The role that cognitive abilities play in group cohesion and stabilization

Dunbar’s argument is that mammals that live in stable social groups use two strategies to keep and reproduce the group, besides group structure itself: grooming, as a form of affiliative interaction, and a suite of higher-order cognitive skills. This dichotomy overlooks the role of “mid-level” mechanisms that combine emotional and cognitive components, such as trust itself or empathy (Pérez-Manrique & Gomila, Reference Pérez-Manrique and Gomila2022, Reference Pérez-Manrique and Gomila2018), among other forms of social bonding and affiliation.

Importantly, and in contrast with cognitive abilities, these emotional dispositions involve a motivational component that is evolutionarily more effective to keep cooperative commitments and avoiding free-riding than cognition itself (Jablonski, Reference Jablonski2021; Frank, Reference Frank1993). In other words, it is wrong to assume that higher cognition is the only way to make cooperation and social coordination possible in bigger groups. As a matter of fact, the paper is full of examples of species that achieve group stability and cohesion without going cognitive. Therefore, the contention that the evolution of cognition in primates is driven by social pressures is not warranted.

For instance, as suggested by the paper itself, instances of group synchronization and entrainment, based on the “do as your neighbor” rule, may turn out to be a very effective solution to group coordination, a solution that does not require higher cognitive abilities. It is found in different species (Bates & Byrne, Reference Bates and Byrne2010), including primates, whose capacity for observation and learning by imitation is well-established (Whiten, Reference Whiten2000; Whiten & van de Waal, Reference Whiten and van de Waal2017), and that imitation induces affiliation (Dignath et al., Reference Dignath, Born, Eder, Topolinski and Pfister2021; Dignath et al., Reference Dignath, Lotze-Hermes, Farmer and Pfister2018).

Another example can be observed in species such as termites, bees, and ants, which secrete chemicals that serve to drive their coordinated actions and cohesion when required by the environment, without needing for cognitive complexity (Barbero, Mannino, & Casacci, Reference Barbero, Mannino and Casacci2023; Mizumoto et al., Reference Mizumoto, Lee, Valentini, Chouvenc and Pratt2021; Pernal, Reference Pernal2021). In primates, this chemical mechanism, present with the secretion of endorphins, is related to the cognitive mechanism because it is linked to social contact (Dunbar, Reference Dunbar2025). However, there are studies showing that primates’ more complex cognitive abilities can be more difficult to coordinate collective action within a group than the chemical mechanism of ants (Dreyer et al., Reference Dreyer, Haluts, Korman, Gov, Fonio and Feinerman2025).

Second, acknowledging these emotional-cum-basic cognitive abilities also entails that the differentiation of structural, behavioral, and cognitive strategies of group coordination is misleading. Just as different kinds of social structures can be distinguished at the same time in humans, as our work on trust networks shows the same point can be made for social mammals, where different structures can be distinguished depending on the kind of link chosen: status, patterns of residence, and affiliation but grooming, but also by empathy.

Likewise, large groups may exist with very weak cohesion. The structural mechanisms that enable the maintenance of relatively stable large groups, are compatible with such groups being comprised of smaller ones that coexist to a certain extent, due to the cognitive limitations of managing an excessively large number of relationships (Operario & Fiske, Reference Operario and Fiske1999), resulting in larger groups maintaining relatively weak internal cohesion (Alvarez-Rodriguez et al., Reference Alvarez-Rodriguez, Battiston, de Arruda, Moreno, Perc and Latora2021). This weakness of the large group may occur when individuals spend more time cultivating their closest relationships, perhaps to secure their ties against other cliques competing for resources, which causes greater distancing between coalitions (Chae, Reference Chae, Kim, Kim, Lim, Kim and Kim2022; Morris-Drake, Reference Morris-Drake, Kennedy, Braga Goncalves and Radford2022). Similarly, in humans, our work shows that a large group may be composed of multiple cliques without much contact among them (Acedo-Carmona & Gomila, Reference Acedo-Carmona and Gomila2019, Reference Acedo-Carmona and Gomila2013a, Reference Acedo-Carmona and Gomila2013b).

Third, the cognitive strategy may be suboptimal when the environment is unstable. For example, we found evidence that the cognitive mechanism of trust creates obstacles for individuals to adapt to rapidly changing environments in the short and medium term (Acedo-Carmona, Munar, & Gomila, Reference Acedo-Carmona, Munar and Gomila2018). Furthermore, trust is also easily susceptible to adversely affect relationships when it is negatively stimulated or when negative experiences are experienced (Davidson & Satta, Reference Davidson and Satta2021; Weiss et al., Reference Weiss, Michels, Burgmer, Mussweiler, Ockenfels and Hofmann2021).

Social structure and group size

In the target paper, Dunbar insists on his well-known argument for an intimate link between group coordination and group size, so that group size is thought to be dependent on the structural, behavioral, or cognitive strategies evolved to achieve group coordination. The assumption is that the bigger the group, the better, and that bigger groups require more cognitive abilities.

While it is true that social structure itself has a limited scope that depends on group size (La Macchia et al., Reference La Macchia, Louis, Hornsey and Leonardelli2016; Sato, Reference Sato1988), it is also true that this limit can be overcome as long as such structures are extended through imitation (Bhoopchand et al., Reference Bhoopchand, Brownfield, Collister, Dal Lago, Edwards, Everett, Fréchette, Oliveira, Hughes, Mathewson, Mendolicchio, Pawar, Pȋslar, Platonov, Senter, Singh, Zacherl and Zhang2023). On the other hand, group size is sensitive to ecological conditions, not only to the strategy available for group coordination. Of course, some habitats may not impose any pressure on group size, but in our work in Northern Ghana, we found that cooperation among different ethnic groups was promoted precisely for the scarcity of resources available. The moral is that group cohesion may turn out to be as high as required by the environment, not a consequence of the suite of cognitive abilities developed.