Michael Tomasello explains the human sense of obligation by the role it plays in negotiating practices of acting jointly and the commitments they underwrite. He draws in his work on two models of joint action, one from Michael Bratman, the other from Margaret Gilbert. But Bratman's makes the explanation too difficult to succeed, and Gilbert's makes it too easy.

The natural history of our moral stance told here in this commentary reveals the close nexus of morality and basic social-cognitive capacities. Big mysteries about morality thus transform into smaller and more manageable ones. Here, I raise questions regarding the conceptual, ontogenetic, and evolutionary relations of the moral stance to the intentional and group stances and to shared intentionality.

Obligation as defined by Tomasello requires mutually capable parties, but one-sided caregiver relationships reveal its developmental and evolutionary precursors. Specifically, “coercive” emotions may prompt protective action by caregivers toward infant primates, and infants show distress toward caregivers when they appear to violate expectations in their relationships. We argue that these early social-relational expectations and emotions may form the base of obligation.

Tomasello proposes that the concept of obligation develops “from the inside-out”: emerging first in experiences of shared agency and generalizing outward to shape children's broader understanding. Here I consider that obligation may also develop “from the outside-in,” emerging as a domain-specific instantiation of a more general conceptual bias to expect categories to prescribe how their members are supposed to behave.

We extend Tomasello's framework by addressing the functional challenge of obligation. If the long-run social consequences of a decision are sufficiently costly, obligation motivates the actor to forgo potential immediate benefits in favor of long-term social interests. Thus, obligation psychology balances the downstream socially-mediated payoffs from a decision. This perspective can predict when and why obligation will be experienced.

Satisfying one's obligations is an important part of being human. However, people's obligations can often prescribe contradictory behaviors. Moral obligations conflict (loyalty vs. fairness), and so do obligations to different groups (country vs. family when one is called to war). We propose that a broader framework is needed to account for how people balance different social and moral obligations.

Tomasello's novel and insightful theory of obligation explains why we sometimes sense an obligation to treat each other equally, but he has not yet explained why human morality also allows and enables much inequality in wealth and power. Ullman-Margalit's (1977) account of norms of partiality suggested a different source and kind of norms that might help to fill out Tomasello's picture.

Tomasello's moral psychology of obligation would be developmentally deepened by greater attention to early experiences of cooperation and shared social agency between parents and infants, evolved to promote infant survival. They provide a foundation for developing understanding of the mutual obligations of close relationships that contribute (alongside peer experiences) to growing collaborative skills, fairness expectations, and fidelity to social norms.

My response to the commentaries focuses on four issues: (1) the diversity both within and between cultures of the many different faces of obligation; (2) the possible evolutionary roots of the sense of obligation, including possible sources that I did not consider; (3) the possible ontogenetic roots of the sense of obligation, including especially children's understanding of groups from a third-party perspective (rather than through participation, as in my account); and (4) the relation between philosophical accounts of normative phenomena in general – which are pitched as not totally empirical – and empirical accounts such as my own. I have tried to distinguish comments that argue for extensions of the theory from those that represent genuine disagreement.

The processes underwriting the acquisition of culture remain unclear. How are shared habits, norms, and expectations learned and maintained with precision and reliability across large-scale sociocultural ensembles? Is there a unifying account of the mechanisms involved in the acquisition of culture? Notions such as “shared expectations,” the “selective patterning of attention and behaviour,” “cultural evolution,” “cultural inheritance,” and “implicit learning” are the main candidates to underpin a unifying account of cognition and the acquisition of culture; however, their interactions require greater specification and clarification. In this article, we integrate these candidates using the variational (free-energy) approach to human cognition and culture in theoretical neuroscience. We describe the construction by humans of social niches that afford epistemic resources called cultural affordances. We argue that human agents learn the shared habits, norms, and expectations of their culture through immersive participation in patterned cultural practices that selectively pattern attention and behaviour. We call this process “thinking through other minds” (TTOM) – in effect, the process of inferring other agents’ expectations about the world and how to behave in social context. We argue that for humans, information from and about other people's expectations constitutes the primary domain of statistical regularities that humans leverage to predict and organize behaviour. The integrative model we offer has implications that can advance theories of cognition, enculturation, adaptation, and psychopathology. Crucially, this formal (variational) treatment seeks to resolve key debates in current cognitive science, such as the distinction between internalist and externalist accounts of theory of mind abilities and the more fundamental distinction between dynamical and representational accounts of enactivism.

The Bayesian brain hypothesis, as formalized by the free-energy principle, is ascendant in cognitive science. But, how does the Bayesian brain obtain prior beliefs? Veissière and colleagues argue that sociocultural interaction is one important source. We offer a complementary model in which “interoceptive self-inference” guides the estimation of expected uncertainty both in ourselves and in our social conspecifics.

We applaud the ambition of Veissière et al.'s account of cultural learning, and the attempt to ground higher order thinking in embodied theory. However, the account is limited by loose terminology, and by its commitment to a view of the child learner as inference-maker. Vygotsky offers a more powerful view of cultural learning, one that is fully compatible with embodiment.

Thinking through other minds creatively situates the free-energy principle within real-life cultural processes, thereby enriching both sociocultural theories and Bayesian accounts of cognition. Here, shifting the attention from thinking-through to becoming-with, we suggest complementing such an account by focusing on the empirical, computational, and conceptual investigation of the multiscale dynamics of social interaction.

Veissière and colleagues suggest that thinking is entirely based on social norms. I point out that despite the fact that social norms are commonly used to alleviate cognitive processing, some individuals are willing and able to go about the costly process of questioning them and exploring other valuable ways of thinking.

Veissière et al. must sacrifice explanatory realism and precision in order to develop a unified formal model. Drawing on examples from cognitive archeology, we argue that this makes it difficult for them to derive the kinds of testable predictions that would allow them to resolve debates over the nature of human social cognition and cultural acquisition.

The authors deploy an epistemic framework to represent culture and model the acquisition of cultural behavior. Yet, the framing inherits familiar problems with explaining the acquisition of norms. Such problems are conspicuous with regard to human societies where norms are ubiquitous. This creates a new difficulty for the authors in explaining change to mutually exclusive organizational structures of human life.

The multicultural experience (i.e., multicultural individuals and cross-cultural experiences) offers the intriguing possibility for (i) an empirical examination of how free-energy principles explain dynamic cultural behaviors and pragmatic cultural phenomena and (ii) a challenging but decisive test of thinking through other minds (TTOM) predictions. We highlight that TTOM needs to treat individuals as active cultural agents instead of passive learners.

Emerging cybertechnologies, such as social digibots, bend epistemological conventions of life and culture already complicated by human and animal relationships. Virtually-augmented niches of machines and organic life promise new free-energy-governed selection of intelligent digital life. These provocative eco-evolutionary contexts demand a theory of (natural and artificial) minds to characterize and validate the immersive social phenomena universally-shaping cultural affordances.

Although we applaud the general aims of the target article, we argue that Affective Social Learning completes TTOM by pointing out how emotions can provide another route to acquiring culture, a route which may be quicker, more flexible, and even closer to an axiological definition of culture (less about what is, and more about what should be) than TTOM itself.

Veissière and collaborators ground their account of culture and social norms in the free-energy principle, which postulates that the utility (or adaptive value) of an outcome is equivalent to its probability. This equivalence would mean that their account entails that complying with social norms has always adaptive value. But, this is false, because many social norms are obviously maladaptive.