NEAT: The Largest Movement Variable Nobody Tracks

Two people of the same height, weight, age and sex, both describing themselves as sedentary office workers, can differ by several hundred calories a day in what they actually burn. Almost none of that gap comes from the gym. It comes from everything else they do while awake: walking to the printer, standing on a call, taking stairs, shifting in a chair, carrying shopping, cooking rather than ordering.

Researchers call this non-exercise activity thermogenesis, usually shortened to NEAT. It is the most variable component of human energy expenditure, the least measured, and the one most likely to explain why a carefully calculated deficit stops working. It is also the component that quietly shrinks when you eat less, which makes it central to the plateau story.

The overfeeding experiment that defined the term

The foundational study is small, elegant and still the clearest demonstration in the literature. Levine, Eberhardt and Jensen, writing in Science (283:212-214, 1999), took sixteen non-obese adults and overfed them by 1,000 kilocalories a day for eight weeks, measuring the components of energy expenditure carefully throughout.

If the body were a simple ledger, every participant should have gained a similar amount of fat. They did not. Fat gain ranged from 0.36 kg at the low end to 4.23 kg at the high end, an eleven-fold spread on an identical surplus. The variable that explained it was NEAT. Changes in non-exercise activity ranged from a fall of 98 kilocalories a day to a rise of 692 kilocalories a day, and the size of that increase predicted how little fat a person gained.

The people who resisted weight gain were not exercising more. They were unconsciously moving more across the whole day, and their bodies were spending a substantial share of the surplus on activity they never decided to do. The people who gained most did not increase movement at all.

Read the direction of that finding carefully, because it is often reversed in summary. The study does not show that willpower determines outcomes. It shows that a largely involuntary response to energy surplus varies enormously between people, and that this variation is a real determinant of who gains weight easily.

A companion result from a different design points the same way. Bouchard and colleagues, in the New England Journal of Medicine (322:1477-1482, 1990), overfed twelve pairs of identical twins by 1,000 kilocalories a day for 100 days. Weight gain ranged from 4.3 kg to 13.3 kg across the group, but pairs resembled each other far more closely than they resembled other pairs. Whatever governs this response is substantially inherited.

Posture, not fidgeting

The popular version of NEAT collapses into advice about tapping your foot. The measurement work suggests something more structural.

Levine and colleagues returned to the question in Science (307:584-586, 2005), using instrumented undergarments that sampled body position and movement every half-second for ten days in lean and mildly obese volunteers, all of whom were self-described non-exercisers. The lean participants spent roughly two and a quarter hours more per day upright and moving. The obese participants spent that time seated instead.

The authors estimated the energy difference at around 350 kilocalories a day. More interestingly, when obese participants lost weight, their posture pattern did not shift toward the lean pattern, and when lean participants gained weight, theirs did not shift either. The tendency looked like a stable trait rather than a consequence of current body size.

So the actionable unit is not fidgeting. It is hours spent upright and in motion, accumulated in fragments nobody records: the shape of a commute, whether a job involves standing, how far the kitchen is from the desk, whether errands happen on foot.

NEAT falls when you cut calories

This is the part that matters most to anyone dieting, and the part least represented in calculator outputs.

When energy intake drops, total expenditure drops by more than the loss of body tissue alone would predict. Leibel, Rosenbaum and Hirsch established the scale of this in the New England Journal of Medicine (332:621-628, 1995), studying obese and never-obese subjects whose weight was deliberately altered under controlled feeding. A 10% reduction in body weight produced a fall in total energy expenditure meaningfully larger than the change in body composition accounted for, and the effect persisted while the reduced weight was maintained.

Rosenbaum and Leibel have since reviewed this literature repeatedly under the heading of adaptive thermogenesis. Some of the reduction sits in resting metabolic rate. A substantial share sits in the efficiency and quantity of movement, which is to say in NEAT.

The behavioural expression is familiar to anyone who has dieted properly. Several weeks into a deficit you take the lift without deciding to, you drive to the shop that you used to walk to, you sit down between sets, you stop pacing on phone calls. Formal training usually survives, because it is scheduled and visible. Everything unscheduled quietly contracts.

This is the mechanism doing most of the work in why weight loss plateaus happen, and it explains a pattern that otherwise looks like a contradiction: a person can be honestly compliant with a target that was correctly calculated, and still stop losing weight, because the denominator moved. Metabolic adaptation explained covers how much of the shortfall is resting rate versus activity.

What this does to a calculated target

Every predictive energy equation, including the ones on this site, estimates resting metabolism and then multiplies by an activity factor. That structure has a specific weakness: it assumes your activity category is stable, and NEAT is the least stable thing about you.

The multipliers are coarse by necessity. For a 42-year-old man of 82 kg and 178 cm, the Mifflin-St Jeor equation gives a resting rate near 1,728 kilocalories. The TDEE calculator then returns about 2,073 at the sedentary factor and about 2,375 at the lightly active factor. That single step between adjacent descriptions is roughly 300 kilocalories a day, which is well inside the range Levine's participants moved spontaneously. Choosing between two labels that both sound like a fair description of your life changes the answer by more than most people's deliberate exercise contributes.

Two things follow. First, an equation output is a starting hypothesis, not a measurement, and it carries roughly ±10% error before NEAT variability is added. Second, the correct response to a stalled deficit is often to restore movement rather than cut food further, since cutting further tends to suppress the component that stalled in the first place. How to create a calorie deficit works through calibrating a target against real results.

Raising it without turning it into exercise

The useful interventions are structural, because NEAT responds to environment more reliably than to intention. Deliberately trying to fidget more does not last. Changing where things are, and what a day requires of you, does.

Walking is the highest-leverage single change because it is cheap, repeatable and easy to accumulate in fragments. On the calories burned calculator, brisk walking for an hour comes to roughly 431 kilocalories for an 82 kg adult, and about 215 for half an hour. Split into three short walks it barely registers as effort, which is precisely the point: the fragments that do not feel like training are the ones that survive a busy week.

Beyond that, the changes worth making are unglamorous. Standing for part of the working day rather than all of it. Taking calls on foot. Parking or alighting further away by default, so the decision is made once rather than daily. Doing household tasks manually where the manual version is not genuinely worse. Keeping stairs as the default rather than a choice.

Two cautions. Standing desks are frequently oversold; standing still burns only marginally more than sitting, and the benefit comes from the movement that standing makes easier, not from the posture itself. And step counters measure ambulation, not NEAT, so a step target is a proxy for one slice of the picture rather than the picture.

Because these changes are habitual rather than motivational, the constraint is usually consistency rather than knowledge. How to build habits that survive month three is the more useful companion piece if that is where the difficulty sits.

What NEAT does not do

It does not replace training. Resistance work protects lean mass during a deficit and NEAT does not, so raising daily movement is a complement to structured exercise rather than a substitute for it. It also does not compensate indefinitely: the overfeeding studies show meaningful variation in the response, not an unlimited buffer, and the participants who increased NEAT most still gained some fat.

Nor is a low NEAT a character flaw. The posture work suggests a large inherited component, and there are plainly circumstances that constrain movement regardless of intent, including desk-bound work, caring responsibilities, disability, chronic pain and illness. The honest framing is that the variable is real and partly modifiable, not that it is fully under anyone's control.

Anyone whose daily activity has fallen sharply without an obvious explanation, or who finds movement newly painful, breathless or exhausting, should raise that with a clinician rather than treat it as a discipline problem, and when a health number warrants a doctor's visit covers how to frame that conversation. And if tracking movement or energy has become a source of distress, that is worth professional support rather than more precise measurement.

The reason NEAT deserves attention is not that it is a hidden trick. It is that the largest genuinely adjustable term in most people's daily expenditure sits outside the hour they think of as exercise, and outside every number they record.