The glass of water on the table looks still.
The person holding it is not.
Their blood pressure shifts as they stand. Body temperature is defended while the room changes. Blood glucose rises and falls around meals. Breathing adjusts to movement. Fluid balance, pH, oxygen delivery, electrolytes and countless cellular processes are being regulated continuously, usually without becoming conscious experience.
This is the first correction homeostasis asks of ordinary language: stability in a living system is not the same thing as immobility.
Definition
Homeostasis is the self-regulating process through which an organism maintains internal variables within viable ranges while adapting to changing internal and external conditions. It depends on sensing, comparison, feedback, coordinated effectors and higher-order regulation. The relevant variables are not all held at one exact value. Many oscillate, vary by time of day, developmental state or context, and are controlled through interacting systems rather than one isolated loop.
The modern term is associated with Walter Cannon, building on Claude Bernard's nineteenth-century idea of the milieu intérieur. Contemporary physiology uses homeostasis as a central organizing principle for understanding how living systems preserve conditions compatible with function.
The thermostat analogy is useful, and insufficient
Homeostasis is often introduced through a thermostat. A sensor detects deviation from a target, a controller compares actual and desired temperature, and a heater or cooler acts to reduce the error.
That analogy teaches negative feedback well. It also risks making physiology look simpler than it is.
A person is not a room with one thermostat. Arterial pressure, temperature, osmolarity, glucose availability, respiratory gases and energy balance interact. Control systems can be redundant. Some responses are anticipatory as well as corrective. Higher neural centers can alter reflexes according to context, learning and behavior. A variable that is tightly controlled in one situation may shift its operating range in another.
Billman's 2020 review therefore emphasizes homeostasis as dynamic regulation, not a static condition. Multiple feedback systems interact, and hierarchical control provides flexibility rather than merely returning every value to a single immutable set point.
What homeostasis is not
Homeostasis is not “being calm.”
A person can be emotionally distressed while many physiological variables remain effectively regulated. Conversely, someone can feel calm while glucose, blood pressure or temperature regulation is impaired.
Homeostasis is not synonymous with health. Disease can involve homeostatic compensation that preserves one variable temporarily while creating costs elsewhere. A regulatory system can also maintain a value through increasingly intensive compensation.
Homeostasis is not the same as allostasis. The two concepts overlap historically and theoretically, but allostasis places greater emphasis on anticipatory, context-sensitive adjustment and on changing regulatory demands across time. Homeostasis asks how essential variables remain within viable bounds. Allostasis asks more explicitly how the organism changes regulatory settings and coordinates resources to meet predicted need.
Nor is homeostasis the same as allostatic load. Allostatic load is a research framework for cumulative multisystem burden associated with repeated or dysregulated adaptation. A single successful homeostatic correction is not cumulative load.
Feedback is the core logic
Negative feedback reduces deviations from a regulated range. If arterial pressure rises, baroreceptor signaling changes autonomic output in ways that tend to reduce pressure. If temperature falls, thermoregulatory responses can reduce heat loss and increase heat production. If blood glucose rises, endocrine responses help move it toward an appropriate range.
Positive feedback also exists in physiology, but it serves a different function. In processes such as blood clotting or parturition, a response can amplify itself until a terminating event occurs. Calling all physiology “negative feedback” is therefore too simple.
Feedforward regulation matters too. The body can respond before a variable has drifted dangerously. Salivation before food reaches the stomach, cardiovascular adjustment at the start of movement, and anticipatory endocrine responses show that regulation can be prepared by cues and prediction.
The distinction matters for Inner Technology because “regulation” is often used as if it meant correcting a failure after it appears. Living systems frequently regulate by preparing.
Measurement follows the variable
There is no single homeostasis score.
Temperature is measured differently from glucose, blood pressure, pH or osmolality. Each variable has its own sensors, effectors, time scales, normal ranges, measurement error and clinically meaningful thresholds. Even within one variable, office measurements, ambulatory measures and continuous monitoring answer different questions.
The homeostatic concept therefore belongs at the level of regulatory organization, not as a new consumer metric.
A wearable can measure heart rate accurately and still tell us little about “whole-body homeostasis.” A glucose sensor can reveal glucose dynamics without measuring thermoregulation, immune function or cardiovascular control. More sensors do not automatically create one valid global regulatory score.
Regulation happens across time scales
Some corrections occur in seconds. Others unfold over hours or days. Still others depend on developmental adaptation.
This creates a common interpretive problem. A snapshot may catch a variable inside its acceptable range while missing the amount of regulatory effort required to keep it there. Repeated measurement can reveal variability, recovery, drift and compensation that a single reading cannot.
The inverse is also true. A temporary excursion outside a usual range is not necessarily dysregulation. Exercise deliberately changes heart rate, ventilation, temperature and metabolism. A meal changes glucose and endocrine activity. Sleep changes autonomic balance and hormone release.
Physiological meaning depends on timing and context.
Homeostasis and behavior are entangled
Humans regulate by acting in the world.
We put on a coat, drink water, seek shade, eat, sleep, move, change posture and alter the environment around us. These behaviors are not separate from physiological regulation. They are part of how regulatory demands are met.
This is one place where simplistic “inside versus outside” models fail. An organism can maintain internal conditions by changing external conditions. Technology enters the same loop. Heating systems, medication reminders, insulin pumps, navigation aids and workplace schedules can all reduce or redistribute regulatory demands.
The relevant scientific question is not whether technology is “natural.” It is which variable is being supported, through what mechanism, with what reliability, and whether external support changes the organism's own regulatory work in ways that matter.
Dysregulation is not one thing
Homeostatic failure can arise through damaged sensors, weak effectors, altered feedback, persistent disturbance, conflicting regulatory demands, medication, disease, environmental exposure or developmental change.
The word dysregulated should therefore be used with care. It is not a diagnosis by itself. It does not identify mechanism. In popular discourse, “my nervous system is dysregulated” can collapse many distinct possibilities into one persuasive phrase.
A scientific account needs to ask: which system, which variable, measured how, over what time scale, compared with what reference, and with what functional consequence?
Without those questions, homeostasis becomes metaphor rather than physiology.
Set points, ranges and regulated variables
The language of a set point can be useful, but it should not be applied mechanically.
Some physiological systems behave as though regulation is organized around a preferred value or narrow range. Others show broader zones, thresholds, seasonal changes, developmental shifts or context-sensitive operating points. The controlled variable can also be maintained through several effectors at once.
For blood pressure, for example, cardiac output, vascular resistance, blood volume, renal regulation and autonomic reflexes interact. A pressure reading is therefore an outcome of coordinated control, not evidence for one hidden dial.
The same caution applies to glucose, temperature and body weight. A model can be useful without implying that one immutable number is stored somewhere in the brain.
This matters because popular “reset your set point” language often moves directly from a control-theory metaphor to intervention claims that have not been demonstrated.
Compensation can preserve the variable while changing the cost
Homeostatic success can hide compensatory work.
A regulated variable may remain inside an acceptable range because multiple systems are working harder, because medication is supplying an external correction, or because behavior is continually compensating for environmental demand.
This is one bridge from homeostasis toward allostasis and allostatic load. A system can preserve immediate viability while the pattern of compensation changes over time.
That does not mean every compensation produces disease. It means the visible variable is not always the whole regulatory story.
For research and design, this is an invitation to look beyond whether a person appears functional. Ask what supports, trade-offs and repeated adjustments are making that function possible.
Homeostatic metaphors should not be imported carelessly into psychology
Psychology often borrows the language of equilibrium. People are said to return to emotional baselines, restore balance or regulate toward a preferred state.
Sometimes those metaphors are useful. They become misleading when they imply a validated physiological controller where none has been measured.
Emotion is not body temperature. A relationship does not have a blood-pressure set point. Meaning-making is not a thermostat.
The Institute can use homeostatic language only when the analogy is explicit and its limits remain visible. Scientific concepts become weaker, not stronger, when they are stretched until everything looks like them.
What the evidence does not show
The evidence does not show that every desirable psychological state corresponds to a single homeostatic optimum. It does not show that feeling calm proves physiological regulation. It does not show that a brief rise in heart rate or cortisol is a failure of homeostasis. It does not show that one wearable-derived score captures whole-body regulatory health. It does not show that every chronic illness can be explained as generic “dysregulation.”
It also does not show that homeostasis and allostasis are mutually exclusive theories. The concepts emphasize different aspects of regulation and are used differently across literatures.
Institute synthesis: stability is active work
The Institute uses the phrase stability is active work as a translation principle.
This is synthesis, not a new physiological construct.
The point is to resist a cultural mistake: treating regulation as the absence of movement. A resilient system is often visibly changing because conditions are changing. It accelerates, inhibits, redistributes, anticipates and recovers.
That matters for human development. A person who can remain capable in a difficult conversation may not feel still. Their achievement may be the ability to stay inside a workable range while emotion, attention and bodily activation move.
From individual physiology to institutional design
Institutions also create regulatory demands.
A workplace can require repeated sleep disruption through scheduling. A school can compress eating, movement and recovery into narrow timetables. A hospital interface can increase cognitive and autonomic load by producing alarms without hierarchy. A digital product can create repeated attentional capture while calling the resulting fatigue a personal self-management problem.
Homeostasis does not make these institutional effects morally self-evident. It does make one principle visible: environments participate in the conditions organisms must continually regulate against.
The better design question is therefore not “How do we make people calmer?”
It is “What regulatory work does this environment repeatedly require, and is that work compatible with health, learning and agency?”
Why this matters for Inner Technology
Homeostasis places a hard scientific floor beneath the language of regulation.
It reminds the Institute that regulation is specific, measurable and embodied. It requires us to distinguish physiological variables from psychological metaphors, momentary activation from dysfunction, and successful compensation from low cost.
It also adds humility.
The body is not trying to become motionless. It is continuously negotiating viable ranges in a world that does not stop moving.
Inner Technology should not teach people to worship stillness. It should help them understand when movement is adaptive, when compensation is becoming costly, and when the environment itself needs to change.
Research references
- Billman GE. Homeostasis: The Underappreciated and Far Too Often Ignored Central Organizing Principle of Physiology. Front Physiol. 2020;11:200. DOI: 10.3389/fphys.2020.00200.
- Cannon WB. The Wisdom of the Body. New York: W.W. Norton; 1932.
- Sterling P. Allostasis: a model of predictive regulation. Physiol Behav. 2012;106(1):5-15. DOI: 10.1016/j.physbeh.2011.06.004.