Laughter is mediated strongly by social context: we are 30 times more likely to laugh if we are with someone else than if we are on our own (Scott et al. 2014: 1).
High quality social relationships correspond with longer, happier, and healthier lives—facts that hold true, as far as anyone knows, regardless of geography or culture (Coan and Sbarra 2015: 1)
I have long been an advocate for collectivism. My basic argument for collectivism is evolutionary rather than political: humans are an obligatorily social species; we evolved over millions of years to live in groups (small and large); and almost every feature of humanity only makes sense in this social perspective.
Moreover, one can see rudiments of the same social structures and processes in our great ape cousins, as well as in elephants, dolphins, and dogs. We also see them in some social birds. Nothing about the evolution of social species says that master-servant relationships are natural. Indeed, humans are the only mammals that enter into such relationships. If social species had a general motto, it would be:
"All for one and one for all.
United we stand, divided we fall."
I learned a great deal about this from Frans de Waal (1948 - 2024), the primate ethologist who regretted ever coining the widely misunderstood term "alpha male" (See this TED Talk video where he explains what it really means). More specifically, de Waal's book The Bonobo and the Atheist sketches out an evolutionary theory of morality as a feature of evolving to live socially (outlined in another TED Talk).
As the saying goes: "It takes a village to raise a child". This is not hyperbole. Apart from a handful of outliers, humans live in communities. Our genes generally get passed on within an extended community. The alternative, individualism, tends to be a pathology in a social species since individualism weakens the group and leaves individuals vulnerable to predators. Moreover, social isolation is known to cause both mental and physical health problems.
More recently, I have picked up on the idea of allostasis. I learned about this from Lisa Feldman Barrett's book How Emotions Are Made, but it also crops up in Karl Friston's free energy principle. I keep meaning to write something about the free energy principle, but haven't gotten to it yet.
In this essay, I want to explore a newish idea that combines evolutionary perspectives on social lifestyles with allostasis. This new theory correlates extremely well with Frans de Waal's account of primate sociability and the evolution of morality. It also correlates well with Robin Dunbar's account of human evolution and the evolution of religion. These kinds of correlations with existing explanations (that I already find useful) are what I look for in a new idea. The best explanations are those that cover the broadest range of topics with the smallest number of assumptions.
I begin with some background on allostasis and the free energy principle, before outlining social baseline theory as it appears in the literature.
Homeostasis and Allostasis
We have long known that certain bodily processes are governed by feedback mechanisms that keep our bodies within specific limits, a goal known as homeostasis. The study of feedback in this sense is called cybernetics.
An obvious example of homeostasis in humans is body temperature, which ideally stays in a 1 °C range: 36.5–37.5 °C. That is, we ideally keep our body temperature to within half a degree of 37 °C. Prolonged periods of too low or high temperatures may be fatal. If we go below 36.5 °C, we get goosebumps and shiver, which warms us; if we go above 37.5 °C, we start to sweat, pant, and/or fan ourselves, which cools us. In this way, we can survive in air temperatures of roughly -50 °C to +50 °C.
However, it soon became apparent that something more active was also going on, especially in the brain. Rather than relying entirely on reactive homeostasis, the brain is actually predicting what will happen next and preparing for that. Predictions are based on expectations formed from past experience. This process, which has the same goal as homeostasis, is called allostasis.
Allostasis is often likened to Bayesian inference, and for some, it is functionally similar to the idea of the Bayesian brain (although see also Mangalam (2025) for a critique). In Bayesian inference, we try to find the most likely outcome by iteratively improving our knowledge and using that knowledge to assess the likelihood of various possible outcomes. This is also similar to the scientific method more generally. We can take it to refer to a process of updating our expectations based on the available knowledge.
Karl Friston's free energy principle suggests that biological systems, right down to the level of individual cells, have Bayesian-like processes that minimise unexpected results, by either refining predictions or changing something (internal or external to the cell). There is a kind of intelligence here that is evident even in single-celled organisms. And Friston proposes that this process is driven by physical processes analogous to thermodynamics.
I should add that this "intelligence" is in no way connected to being aware or self-aware. The fact that we can see it in single-celled organisms means that it's a purely mechanistic process, albeit one of dazzling complexity (even at that scale). That said, it's difficult to watch, say, an amoeba pursuing its prey without imagining some kind of agency. However, this reflects a limit on the human imagination, not a statement about nature generally. Humans have a strong tendency to see agency where none exists, as explained by Justin L. Barrett in Why Would Anyone Believe in God? (which I have cited many times over the years).
We've probably all experienced "getting up top fast", where, rising from sitting to standing, we briefly get light-headed and dizzy. This is a minor failure in allostasis. What usually happens is that the desire to stand up occurs and the body automatically begins to prepare for it by, for example, raising your blood pressure. Better to be prepared and then not need it than to pass out from lack of blood in the brain every time you stand up in a hurry.
We need to be clear that the brain is not acting like a little person in this scenario. The brain, per se, does not know what's going on. The brain, per se, is not aware separately from our first-person perspective. At the end of the day, all the brain does is accept inputs and generate outputs. In human self-awareness, the functioning of the brain is entirely transparent to our first-person perspective. The brain was not even suspected to be where thinking goes on until the late 19th century.
To date, we can only describe the functioning of the brain as a kind of "black box". The brain receives millions of inputs from nerves and its own neurons, it weighs and evaluates those inputs, and it then produces a variety of outputs, including actions such as changing the balance of substances in our blood, changing our respiration, and moving our body around. A movement like walking involves thousands of signals per second, going to hundreds of muscles, which must contract and relax in a very specific sequence, or we simply fall over. Balancing on two legs is an incredibly complex process, which is only made more difficult by moving around. However, such processes are also entirely transparent to our conscious mind, because conscious coordination on that scale would be far too slow to be useful, if it were possible at all.
As well as coordinating nerve inputs and outputs, the brain also coordinates energy inputs and outputs in the body. The brain tries to anticipate energy expenditure and tries to ensure that the body has sufficient resources. And this part is generally not under our conscious control. It involves feelings like hunger and satiation. It makes sense that the brain would seek to minimise or at least optimise energy expenditure when seeking any goal.
For example, we walk most places rather than sprinting at full speed. Sprinting requires very intensive use of resources that cannot be sustained over more than about 15 seconds, usually with a recovery time of several hours, which limits how often one can sprint. Fanny Blankers-Koen (Netherlands) won four sprint races in one day at the 1948 London Olympics, running a total of 580 m. Although note, she also went on a celebratory shopping spree before the final relay, arriving back at the stadium with only 10 minutes to spare. Blanker-Koen might have run in other races that day, but the organisers limited her to three individual races and one relay. Still there are limits.
By contrast, humans can easily walk for several hours every day, day after day. Energy budgeting turns out to be a very powerful way of thinking about what the brain does. Slow and steady does win the race over long distances.
One of the striking features of Karl Friston's free energy principle is that cells, brains, and organisms use the same basic architecture to optimise energy expenditure. Hence, the incredible explanatory power of the free energy principle.
Friston says that optimising energy expenditure is based on making accurate predictions, or in other words, based on minimising surprises. Discrepancies between expectation and reality (aka surprises) are dealt with in one of two ways: change the prediction or change the input (i.e do an action). The less we have to change predictions or take unexpected actions, the better we can manage our energy expenditure.
I am still a bit vague about the next part of the argument. As I understand it, at present, Friston's most important contribution has been to mathematically describe this process of minimising surprise. This enabled him to show that minimising surprise is mathematically the same idea as minimising free energy in information theory. Moreover, both are mathematically equivalent to Bayesian inference.
If you watch interviews with Friston, he tends to switch between three approaches: his own mathematical description of surprise, information theory, and Bayesian theory. In order to follow his train of thought, one needs to know these topics quite well. And, honestly, I don't.
Another important concept is allostatic load, a term coined by Bruce McEwen and Eliot Stellar (1993). I have written about this before without knowing what it was called (see Rumination, the Stress Response, and Meditation. 22 January 2016). McEwen and Stellar (1993: 2093) define allostatic load as:
...the cost of chronic exposure to fluctuating or heightened neural or neuroendocrine response resulting from repeated or chronic environmental challenge that an individual reacts to as being particularly stressful.
Allostatic load has measurable consequences, such as changes in blood pressure and cortisol levels. It is a physiological process whereby an organism can become depleted by constant stress. I first observed this in a biology class in 1982 while studying earthworms. I especially noted that fight or flight responses rapidly drain our resources (especially in the endocrine system), which take some time to recover. For example, physiological arousal causes our adrenal glands to secrete adrenaline (aka epinephrine). The adrenaline is released in milliseconds, but it takes several hours to replenish. While we try to keep some in reserve, chronic arousal depletes the store. If our store of adrenaline is exhausted, the body tries to become aroused, i.e. ready to fight or flee, but it cannot respond.
Repeated stimulation of these responses without time to recover leads to responses of diminishing intensity until we become unable to respond. In my earlier essay, I made a connection from this to what doctors in the UK call major depression (aka clinical depression). And I characterised depression as a collapse of our ability to respond to stimuli as the result of hyper-stimulation or what I would now call allostatic overload. Many have noted that the symptoms of depression are as much physical as mental. Physical exhaustion is a common symptom of depression, consistent with adrenaline depletion. A recent review of many studies (Gou et al 2025) confirms my suspicion:
High [allostatic load] is positively associated with increased risks of depression, anxiety, and suicide, highlighting its potential as a predictive tool in mental health.
With this preamble complete, we can now explore an idea that applies this same kind of insight on a higher level of organisation: the social group.
Social Baseline Theory
Social Baseline Theory was first proposed by Lane Beckes and James A. Coan in 2011. With subsequent contributions from David A. Sbarra and others. I should emphasise that social baseline theory is still relatively new. While it emerges from considering patterns in neuroscience evidence, it is still somewhat speculative. I'm enthusiastic because what has emerged to date is very consistent with ideas that I've been writing about for a decade or more. To me, it makes good sense, and I find it a useful addition to how I think about humanity.
The basic insight of social baseline theory is that, having evolved to live in groups, one of the things we expect in our optimisation of energy expenditure is social input. In this view, having good social connections is part of the baseline of being social. Our social group is perceived by the body as a resource that it can draw on. A corollary of this is that not having those connections is actually a drain on us.
As noted, Friston sees the same kind of allostatic mechanism operating in both cells, brains, and organisms. Here, Beckes and Coan extend this same reasoning to social groups. The abstract from the original paper is an excellent (if jargon-heavy) summary:
Social proximity and interaction attenuate cardiovascular arousal, facilitate the development of non-anxious temperament, inhibit the release of stress hormones, reduce threat-related neural activation, and generally promote health and longevity. Conversely, social subordination, rejection and isolation are powerful sources of stress and compromised health. Drawing on the biological principle of economy of action, perception / action links, and the brain’s propensity to act as a Bayesian predictor, Social Baseline Theory (SBT) proposes that the primary ecology to which human beings are adapted is one that is rich with other humans. Moreover, SBT suggests that the presence of other people helps individuals to conserve important and often metabolically costly somatic and neural resources through the social regulation of emotion. (Beckes & Coan 2011)
This might be too technical for some, so let's unpack it by working through their argument. In this view, then, humans require the presence of other humans to help regulate our bodies.
...the human brain is designed to assume that it is embedded within a relatively predictable social network characterized by familiarity, joint attention, shared goals, and interdependence. (Beckes and Coan 2011: 976-977)
The brain was not "designed". I would say that the brain evolved to expect being socially embedded. Other than this quibble, this part of the explanation seems reasonably clear. The fact that we evolved to live in communities means that the presence of a community is something that we expect to be present. It would be weird not to. It also makes the absence of a community notable. In short, evolution has optimised our bodies and minds for living in close-knit communities. And as Frans de Waal noted, social animals have two capacities in common: empathy and reciprocity. We respond emotionally to our community and they to us.
The presence or absence of a supportive social environment is evident in our blood, especially during stressful events. Individuals carry the whole load of the stress, communities distribute it.
What Beckes and Coan are saying is that allostasis which we can see operating in cells and brains, can also be seen to operate at the level of communities.
Moreover, Beckes, Coan and others have observed that when social support is available, the parts of the brain believed to be involved in self-regulating emotions are less active. Which suggests that the regulation of emotions in social situations works by a different mechanism. This leads to socialised awareness:
When proximity [to a supportive community] is maintained or reestablished, the brain is simply less vigilant for potential threats, because it is embedded within the social environment to which it is adapted. (Beckes and Coan 2011: 977)
There are specific biological mechanisms that make being in a social group more energy efficient for all the individuals involved.
Both social baseline theory and the free energy principle assume that one of the main roles of the brain is optimising the body's energy use. And many studies in both humans and animals support this conclusion (Beckes and Coan 2011: 978).
There are two principal benefits to this: risk distribution and load sharing.
Even though group living can be costly in terms of resource competition, overall fitness is enhanced in groups by decreasing the risk of predation, injury, and other potential threats (Krebs & Davies, 1993)... Thus, social groups typically settle into sizes that optimize access to food against the distribution of risk across group members. (Beckes and Coan 2011: 978)
Risk distribution is more or less just safety in numbers. If the task is spotting predators, for example, then having a dozen pairs of eyes is more efficient than just one. This means that when someone "has your back", you are able to use energy more efficiently.
Loading sharing is a distribution of effort towards common goals (Coan & Sbarra 2015: 2). This takes on more significance in the light of allostatic load. Sharing the allostatic load decreases the cost of responding to stress, by distributing it across the group.
Load sharing means that the activities of other group members provide benefits to the individual, whether the individual participates or not. For example, in foraging societies, we often see men focused on hunting and women focused on gathering (and sometimes cultivating) plants. Women benefit from having meat, and men benefit from having vegetables. And both benefit from the specialisation that leads to enhanced expertise.
We feel less threatened by surprising stimuli when we are in the presence of someone with whom we share a bond. In this view, our social network becomes an extension of our self. Normally, the presence of other people helps us to regulate our emotions in ways that are more efficient than simple self-regulation.
Many theorists have suggested that the self is “expanded” by relationships with others [26]. This may be literally true at the neural level. For example, the brain encodes threats directed at familiar others very similarly to how it encodes threats directed at the self—but no such similarity obtains for strangers [27]. (Coan & Sbarra 2015: 3)
This appears to be consistent with the idea that identity in some societies is more social than amongst industrialised Europeans and colonists. Indeed, the rampant individualism associated with the Neoliberal Revolution can be seen as a pathological state for a social species. We cannot survive alone, much less attain our broader aims in life.
An extension of social baseline theory involving the concept of allostatic load was proposed by Saxbe et al (2020). They proposed that allostatic load has a social analogue. This makes sense. In a social group characterised by the capacities for empathy and reciprocity, we might imagine that emotional contagion generalises.
Robin Dunbar (2014, 2023) has observed that group bonding in non-human primates is strongly based on one-to-one activities, especially grooming. Grooming stimulates endorphins, which mediate our sense of well-being. Grooming calms both the individuals involved and creates a sense of bonding (or attachment).
However, Dunbar points out that human groups have grown too large for one-to-one grooming to be efficient. We simply don't have time for ~150 individuals to spend enough time grooming each other for this to work. What we have instead are group activities that achieve the same end. Notably, collective practices such as singing and dancing together have the same effect on our sense of well-being as grooming does in other animals.
The idea of social allostatic load is that our relationships help us to regulate our emotions. And the loss of relationships then puts additional stress on the individual.
Conclusion
None of this seems surprising or counterintuitive. It is entirely consistent with the idea that we evolved, over millions of years, to live in social groups. It is also consistent with the idea that we employ both homeostatic and allostatic processes at a variety of scales.
...the brain manages energy and behavior by making predictions about outcomes given (i) the current situation (particularly constraints, risks, and opportunities), (ii) the predicted possible future situation(s), (iii) situational goals, (iv) current energy states, and (v) expected future energy states (Beckes and Coan 2011: 980-981).
Or as Coan and Sbarra (2015: 1) put it:
... the human brain assumes proximity to social resources—resources that comprise the intrinsically social environment to which it is adapted. Put another way, the human brain expects access to relationships characterized by interdependence, shared goals, and joint attention.
My comment here is that the human brain is like this because we evolved to live in communities.
Beckes and Coan also make the link from Social Baseline Theory to another popular model: Attachment Theory. This is a topic for another essay, I just wanted to note the connection.
Some readers may recall that I often cite Ariel Glucklich (1997) when talking about magic. In his book The End of Magic, he says
Magic is based on a unique type of consciousness: the awareness of the interrelatedness of all things in the world by means of simple but refined sense perception... magical actions... constitute a direct, ritual way of restoring the experience of relatedness in cases where that experience has been broken by disease, drought, war, or any number of other events. (1997: 12)
However, in refreshing my memory of that first citation in 2008, I noticed something that I usually forget. The next part of the story is:
[Magic] is a natural phenomenon, the product of our evolution as a human species and an acquired ability for adapting to various ecological and social environments. (1997: 12)
As I say, I find an idea more compelling when it fits with other explanations I find useful. One idea that explains numerous phenomena is more compelling than an idea that explains one phenomenon in isolation. We could now say that "magic" exploits the effects described by social baseline theory. A sense of being securely connected to our social environment is essential to health and well-being.
And thus the general sense of alienation and isolation that pervades British and (more so) US culture is a symptom of a deep malaise. The Neoliberals sought to downplay "society" in order to foster extreme forms of individualism (including libertarians and billionaires), which have been all too clearly detrimental to both Britain and the USA.
Stressed by constant change, constant threats to our livelihood, and threats to our communities, we begin to experience allostatic overload. Depression and anxiety are some of the most obvious results. Given the pace of change in my lifetime and the disruptions caused by the Neoliberal revolution, we might expect to see sharp increases in mental health problems after the 1980s. And we do.
And this leads me to one last connection. In an article that dramatically changed how I see modern politics, Karen Stenner (2020) describes how and why some people opt to support authoritarian leaders. In Stenner's terms, people with the authoritarian disposition make up around 30% of the population. They are characterised by low scores on IQ tests and low scores on the Big 5 criterion of openness to experience. Such people want sameness (group conformity) and oneness (group authority). When that expectation is denied and combined with, say, perceived economic hardship (such as a decline in living standards), then they support the first authoritarian ruler who comes along promising a return to the "good old days". Thus, we can see the rise of fascism in the 21st century as a response to social allostatic overload caused by the neoliberal revolution.
As I say, social baseline theory is relatively new still. More research needs to be done to establish it in the scientific community. That said, the idea that we rely on other people is a no-brainer. We do. Of course we do. What we are doing here is not stating some new and hitherto unsuspected truth. Rather, we are finessing and "putting the eyebrows on" something that everyone knows.
Social baseline theory emphasises that, for an obligatorily social animal, the lack of a social context prevents us from optimising our energy expenditure and is detrimental to our health and well-being.
~~Φ~~
Bibliography
Barrett, Justin L. (2004). Why Would Anyone Believe in God? Altamira Press.
Beckes, L., & Coan, J. A. (2011). "Social baseline theory: The role of social proximity in emotion and economy of action." Social and Personality Psychology Compass, 5(12), 976–988. https://doi.org/10.1111/j.1751-9004.2011.00400.x
Coan, J. A., & Sbarra, D. A. (2015). "Social Baseline Theory: The Social Regulation of Risk and Effort." Current Opinion in Psychology 1: 87–91. https://doi.org/10.1016/j.copsyc.2014.12.021 https://pmc.ncbi.nlm.nih.gov/articles/PMC4375548/
Dunbar, Robin. (2014). Human Evolution: A Pelican Introduction. Pelican.
Dunbar, Robin. (2023). How Religion Evolved: And Why It Endures. Penguin.
Glucklich, Ariel. (1997). The end of magic. New York, Oxford University Press.
Gou, Y., Chenga, S., Kanga, M. et al. (2025). "Association of Allostatic Load With Depression, Anxiety, and Suicide: A Prospective Cohort Study." Archival Report 97(8): 786-793. https://doi.org/10.1016/j.biopsych.2024.09.026
Mangalam, M. (2025). "The myth of the Bayesian brain." European Journal of Applied Physiology, 125(10), 2643–2677. https://doi.org/10.1007/s00421-025-05855-6
McEwen, B.S. & Stellar, E. (1993). "Stress and the individual. Mechanisms leading to disease". Archives of Internal Medicine 153(18): 2093–2101. doi:10.1001/archinte.153.18.2093.
Saxbe, D. E., Beckes, L., Stoycos, S. A., & Coan, J. A. (2020). "Social Allostasis and Social Allostatic Load: A New Model for Research in Social Dynamics, Stress, and Health." Perspectives on psychological science: a journal of the Association for Psychological Science 15(2): 469–482. https://doi.org/10.1177/1745691619876528
Scott, S. K., Lavan, N., Chen, S., & McGettigan, C. (2014). "The social life of laughter." Trends in Cognitive Sciences 18(12): 618–620. https://doi.org/10.1016/j.tics.2014.09.002
Stenner, Karen. 2020. "Authoritarianism." Hope Not Hate. https://hopenothate.org.uk/2020/11/01/authoritarianism/
