Moving beyond distinguishing knowledge and beliefs, we propose two lines of inquiry for the next generation of theory of mind (ToM) research: (1) characterizing the contents of different mental-state representations and (2) formalizing the computations that generate such contents. Studying how children reason about what others think of the self provides an illuminating window into the richness and flexibility of human social cognition.

Supporting the central claim that knowledge representation is more basic than belief representation, we focus on the emerging evidence for preverbal infants' active and selective communication based on their representation of both knowledge and ignorance. We highlight infants' ontogenetically early deliberate information seeking and information transmission in the context of active social learning, arguing that these capacities are unique to humans.

Phillips and colleagues claim that the representation of knowledge is more basic than the representation of belief, presupposing them to be categorically distinct mental states with distinct evolutionary purposes. We argue that the relationship between the two is much more complex, is further shaped by culture and language, and leaves its mark on manifestations of theory of mind and teaching.

This commentary on the paper “Knowledge before belief” argues that it is not only in the cognitive sciences that knowledge should be separated into a separate category from belief, but also in rational decision theory. It outlines how knowledge-as-commitment – as distinct from knowledge-as-belief – can be built into an extension of the economic theory of revealed preference.

This commentary supports knowledge prior to belief, but from a different angle, supplementing the target article's central thesis. The target article evaluates belief-representations versus knowledge-representation in others. This commentary considers one's own unconscious knowledge, which can be prior to belief of any sort. Two examples are offered: one from clinical psychoanalysis, another involving a cognitive psychology duck/rabbit experiment.

The knowledge-centric theory of mind research program suggested by Phillips et al. stands to gain significant value by embracing a neurocognitive approach that takes full advantage of techniques such as fMRI and EEG. This neurocognitive approach has already begun providing important insights into the mechanisms of knowledge attribution, insights which support the claim that it is more basic than belief attribution.

In arguing for knowledge representation before belief, Phillips et al. presuppose a representational theory of knowledge, a view that has been extensively criticized. As an alternative, we propose an action-based approach to knowledge, conceptualized in terms of skill. We outline the implications of this approach for children's developing social understanding, beginning with sensorimotor interaction and extending to the verbal level.

Autistic, developmental, and nonhuman primate populations fail tasks that are thought to involve attributing beliefs, but not those thought to reflect the representation of knowledge. Instead of knowledge representations being more basic than belief representations, relational mentalizing may explain these observations: The tasks referred to as reflecting “belief” representation, but not the “knowledge” representation tasks, are social conflict designs. They involve mental conflict monitoring after another's mental state is represented – with effects that need to be accounted for.

The authors distinguish knowledge and belief attributions, emphasizing the role of the former in mental-state attribution. This does not, however, warrant diminishing interest in the latter. Knowledge attributions may not entail mental-state attributions or metarepresentations. Even if they do, the proposed features are insufficient to distinguish them from belief attributions, demanding that we first understand each underlying representation.

Building on Phillips and colleagues' case for the primacy of knowledge, we advocate for attention to diversity in mentalizing constructs within, as well as between, knowledge and belief. Ultimately, as great apes and other animals show, the development and evolution of theory of mind may reflect a much greater range of incremental elaborations of representational or computational complexity.

Phillips et al. discuss whether knowledge or beliefs are more basic representations of others' minds, focusing on the primary function of knowledge representation: learning from others. We discuss links between emotion and “knowledge versus belief,” and particularly the role of emotions in learning from others in mechanisms such as “social epistemic emotions” and “affective social learning.”

In this commentary, we argue that Phillips et al.'s findings can be used to provide new important insights in the debate between externalists’ theories of cognition. In particular, we claim that the results presented in this target article may offer us the conceptual palette needed for a sustained defence of an extended account of cognition over an enactive one.

I argue for three points: First, evidence of the primacy of knowledge representation is not evidence of primacy of knowledge. Second, knowledge-oriented mindreading research should also focus on misrepresentations and biased representations of knowledge. Third, knowledge-oriented mindreading research must confront the problem of the gold standard that arises when disagreement about knowledge complicates the interpretation of empirical findings.

The target article presents strong empirical evidence that knowledge is basic. However, it offers an unsatisfactory account of what makes knowledge basic. Some current ideas in cognitive neuroscience – predictive coding and analysis by synthesis – point to a more plausible account that better explains the evidence.

We summarize research and theory to show that, from early in human ontogeny, much information about other minds can be gleaned from reading the eyes. This analysis suggests that eyes serve as uniquely human windows into other minds, which critically extends the target article by drawing attention to what might be considered the neurodevelopmental origins of knowledge attribution in humans.

Phillips and colleagues argue that knowledge representations are more fundamental than belief representations because they better facilitate social learning. We suggest that existing theory of mind paradigms may be ill-equipped to adequately evaluate this claim. Future study should explore learning in situations where there is uncertainty about one's own and others’ knowledge, which better mirror real-world social learning contexts.

We welcome Phillips et al.'s proposal to separate the understanding of “knowledge” from that of “beliefs.” We argue that this distinction is best specified at the level of the cognitive mechanisms. Three distinct mechanisms are discussed: tagging one's own representations with those who share the same reality; representing others' representations (metarepresenting knowledge); and attributing dispositions to provide useful information.

Recent findings from anticipatory-looking false-belief tests have shown that nonhuman great apes and macaques anticipate that an agent will go to the location where the agent falsely believed an object to be. Phillips et al.'s claim that nonhuman primates attribute knowledge but not belief should thus be reconsidered. We propose that both knowledge and belief attributions are evolutionary old.

The knowledge-first approach is attractive and consistent with a wide variety of evidence. So is the opposing belief-first picture. I explain why the target article's criticisms of the latter fail, and argue that the outcome is a stalemate.

Phillips and colleagues claim that the capacity to ascribe knowledge is a “basic” capacity, but most studies reporting linguistic data reviewed by Phillips et al. were conducted in English with American participants – one of more than 6,500 languages currently spoken. We highlight the importance of cross-cultural and cross-linguistic research when one is theorizing about fundamental human representational capacities.