(Edited to clarify question, I am a math professor not a biologist so I appreciate help clarifying the language to match the knowledge base present here.)
As I understanding nutrition labels, if a label claims a product P (e.g. an orange) has 50% of a recommended daily diet of some nutrient N (e.g. Vit. C), it is simply saying that an idealized diet recommends twice as much as of the nutrient (Vit. C) found in that product (Orange). This is different from saying that by eating the orange you body will receive 50% of that nutrient for several reason, including:
- Most digestion is not idealized so the comparison is at best an attempt at an average normalized to a specific capacity (2000 cal diet).
- Just because the product contains 50% of the nutrient does not mean the body will absorb all of that nutrient in digestion.
Setting aside the effects on individuals, I am interested in the behavior of 2. Are their products and nutrients where (on average) eating larger servings leads to decreased (or increased) absorption of a fixed nutrient? If so how much can the the amount absorbed differ from the naive multiple of the listed nutrition quantity.
In mathematical language, if A(p) represents the percent of recommended daily diet typically absorbed from consuming p units of a product, is the the derivative A'(p) generally considered to be constant (which we call "additive" or "linear")?
So in a sense, if A(p) was plotted as a graph then the nutrient label is like a tangent line locally approximating the graph at A(1). I want to understand how well the actual nutrition absorption compares to this tangent line approximation in real life examples. Are they generally good approximations, within the significant figures of the displayed data, or are they sometimes very far from average absorption as p varies.
Though possibly not relevant, USDA literature here documents various conversions between different units, raw and cooked food (p. 11-12), how much of a product is refuse (p. 5), and each appears to be additive. Though I did not read every page, this seems to suggest that any non-linear behavior is in fact owed to biological/chemical reasons and not a reflection of how the data is reported in labels.