interesting. small sample size though, and probably not obtain from a large spectrum of society. the selection of slim people might have also been a limiting factor in coming to a determination of results.
pre-existing health conditions were also not considered that may influence diets, and neither was there an indication which showed that there was complete monitoring of caloric intake (fine and dandy to say you must eat 10k worth of calories per day, but if nobody enforces that on you, you'll stop after reaching the halfway mark due to you feeling sick. believe you me).
would love to see an expanded study on this, though.