For neuroscientists dedicated to explain brain function and motivation as mechanistically emergent effects, this paper provides support for a potentially revolving insight: a strictly cognitive process, the individual’s notion of the state of their social environment, can cause such profound stress that it can induce temporary infertility among females. The disruption of the reproductive function depends on the hypothalamus, implying that cognitive perception of social stress is a critical factor for hypothalamic function and therefore mood (Bao & Swaab, Reference Bao and Swaab2019). Dunbar (Reference Dunbar2025) also shows evidence that the social cognitive processes among primates can drive the individual to adopt almost any behavior to be able to defuse perceived social stress. The resulting behavioral strategies and the emergent “glass ceilings” that limit the maximum group sizes apparently depend on the cognitive limits of the species to find behavioral strategies to defuse the social stress, which scales with the social group size.

From a neurophysiological and psychiatric perspective, the paper presents an important opportunity to formulate new perspectives on how the internal brain signaling of stress and release from stress, potentially a powerful reward signal in itself, is organized. The physiological and evolutionary origin of stress is the need of the brain to regulate the activity distribution across our body and inner organs in relation to what actions we currently engage in, ultimately mediated via the autonomic nervous system (ANS). For example, when perceiving stressful situations such as the presence of a predator, it is important to prepare by redirecting more blood to the muscles rather than the gut. There are tremendous numbers of sensors across all inner organs (Wang & Chang, Reference Wang and Chang2024), the skin, and the entire musculoskeletal body, which are dedicated to directly signal potential harm such as tissue damage, fatigue, or physiological/homeostatic imbalances, which the brain uses to redirect activity focus across our body (Mårtensson et al., Reference Mårtensson, Enander, Rongala and Jörntell2025). In contrast, the notion of social wear and tear has no ground truth sensor signals. Our estimates of potential social imbalances instead must originate from our perceptual and reasoning capabilities in the cortex.

Hence, the findings of Dunbar (Reference Dunbar2025) demonstrate that the cognitive analyses of the social conditions can be the origin of stressful states internally in our minds. These internal states will be reflected in the activity of the brain circuitry, in the examples given by Dunbar (Reference Dunbar2025), even to the extent that hypothalamic function is impacted. In analogy with the ANS, these internal brain states continually lead to a rebalancing of activity across the brain for it to cope with the most acute contextual aspects, and the mediators of this rebalancing (Breton-Provencher et al., Reference Breton-Provencher, Drummond, Feng, Li and Sur2022) therefore can be viewed as increasing or decreasing their activity merely as a secondary, consequential effect. This would have revolving implications for psychiatry, which otherwise widely rely on what may be referred to as some form of “biochemical” pushbutton hypothesis, i.e., that our mood disorders are caused by imbalances in some specific brain-internal mediator (such as endorphin, oxytocin, dopamine, serotonin, histamine, noradrenaline, and acetylcholine). But it is widely known among practicing psychiatrists that the biochemical pushbutton hypothesis does not rhyme well with effects observed in reality, and other than pain relief, over the last few years, spectacular retrospective analyses demonstrated its inefficacy over placebo (Leichsenring et al., Reference Leichsenring, Steinert, Rabung and Ioannidis2022; Moncrieff et al., Reference Moncrieff, Cooper, Stockmann, Amendola, Hengartner and Horowitz2023).

An important overlooked fact is that none of these mediator systems has any proven intrinsic mechanism to judge whether we are in a stressful or a beneficial situation. Current hypotheses that these mediators are primary causes of stress have instead implicitly assumed a deus ex machina function in them, such that they somehow magically “know” what a reward or a stressful situation would look like and regulate their signals based on that. But the only brain system that has a sufficiently rich intrinsic working memory (Goldman-Rakic, Reference Goldman-Rakic1995; Kristensen et al., Reference Kristensen, Kesgin and Jörntell2024) to evaluate whether a complex, abstract, and temporally evolving phenomenon like social network state is potentially harmful or beneficial is the cortical network. The cortex exerts its influences widely across the central nervous system (CNS), including on the limbic system and hypothalamus, and can concertedly control the release of the mediators listed above to balance the brain activity distribution depending on the perception of contextual state (Stratmann et al., Reference Stratmann, Albu-Schäffer and Jörntell2018). Hence, there is a mechanistic infrastructure in the brain through which the cognitive perception of social stress can control what we traditionally think of as mood-related brain signals–but the mood may depend on the cortex itself.

That social interactions are stressful naturally depends on that we cannot control other individuals, and the social interaction is therefore inherently unpredictable. But then, why are social networks formed at all? What is the factor that can defuse the stress of the first social interaction we would ever have and thus allowing our cognitive systems to learn that social interactions can be beneficial? At least in mammals and marsupials, there is a greater stress that the first social interaction can defuse: nutritional needs, signaled by the hypothalamus. These needs can be relieved by breastfeeding or by providing for nutrient intake by other means. In animals, the process of suckling would hence lead to an early foundation in the cognitive system that social interactions have an element of major potential goodness (release from stress). This could, in turn, lead to a natural cognitive drive later in life to explore social interactions as a means of defusing other stress factors. Indeed, the nutritional stress (or gain) is indeed a key factor across different species of primates that drives the social network cohesiveness and the specific strategies adopted by the social group–and depending on the size of their cortex, these social networks can take on different levels of complexity (Dunbar, Reference Dunbar2025). Indeed, whales can also exhibit highly differentiated social behaviors, where the feeding behavior and strategy tend to be unique for each social group (Jourdain et al., Reference Jourdain, Karoliussen, Fordyce Martin, Langangen, Robeck, Borgå and Foote2024). A related question is at what lower limit of cortex size can we begin to speak of true social behavior rather than herd-formation (Dunbar, Reference Dunbar2025)?