The Dose-Response of Eating: What the Literature Actually Says

At mastery, you no longer need rules; you need the underlying model, its uncertainty, and the judgment to bend it. What follows is the mechanistic and meta-analytic picture, with the caveats the headlines leave out.

Energy balance is real — but the inputs aren’t fixed

The first law of thermodynamics is not optional: sustained fat loss requires a sustained energy deficit. The error in popular discourse is treating "calories in" and "calories out" as independent constants. They aren’t. Energy expenditure is a dynamic system: as you lose mass, basal metabolic rate falls; under-eating suppresses non-exercise activity thermogenesis (NEAT) — the fidgeting, posture, and spontaneous movement that can swing daily expenditure by hundreds of calories; and the thermic effect of feeding varies by macronutrient. The ledger balances, but the body actively moves the columns to defend its stores. This adaptive component (sometimes called adaptive thermogenesis) is why a deficit that worked in week 2 may stall by week 10.

The protein curve, read honestly

The most-cited anchor is Morton and colleagues’ 2018 meta-analysis (49 studies, 1,863 participants), which reported a "breakpoint" around 1.62 g/kg/day beyond which supplemental protein produced no further gains in fat-free mass during resistance training. This is widely repeated as a hard ceiling. It isn’t.

Practical synthesis: ~1.6 g/kg/day is a sensible floor for trained lifters seeking hypertrophy; pushing toward ~2.2–3.0 g/kg costs little and plausibly helps at the margins, especially in a deficit (protein’s muscle-sparing and satiety effects both scale up when energy is scarce) and with age, where anabolic resistance raises the per-meal threshold for maximally stimulating muscle protein synthesis.

Why protein dominates satiety: the leverage hypothesis

The protein leverage hypothesis (Simpson & Raubenheimer) proposes that humans prioritize hitting a protein target and will keep eating until they reach it. In food environments engineered to be protein-dilute and energy-dense, satisfying the protein drive means over-consuming total energy. It’s a compelling explanation for why high-protein, whole-food diets self-regulate intake without conscious restriction — and why ultra-processed environments don’t.

Partitioning, fiber, and the gut

Where a surplus or deficit lands — muscle versus fat, the "p-ratio" — is modulated by training status, the protein and resistance-training stimulus, sleep, and genetics, not by calories alone. Fiber, meanwhile, is doing more than adding bulk: fermentable fibers feed gut microbiota that produce short-chain fatty acids influencing satiety signaling and insulin sensitivity. The 25–35 g/day target is a floor with real downstream metabolic consequences, not a digestion footnote.

The honest limits

Most nutrition RCTs are short, small, run on young men, and rely on self-reported intake — which is systematically under-reported. Individual variation in metabolic response is large and partly genetic. Compensatory behaviors confound free-living studies. The defensible posture is to treat the literature as a Bayesian prior — strong on direction (energy balance, protein adequacy, fiber, food quality), genuinely uncertain on exact magnitudes — and to weight your own well-tracked n=1 response heavily once you have it.

The science gives you the shape of the curve. Your own data tells you where on it you sit.

Sources

• A systematic review, meta-analysis & meta-regression of protein supplementation (Morton et al., 2018)Br J Sports Med · PMC
• Dose–response relationship between protein intake and muscle mass (Tagawa et al., 2020)Nutrition Reviews
• Protein Science: beyond the 1.6–2.2 g/kg ruleStronger by Science
• Recent Perspectives on Dietary Protein for Muscle HypertrophyNutrients (MDPI)