation Psychology: The Philippine Experience. Quezon City: University of the Philippines Press, 1992.
Henrich, Joseph. The WEIRDest People in the World: How the West Became Psychologically Peculiar and Particularly Prosperous. New York: Farrar, Straus and Giroux, 2020.
Henrich, Joseph, Steven J. Heine, and Ara Norenzayan. "The Weirdest People in the World?" Behavioral and Brain Sciences 33, no. 2-3 (2010): 61–83. Summary in Nature 466 (2010): 29.
Hong, Ying-yi, Michael W. Morris, Chi-yue Chiu, and Veronica Benet-Martínez. "Multicultural Minds: A Dynamic Constructivist Approach to Culture and Cognition." American Psychologist 55, no. 7 (2000): 709–720.
Hwang, Kwang-kuo. Foundations of Chinese Psychology: Confucian Social Relations. New York: Springer, 2012.
Ji, Li-Jun, Richard E. Nisbett, and Zhiyong Zhang. "Is It Culture or Is It Language? Examination of Language Effects in Cross-Cultural Research on Categorization." Journal of Personality and Social Psychology 87, no. 1 (2004): 57–65.
Kitayama, Shinobu, Sean Duffy, Tadashi Kawamura, and Jeff T. Larsen. "Perceiving an Object and Its Context in Different Cultures: A Cultural Look at New Look." Journal of Personality and Social Psychology 85, no. 5 (2003): 893–913.
Lévy-Bruhl, Lucien. Les fonctions mentales dans les sociétés inférieures. Paris: Félix Alcan, 1910. (Discredited precursor.)
Lloyd, Geoffrey E. R. Demystifying Mentalities. Cambridge: Cambridge University Press, 1990.
Lloyd, Geoffrey E. R., and Nathan Sivin. The Way and the Word: Science and Medicine in Early China and Greece. New Haven: Yale University Press, 2002.
Markus, Hazel R., and Shinobu Kitayama. "Culture and the Self: Implications for Cognition, Emotion, and Motivation." Psychological Review 98, no. 2 (1991): 224–253.
Masuda, Takahiko, and Richard E. Nisbett. "Attending Holistically versus Analytically: Comparing the Context Sensitivity of Japanese and Americans." Journal of Personality and Social Psychology 81, no. 5 (2001): 925–934.
Nisbett, Richard E. The Geography of Thought: How Asians and Westerners Think Differently...and Why. New York: Free Press, 2003.
Nisbett, Richard E., Kaiping Peng, Incheol Choi, and Ara Norenzayan. "Culture and Systems of Thought: Holistic versus Analytic Cognition." Psychological Review 108, no. 2 (2001): 291–310. Target article in Science 291, no. 5502 (2001): 188–189.
Norenzayan, Ara, Richard E. Nisbett, Edward E. Smith, Beom Jun Kim, and Solomon Nof. "A Cross-Cultural Investigation of the Role of Theoretical Frameworks in Categorisation." Perspectives on Psychological Science 2, no. 3 (2007): 270–283.
Peng, Kaiping, and Richard E. Nisbett. "Culture, Dialectics, and Reasoning about Contradiction." American Psychologist 54, no. 9 (1999): 741–754.
Sinha, Durganand. "Indigenizing Psychology." In Handbook of Cross-Cultural Psychology, 2nd ed., vol. 1, edited by John W. Berry, Ype H. Poortinga, and Janak Pandey, 129–169. Boston: Allyn & Bacon, 1997.
Vygotsky, Lev S. Myshlenie i rech [Thought and Language]. Moscow: Sotsekgiz, 1934.
Witkin, Herman A., Ruth B. Dyk, Herbert F. Faterson, Donald R. Goodenough, and Stephen A. Karp. Psychological Differentiation: Studies of Development. New York: Wiley, 1962.

Prerequisites

29.12.01

Tier anchors

beginner: Erin Meyer — The Culture Map (PublicAffairs, 2014); Joseph Henrich — The WEIRDest People in the World (Farrar, Straus and Giroux, 2020), Parts I-II
intermediate: Nisbett, R. E. — The Geography of Thought (Free Press, 2003); Heine, S. J. — Cultural Psychology, 4th ed. (W. W. Norton, 2020), Chs. 3-6
master: Markus and Kitayama 1991 Psych. Rev. 98:224; Nisbett, Peng, Choi, Norenzayan 2001 Science; Masuda and Nisbett 2001; Ji, Nisbett, Zhang 2004; Choi, Nisbett, Norenzayan 2000; Peng and Nisbett 1999; Norenzayan et al. 2007; Henrich, Heine, Norenzayan 2010 Nature 466:29; Henrich 2020 The WEIRDest People in the World

References

Markus, H. R., and Kitayama, S. — Culture and the Self: Implications for Cognition, Emotion, and Motivation, Psychological Review 98, 224-253 (1991) · the foundational statement of independent vs interdependent self-construals
Nisbett, R. E., Peng, K., Choi, I., and Norenzayan, A. — Culture and Systems of Thought: Holistic versus Analytic Cognition, Psychological Review 108, 291-310 (2001) · the programmatic statement and synthesis of the cross-cultural cognition program
Masuda, T., and Nisbett, R. E. — Attending Holistically versus Analytically: Comparing the Context Sensitivity of Japanese and Americans, Journal of Personality and Social Psychology 81, 925-934 (2001) · the underwater-scene eye-tracking paradigm
Henrich, J., Heine, S. J., and Norenzayan, A. — The Weirdest People in the World?, Behavioral and Brain Sciences 33, 61-83 (2010); summary in Nature 466:29 (2010) · the WEIRD critique
Ji, L.-J., Nisbett, R. E., and Zhang, Z. — Is It Culture or Is It Language? Examination of Language Effects in Cross-Cultural Research on Categorization, Journal of Personality and Social Psychology 87, 57-65 (2004) · family-resemblance vs rule-based categorisation
Choi, I., Nisbett, R. E., and Norenzayan, A. — Causal Attribution across Cultures: Variation and Universality, Psychological Review 106, 747-763 (1999) · the attenuation of the fundamental attribution error in East Asian samples
Peng, K., and Nisbett, R. E. — Culture, Dialectics, and Reasoning about Contradiction, American Psychologist 54, 741-754 (1999) · dialectical vs formal-logical reasoning
Nisbett, R. E. — The Geography of Thought: How Asians and Westerners Think Differently...and Why (Free Press, 2003) · the synthetic monograph
Henrich, J. — The WEIRDest People in the World: How the West Became Psychologically Peculiar and Particularly Prosperous (Farrar, Straus and Giroux, 2020) · the extended WEIRD synthesis
Witkin, H. A., Dyk, R. B., Faterson, H. F., Goodenough, D. R., and Karp, S. A. — Psychological Differentiation (Wiley, 1962) · field-dependence/independence as a precursor
Berry, J. W. — Ecological and Cultural Factors in Spatial Perceptual Development, Canadian Journal of Behavioural Science 2, 196-206 (1970) · the ecocultural model

Estimated time

beginner: 25m
intermediate: 50m
master: 85m