The Portion Size Effect

What it is

The clean experimental demonstration belongs to Barbara Rolls and colleagues, in 2002. Fifty-one adults were served macaroni cheese at four different portion sizes across four weekly lunches, and were free to eat as much or as little as they wanted. Larger portions produced larger intakes. Participants did not report feeling any more full for having eaten more.

Two terms are worth glossing, because both get used loosely. Portion size is the amount put in front of a person, before they decide anything. Satiation is the sense of having had enough that builds during a meal and stops it. The finding is that the first shifts how much gets eaten without shifting what the second reports, which is why nobody experiences this happening to them.

Portion size is established. Plate size is a different claim and is not. The popular version of the plate claim, that a smaller dish makes a smaller meal feel like more, runs through a laboratory whose data cannot be trusted, and this site keeps the two apart deliberately. See Plate Size.

In effect

The review

Hollands, G. J., Shemilt, I., Marteau, T. M., Jebb, S. A., Lewis, H. B., Wei, Y., Higgins, J. P. T., & Ogilvie, D. (2015). "Portion, package or tableware size for changing selection and consumption of food, alcohol and tobacco", Cochrane Database of Systematic Reviews. Sixty-one studies and 6,711 participants. People consistently consumed more when given larger portions or packages. The effect was small to moderate, and the evidence was graded moderate quality, marked down for incomplete reporting in the underlying studies.

One figure that needs handling

The review's authors estimated that removing larger portions altogether might reduce daily energy intake by up to roughly a sixth in adults in the United Kingdom, and rather more in the United States. That is a modelled projection rather than something anyone measured, and it should never be quoted as a measured result. It is the kind of number that travels well and means less than it sounds.

What this describes

An environment. Restaurants, packages, serving dishes and default sizes are the things this evidence is about. The effect is reliable at the level of a population eating from a food supply, and it says nothing about what any individual should put on their own plate.

What it does not say

It does not say larger portions make people feel fuller. The point of the finding is the opposite. Intake moves and the reported sense of fullness does not, which is why self-report is a poor instrument here. See The Self-Report Problem.

It does not say the effect is large. Small to moderate, on moderate-quality evidence, is the honest description.

It does not support the plate-size claim. That is a separate proposition with a separate and much weaker record.

It is not a weight-loss technique and this site does not present it as one. Nothing here licenses telling anyone what to serve themselves, and no page on this site attaches calorie targets or eating plans to it. See YMYL.

---