While I haven’t read the study, this is an intriguing experimental approach to the study of food consumption! I’ve always been curious why – apart from the everpresent financial or practical reasons – I haven’t seen more studies that eschewed survey data. Keeping the menu the same but manipulating portions and ingredients seems to be a pretty plausible way to control the nutrient ratio with a minimum of intervening factors. Maybe there are plenty of similar studies and I just haven’t run across them? I sure hope so.