Researchers must infer “what babies know” based on what babies do. Thus, to maximize information from doing, researchers should use tasks and tools that capture the richness of infants' behaviors. We clarify Gibson's views about the richness of infants' behavior and their exploration in the service of guiding action – what Gibson called “learning about affordances.”

Why have core knowledge? Standard answers typically emphasize the difficulty of learning core knowledge from experience, or the benefits it confers for learning about the world. Here, we suggest a complementary reason: Core knowledge is critical for learning not just about the external world, but about the mind itself.

What Babies Know (WBK) argues that core knowledge has a unique place in cognitive architecture, between fully perceptual and fully conceptual systems of representation. Here I argue that WBK's core knowledge is on the perception side of the perception/cognition divide. I discuss some implications of this conclusion for the roles language learning might play in transcending core knowledge.

What language devises, it might divide. By exploring the relations among the core geometries of the physical world, the abstract geometry of Euclid, and language, I give new insight into both the persistence of core knowledge into adulthood and our access to it through language. My extension of Spelke's language argument has implications for pedagogy, philosophy, and artificial intelligence.

Spelke convincingly argues that we should posit six innate modular systems beyond the periphery (i.e., beyond low-level perception and motor control). I focus on the case of spatial navigation (Ch. 3) to claim that there remain powerful considerations in favor of positing additional innate, nonperipheral modules. This opens the door to stronger forms of nativism and nonperipheral modularism than Spelke's.

Decades of research have pushed us closer to understanding what babies know. However, a powerful approach – representational similarity analysis (RSA) – is underused in developmental research. I discuss the strengths of this approach and what it can tell us about infant conceptual knowledge. As a case study, I focus on numerosity as a domain where RSA can make unique progress.

I focus here on concepts that are not part of core knowledge – the ability to treat people as social agents with shareable mental states. Spelke proposes that learning language from another might account for the development of these concepts. I suggest that homesigners, who create language rather than learn it, may be a potential counterexample to this hypothesis.

This commentary challenges Spelke's view on the early development of social cognition from a neuroscience perspective by presenting an overlooked body of evidence from neuroimaging research on joint attention with human infants. Indeed, evidence demonstrating adult-like, neural sensitivity to joint attention in young infants, supports alternative theoretical views concerning the origins of uniquely human forms of social cognition.

What we know about what babies know – as represented by the core knowledge proposal – is perhaps missing a place for the baby itself. By studying the baby as an actor rather than an observer, we can better understand the origins of human intelligence as an interface between perception and action, and how humans think and learn about themselves in a complex world.

Spelke's What Babies Know masterfully describes infants’ impressive repertoire of core cognitive concepts, from which the suite of human knowledge is eventually built. The current commentary argues for the existence of a core concept that Spelke claims preverbal infants lack: social goal. Core social goal concepts, operative extremely early in human development, underlie infants’ basic abilities to interpret and evaluate entities within the moral world; such abilities support claims for a core moral domain.

Central to What Babies Know (Spelke, 2022) is the thesis that infants' understanding is divided into independent modules of core knowledge. As a test case, we consider adding a new domain: core knowledge of substances. Experiments show that infants' understanding of substances meets some criteria of core knowledge, and they raise questions about the relations that hold between core domains.

We argue that the core social being system is unlike other core systems in that it participates in frequent, widespread learning. As a result, the social being system is less constant throughout the lifespan and less informationally encapsulated than other core systems. This learning supports the development of the precursors of bias, but also provides avenues for preempting it.

We examine Spelke's core knowledge taxonomy and test its boundaries. We ask whether Spelke's core knowledge is a distinct type of cognition in the sense that the cognitive processes it includes and excludes are biologically and mechanically coherent.

Somewhat questioning Elizabeth Spelke's attempt to account for infants’ social knowledge, our commentary argues that social cognition might be divided into several specialized systems. In addition to the core system dedicated to the intersubjective dimension of close relationships, infants could be prewired to process social relationships, such as dominance, characterized by their impersonal, normative dimension.

We question Spelke's key claim that the medium, in which contents from different core knowledge systems can be represented and combined, is language-based. Recalling an episodic memory, playing chess, and conducting mental rotation are tasks where core knowledge information is represented and combined. Although these tasks can be described by means of language, these tasks are not inherently language-based. Hence, language may be an important subset of an abstraction medium – not the medium as such.

Core knowledge systems play an important role in theories of cognitive development. However, recent studies suggest that fundamental principles of the object and agent systems can be revised by adults and preschoolers, when given small amounts of counterevidence. We argue that not all core knowledge systems are created equal, and they may be subject to revision throughout development.

While Spelke provides powerful support for concept nativism, her focus on understanding concept nativism through six innate core knowledge systems is too confining. There is also no reason to suppose that the curse of a compositional mind constitutes a principled reason for positing less innate structure in explaining the origins of concepts. Any solution to such problems must take into account poverty of the stimulus considerations, which argue for postulating more innate structure, not less.

Elizabeth Spelke's What Babies Know is a scholarly presentation of core knowledge theory and a masterful compendium of empirical evidence that supports it. Unfortunately, Spelke's principal theoretical assumption is that core knowledge is simply the innate product of cognitive evolution. As such, her theory fails to explicate the developmental mechanisms underlying the emergence of the cognitive systems on which that knowledge depends.

What Babies Know outlines a compelling case for why infancy research is fundamental for conceptualizing what it is to be human. There is another period in human development that is relatively inaccessible, yet is more important. In order to truly understand the nature of core knowledge, perception, and cognition, we must start not with the infant, but with the fetus.

Some core knowledge may be rooted in – or even identical to – well-characterized mechanisms of mid-level visual perception and attention. In the decades since it was first proposed, this possibility has inspired (and has been supported by) several discoveries in both infant cognition and adult perception, but it also faces several challenges. To what degree does What Babies Know reflect how babies see and attend?