Cultural Intelligence Hypothesis

The cultural intelligence hypothesis is the claim — formalized by Herrmann, Call, Hernández-Lloreda, Hare, and Tomasello in a 2007 Science paper — that humans differ cognitively from our nearest primate relatives not primarily because of higher general intelligence, but because of a species-specific, early-emerging set of social-cognitive skills for learning from and coordinating with other minds. These skills are the evolutionary foundation of cumulative culture — and, by extension, of every knowledge-management practice humans have ever invented, from the commonplace book to the Zettelkasten to the AI-assisted vault.

The Core Claim

Two rival accounts of human cognitive uniqueness:

General intelligence hypothesis. Humans are just better at thinking across the board — more working memory, stronger reasoning, better problem-solving across every domain.
Cultural intelligence hypothesis. Humans and great apes are comparable on physical cognition (understanding objects, space, causality, quantity) but diverge sharply on social cognition (understanding other minds, learning from them, communicating intent).

Herrmann et al.'s 2007 Primate Cognition Test Battery (PCTB) tested both on 243 subjects (106 chimpanzees, 32 orangutans, 105 human children at 2.5 years, pre-literacy and pre-schooling). Top-level results (proportion correct): physical domain — humans 0.68, chimps 0.68, orangutans 0.59; social domain — humans 0.74, chimps 0.36, orangutans 0.33. The social-domain effect size is η² = 0.72 — one of the largest reported in cross-species cognitive research. The pattern supports cultural intelligence and contradicts general intelligence.

The striking sub-findings

Three specific results sharpen the picture:

Social learning collapses the ape species entirely. Humans 0.86; chimps 0.10; orangutans 0.07. The gap is nearly complete, not merely statistically significant.
Chimpanzees beat humans on tool use (0.74 vs 0.23) and on two other physical tasks (transposition, addition numbers). Humans are not globally better even within physical cognition; they win on causal-reasoning tasks that do not require tool manipulation.
The pointing-cups communication task shows no species difference at all (humans 0.72, chimps 0.74, orangutans 0.73). Not all social-cognition tasks produce the dramatic gap; the gap is concentrated in social learning, theory-of-mind, and attentional-state tasks.

The "social-cognitive skills" the hypothesis points to include (as developed across Tomasello's broader corpus):

Shared intentionality — the capacity to jointly attend to something with another mind, knowing that you both know you're attending together
Theory of mind — reasoning about other minds as having beliefs, desires, intentions distinct from one's own
Natural pedagogy — the evolved readiness to teach and to learn-from-being-taught (ostensive cues, pointing, eye contact)
Imitative social learning — high-fidelity copying of what others do, preserving conventions even when they are not strictly optimal
Collaborative communication — language and gesture as collaborative acts requiring mutual understanding of intentions

These are not merely cultural achievements; they are early-emerging, developmentally robust, cross-culturally universal traits of human ontogeny.

Cumulative Culture

The downstream consequence of these skills is cumulative culture: knowledge, practices, and tools that accumulate across generations because humans can learn from others with enough fidelity that each generation starts roughly where the previous left off, then adds marginal improvements. Other apes have cultures (local behavioral variants transmitted by social learning), but they are not cumulative in this sense — they do not ratchet upward over time.

PKM is downstream of cumulative culture in an obvious way: the whole point of note-taking, external memory, commonplace books, and now AI-augmented vaults is to accumulate knowledge — within a life and across lives — in ways that exceed what any individual can hold. If the cultural intelligence hypothesis is right, this accumulation drive is evolved, not learned.

Implications for PKM

The hypothesis reframes several PKM debates:

PKM is not a modern invention fighting against human cognition; it is an amplification of evolved tendencies. The deep history of note-taking from Aristotle's students through Luhmann is continuous with the evolved cognitive architecture, not a break from it. We take notes because we are built to participate in and extend culture.
Solo thinking fights biology. If human cognition is evolved for cultural participation (shared attention, joint intentionality, communication), then the solo-thinker model — write alone, think alone, produce alone — fights against our actual architecture. Practices like public learning, networked thought, writing groups, and third places for thinking work because they restore the social substrate.
Imitation is not a weakness. The hypothesis frames high-fidelity imitation as one of the core adaptations that make cumulative culture possible. Contemporary emphasis on "originality" sometimes disparages progressive summarization, quoting, and faithful reproduction — but these are precisely the skills cultural learning depends on. Originality sits on top of fidelity.
AI as cultural partner (with caveats). Interacting with LLMs activates the same evolved social-learning circuitry. This is both useful (natural engagement, rapid context transfer) and dangerous (sycophancy exploits pedagogical expectations; overreliance bypasses active learning).
Compounding knowledge is the cultural mechanism, in miniature. The same ratchet that operates across generations (cumulative culture) operates across a single knowledge worker's lifetime (compounding vault). The cognitive architecture that makes the cross-generational ratchet possible also makes the intra-life one possible — we are, in a sense, running the cultural algorithm on ourselves over time.

Alternative the authors themselves flag

In the Discussion, Herrmann et al. note that children outperformed apes not only on social-cognition tasks but also on causality tasks that do not involve active tool manipulation (noise, shape, tool properties). This means what is distinctive may be "the ability to understand unobserved causal forces in general, including (as a special case) the mental states of others as causes of behavior." The authors suggest even this generalized causal-reasoning ability likely evolved first for social/mental-state reasoning and only later generalized — leaving cultural intelligence as the upstream driver either way.

Supporting evidence: the dog case

Hare et al. (2002) found that domestic dogs underperform chimps on physical cognition but outperform them on social cognition. Dogs were not evolutionarily selected for cultural cognition but were selected for social interaction with humans. This provides parallel evidence that social cognition can be adaptively elaborated independently of general intelligence.

Since 2007

Supported and extended — Tomasello's A Natural History of Human Thinking (2014) and Becoming Human (2019) develop the shared-intentionality account at length.
Partially competed-with — Heyes's Cognitive Gadgets (2018) argues many of these skills are culturally transmitted rather than genetically specified. If Heyes is right, the gap is still real but its causation is different.
Nuanced — subsequent comparative work finds ape social cognition richer than the 2007 paper suggested, narrowing but not closing the gap.

For PKM purposes, the core insight survives: human cognition is distinctively adapted for participation in cultural groups that accumulate knowledge across time. The fine-grained magnitude of the gap is less important than the directional claim.

Key Points

Cultural intelligence hypothesis: human cognitive uniqueness comes from species-specific social-cognitive skills, not general intelligence
Empirical basis: Herrmann et al. 2007 found 2.5-year-olds ≈ chimps on physical cognition, >> apes on social cognition
Core skills: shared intentionality, theory of mind, natural pedagogy, imitative learning, collaborative communication
These skills enable cumulative culture — knowledge that accumulates across generations
PKM is continuous with this evolved architecture, not a fight against it
The hypothesis has been developed, extended, and partially challenged, but the directional claim is robust

Open Questions

How much of the contemporary solo-PKM paradigm is fighting against the evolved architecture? Should more practices be explicitly social?
Does LLM-augmented knowledge work activate the social-cognition circuitry in useful ways, in misleading ways, or both?
Is the "ratchet" that operates across generations fundamentally the same mechanism as the "compounding" that operates across a single life?
What would a PKM practice designed from the cultural intelligence hypothesis — rather than retrofitted onto it — look like?

References

Herrmann, Call, Hernández-Lloreda, Hare, Tomasello, "Humans Have Evolved Specialized Skills of Social Cognition: The Cultural Intelligence Hypothesis," Science 317 (2007), pp. 1360-1366
Michael Tomasello, A Natural History of Human Thinking (Harvard, 2014)
Michael Tomasello, Becoming Human: A Theory of Ontogeny (Harvard, 2019)
Cecilia Heyes, Cognitive Gadgets: The Cultural Evolution of Thinking (Harvard, 2018) — partially competing account
Joseph Henrich, The Secret of Our Success (Princeton, 2015) — broader cultural-evolution context