This post describes homeostasis as a fundamental principle in behaviour and motivation.
The fixity of the milieu supposes a perfection of the organism such that the external variations are at each instant compensated for and equilibrated…. All of the vital mechanisms, however varied they may be, have always one goal, to maintain the uniformity of the conditions of life in the internal environment…. The stability of the internal environment is the condition for the free and independent life.
Claude Bernard (1813-1878)
What is homeostasis?
Sixty-one years after Bernard (1865) wrote about the ‘internal milieu’, Walter B. Cannon (1926) coined the term ‘homeostasis’. Then, 16 years later, psychobiologist Curt Richter (1942) expanded the homeostasis idea to include behavioural or ‘ total organism regulators’ in the context of feeding. From this viewpoint, ‘external’ behaviours that are responses to environmental stimuli lie on a continuum with ‘internal’ physiological events. For Richter, behaviour includes all aspects of feeding necessary to maintain the internal environment. Bernard, Cannon and Richter all focused on a purely physiological form of homeostasis, ‘H[Φ]’. I wish to convince the reader that the idea of the ‘external milieu’, the proximal world of socio-physical action, is equally important.
A General Theory of Behaviour (AGTB) extends homeostasis to all forms of behaviour. Psychological homeostasis can be explained in two stages, starting with the classic version of homeostasis in Physiology, H[Φ], followed by the operating features of its psychological sister, H[Ψ]. The essential features are illustrated in Figure 2.1.
Figure 2.1 Upper panel: A representation of Physiological (Type I) Homeostasis (H[Φ]). Adapted from Modell et al. (2015). Lower panel: A representation of Psychological (Type II) Homeostasis (H[Ψ]).
To be counted as homeostasis, H[Φ], a system is required to have five features:
- It must contain a sensor that measures the value of the regulated variable.
- It must contain a mechanism for establishing the “normal range” of values for the regulated variable. In the model shown in Figure 2.1, this mechanism is represented by the “Set point Y”.
- It must contain an “error detector” that compares the signal being transmitted by the sensor (representing the actual value of the regulated variable) with the set range. The result of this comparison is an error signal that is interpreted by the controller.
- The controller interprets the error signal and determines the value of the outputs of the effectors.
- The effectors are those elements that determine the value of the regulated variable. The effectors may not be the same for upward and downward changes in the regulated variable.
Identical principles apply to Psychological (Type II) Homeostasis (H[Ψ] with two notable differences (Figure 2.1, lower panel). In Psychological Homeostasis, there are two sets of effectors, inward and outward, and the conceptual boundary between the internal and external environments lies between the controller and the outward effectors of the somatic nervous system, i.e. the muscles that control speech and action. Furthermore, Psychological Homeostasis operates with intention, purpose, and desire.
The individual organism extends its ability to thrive in nature with Type II homeostasis. Self-extension by niche construction creates zones of safety, one of the primary goals of Type II homeostasis. Niche construction amplifies the organism’s ability to occupy and control the environment proximally and distally. The use of tools for hunting, weapons for aggression, fire for cooking, domestication of animals, the use of language, money, goods for trade and commodification, agriculture, science, technology, engineering, medicine, culture, music literature and social media are all methods of expanding and projecting niches of safety, well-being and control. Individual ownership of assets such as land, buildings, companies, stocks and shares reflect a universal need to extend occupation, power and control but these possessions do not necessarily increase the subjective well-being of the owner [AP 007].
Initiated by the brain and other organs, homeostasis of either type can often act in anticipatory or predictive mode. One principal function of any conscious system is prediction of rewards and dangers. A simple example is the pre-prandial secretion of insulin, ghrelin and other hormones that enable the consumption of a larger nutrient load with minimal postprandial homeostatic consequences. When a meal containing carbohydrates is to be consumed, a variety of hormones is secreted by the gut that elicit the secretion of insulin from the pancreas before the blood sugar level has actually started to rise. The blood sugar level starts lowering in anticipation of the influx of glucose from the gut into the blood. This has the effect of blunting the blood glucose concentration spike that would otherwise occur. Daily variations in dietary potassium intake are compensated by anticipative adjustments of renal potassium excretion capacity. That urinary potassium excretion is rhythmic and largely independent on feeding and activity patterns indicates that this homeostatic mechanism behaves predictively.
Similar principles operate in Type II homeostasis acting together with the brain as a “prediction machine”. When we anticipate a pleasant event such as a birthday party, there is a preparatory ‘glow’ which can change one’s mood in a positive direction, or thinking about an impending visit to the dentist may be likely to produce feelings of anxiety, or the receipt of a prescription of medicines from one’s physician may lead to improvements in symptoms, even before the medicines are taken.
At societal level, anticipation enables rational mitigation, e.g. anticipation of demographic changes influences policy, threat from hostile countries influences expenditure on defence, and the threat of a new epidemic influences programmes of prevention. [AP 008].
Homeostasis involves several interacting processes in a causal network. A homeostatic adjustment in one process necessitates a compensatory adjustment in one or more of the other interacting processes. To illustrate this situation, consider what happens in phosphate homeostasis (Figure 2.2). Many REF-behaviours that we shall refer to are isomorphic with the 4-process structure in Figure 2.2. However, in nature there is no restriction on the number of interconnected processes and any process can belong to multiple homeostatic networks.
Figure 2.2 Phosphate homeostasis. A decrease in the serum phosphorus level causes a decrease in FGF23 and parathyroid hormone (PTH) levels. Increase in serum phosphorus leads to opposite changes. Calcitriol increases serum phosphorus and FGF23, while it decreases PTH. Increase in FGF23 leads to decrease in PTH and calcitriol levels. PTH increases calcitriol and FGF23 levels. Reproduced from Jagtap et al. (2012) with permission.
Homeostasis never rests. It is continuous, comprehensive and thorough. With each round of the REF, all of the major processes in a network are reset to maintain stability of the whole system. The REF process goes through a continuous series of ‘reset’ cycles each of which stabilizes the system until the next occasion one of the processes falls outside its set range and another reset is required.
Processes in Type II homeostasis may vary along quantitative axes or they can have discrete categorical values. For example, values, beliefs, preferences and goals can have discrete values, as does the state of sleep or waking.
Any change in a categorical process involves change throughout the network to which is belongs. [AP 009].
Such changes may be rapid, in the millisecond range, e.g. a changed preference from chocolate chip cookie flavoured ice cream to Madagascar vanilla that may occurs an instant after arriving at the ice-cream kiosk. At the other end of the spectrum of importance, in buying a new apartment, the final choice might also occur in the instant the preferred option is first sighted. Or the decision could take months or years even though it is of precious little consequence, e.g. deciding that one is a republican rather than a monarchist, or it may never occur because we simply do not care one way or the other. These considerations lead to a surprising proposition that:
The speed of a decision is independent of its subjective utility [AP 010].
One objective of A General Theory of Behaviour is to explain the relevance of the REF system to Psychology. We know already that the regulation of action is guided by three fundamental systems: (i) the brain and central nervous system (CNS), (ii) the endocrine system (ES) and (iii) the immune system (IS). It is proposed in A General Theory that, as a ‘meta-system’ of homeostatic control, these systems collectively govern both physiology and behaviour using the two types of homeostasis, H[Φ] and H[Ψ], respectively. We can understand how this might be possible in light of a recently discovered ‘central homeostatic network’.
THE CENTRAL HOMEOSTATIC NETWORK
Recent analyses of the CNS have explored new methods for discovering cortical and subcortical networks in the brain’s anatomical connectivity termed the ‘connectome’. These studies of the connectome are revolutionary in showing that the CNS is at once both more complex and more simple that previously assumed. Let me explain why.
Regions of interest (ROI) are observed as coherent fluctuations in neural activity at rest as well as distributed patterns of activation or ‘networks’. A network is any set of pairwise relationships between the elements of a system—formally represented in graph theory as ‘edges’ linking ‘nodes’. Neurobiological networks occur at different organizational levels from cell-specific regulatory pathways inside neurones to interactions between systems of cortical areas and subcortical nuclei. Architectures which support cognition, affect and action are normally found at the highest level of analysis. In a landmark study, Brian Edlow and his colleagues investigated the limbic and forebrain structures that form the ‘Central Homeostatic Network’. The Central Homeostatic Network (CHN) plays a major role in autonomic, respiratory, neuroendocrine, emotional, immune, and cognitive adaptations to stress. Collectively, these forebrain structures include the limbic system close to the hypothalamus with strong mono- and/or oligo-synaptic connectivity to one another, and shared participation in homeostasis. Homeostatic forebrain nodes receive sensory information concerning extrinsic threats and interoceptive information from the brainstem, resulting in arousal, attention and vigilance during waking, and visceral and somatic motor defences.
There is complexity here but a well-organized complexity. CHN connectogram shows all six brainstem seed nuclei are interconnected with all seven limbic forebrain target sites, but with markedly different streamline probabilities (SPs) (Figure 2.3). The SP measures the probability of a streamline connecting a seed ROI and target ROI, but does not reflect the strength of the neuroanatomic connection. To ensure that the target ROI size was not the only factor contributing to the SP, Edlow and colleagues verified that the SP measurements were derived from anatomically plausible pathways from animal or other studies of subcortical pathways in the human brain.
Figure 2.3. The connectogram of the human Central Homeostatic Network (CHN). Brainstem seed nodes are displayed on the outside of the connectogram and limbic forebrain target nodes at its center. Connectivity is represented quantitatively, with line thickness being proportional to the streamline probabilities for each dyad. Brainstem seed nodes consist of 7 structures as follows: the hippocampus (Hypo); amygdala (Amg); subiculum (Sub); entorhinal cortex (Ent); superior temporal gyrus (anterior) (STGa); superior temporal gyrus (posterior) (STGp); and insula (Ins). Connectogram lines go to the brainstem nucleus of origin: dorsal raphe DR; median raphe MR; locus coeruleus, LC; paragigantocellularis lateralis, PGCL; caudal raphe, CR; vagal complex, VC. Reproduced in slightly adapted form by permission from Edlow, McNab, Witzel & Kinney (2016).
Brian Edlow’s group study findings suggest that H[Φ] is mediated by ascending and descending interconnections between brainstem nuclei and forebrain regions, which together regulate autonomic, respiratory, and arousal responses to stress. The limbic system has been regarded as the neuroanatomic substrate of ‘emotion’, but its role in the regulation of homeostasis is also now being recognized, and the limbic system has been added to the central autonomic network of “flight, fight or freeze”. Edlow et al. concluded as follows: “connectivity between forebrain and caudal brainstem regions that participate in the regulation of homeostasis in the human brain. These nodes and connections form, we propose, a CHN because its nodes not only regulate autonomic functions such as ‘‘fight or flight’’ and arousal (e.g., median and dorsal raphe, and locus coeruleus) but also non-autonomic homeostatic functions such as respiration (i.e., PGCL) and regulation of emotion/affect (e.g. amygdala)” (Edlow et al., op cit., p. 196). This study supports the idea that interconnected brainstem and forebrain nodes form an integrated Central Homeostatic Network in the human brain. To put this in the simplest terms, the forebrain is involved in homeostatic regulation of both autonomic (Type I) and non-autonomic (Type II) human responses to disturbances of equilibrium. These observations demonstrate that the forebrain provides a common central mechanism for both types of homeostasis, H[Φ] and H[Ψ].
Principle III (Communality): Homeostasis of Types I and II are controlled by a single executive controller in the forebrain.
That the forebrain evolved to control both types of homeostasis, inside the body and in outwardly directed behaviour, supports our contention that homeostasis is a unifying concept across Biology and Psychology. Everything we know about the executive role of the forebrain in action planning and decision-making suggests that this must indeed be the case. Why have two control systems when only one is necessary? The simplicity is beautiful.
HOMEOSTASIS A UNIFYING PRINCIPLE
In the Epilogue to ‘The Wisdom of the Body’, Walter Cannon inquired whether there are any general principles of homeostasis acting across industrial, domestic and social forms of organization? He suggested that the homeostasis of individual humans is dependent on ‘social homoeostasis’ via cooperation within communities. He talks analogously of the system of distribution of goods in society as a stream: “Thus the products of farm and factory, of mine and forest, are borne to and fro. But it is permissible to take goods out of the stream only if goods of equivalent value are put back in…Money and credit, therefore, become integral parts of the fluid matrix of society” (p. 314). He believed that “steady states in society as a whole and steady states in its members are closely linked.” (p. 324).
Compared to more economically stable societies, societies in steep economic growth or decline are expected to have a relatively high prevalence of mental illness [AP 011].
Compared to more egalitarian societies, societies with high levels of inequality are expected to have a relatively high prevalence of mental illness [AP 012].
Ludwig von Bertalanffy (1968) was critical of these externally directed, social forms of homeostasis (Type II). He did not support the idea that homeostasis could be applied to spontaneous activities, processes whose goal is not reduction but building up of tensions, growth, development, creation, and in human activities which are non-utilitarian. There are good reasons to think that von Bertalanffy was wrong. The reach of homeostasis extends well beyond Physiology into many realms of Psychology and even into Society as a whole. H[Φ] and H[Ψ] serve identical stabilizing functions internally in the body and externally in socio-physical interactions of behaviour respectively. With Cannon, we accept that “steady states in society as a whole and steady states in its members are closely linked.” H[Φ] and H[Ψ] exist in a complementary relationship of mutual support. It could not be otherwise.
Principle IV (Steady Stable State): Homeostasis Type II serves the same function for Behaviour as Homeostasis Type I serves for Physiology: the production of a stable and steady state.
According to this principle, behaviour produced by most people most of the time is intended to generally calm ‘waves of unrest’ rather than to make the waves larger, to reduce conflict and to produce cooperation, safety and stability. People with high levels of self-control tend to create social stability and have more, and longer-lasting, friendships than people with relatively low levels of self-control. [AP 013].
Individual set ranges for any particular process vary across people and are not the same for all individuals. Individual set ranges are based on unique interactions of genetics, epigenetics and early infant experience. Set ranges may be changed in a few specific disorders and individual differences exist in the rate and extent of the reset following perturbations to equilibrium. The General Theory carries the expectation of wide individual differences across time and space in set ranges, rates of reset, and adaptations over time.
1) All behaviour involves Type II homeostasis, which strives for a stable and steady state
in the socio-physical world.
2) A single executive controller in the forebrain regulates both type of homeostasis.
3) Individual set ranges are based on genetics, epigenetics and early infant experience. They are normally fixed, changing only with major disorders of function.
 Cannon, W.B. (1926). Physiological regulation of normal states: some tentative postulates concerning biological homeostatics. In A. Pettit. A Charles Richet : ses amis, ses collègues, ses élèves. Paris: Les Éditions Médicales. p. 91.
 Richter, C. P. (1942). Increased dextrose appetite of normal rats treated with insulin. American Journal of Physiology-Legacy Content, 135(3), 781-787.
 It is accepted that so-called ‘set points’ are really ‘set ranges’, e.g. the “normal” human body temperature is a range from 97°F (36.1°C) to 99°F (37.2°C). We use the terms ‘set point’ and ‘set range’ interchangeably.
 Moore-Ede, M. C., & Herd, J. A. (1977). Renal electrolyte circadian rhythms: independence from feeding and activity patterns. American Journal of Physiology-Renal Physiology, 232(2), F128-F135.
 Unless stated otherwise, an arrow in any diagram in this book represents a causal effect.
 Jagtap, V. S., Sarathi, V., Lila, A. R., Bandgar, T., Menon, P., & Shah, N. S. (2012). Hypophosphatemic rickets. Indian journal of endocrinology and metabolism, 16(2), 177.
 The term ‘homeorhesis’, meaning a stabilized flow, has also been proposed because reference sets are liable to change. The terms “allostasis” and “heterostasis,” are overlapping with “homeostasis” but are not generally adopted. See: Day, TA (2005). Defining Stress as a Prelude to Mapping Its Neurocircuitry: No Help from Allostasis, Progress in Neuro-psychopharmacology and Biological Psychiatry, 29, 1195–1200.
 Petersen, S.E. & Sporns, O. (2015) Brain networks and cognitive architectures. Neuron 88, 207 – 219.
 Edlow, B. L., McNab, J. A., Witzel, T., & Kinney, H. C. (2016). The structural connectome of the human central homeostatic network. Brain connectivity, 6(3), 187-200.
 Evidently this is the opinion of one of Bill Gates who holds that foreign aid helps to stabilize the developing world and thereby the security and stability of the USA. See: http://time.com/4704550/bill-gates-cutting-foreign-aid-makes-america-less-safe/
 Von Bertalanffy, L. (1968). General system theory. New York. See p. 210.