That's not at all what that paper says. It describes that blood glucose response to various foods can differ between people (and shows to extreme outliers as examples). Your interpretation with "3000 calories of cookies" vs rice is both not supported by the study, and also nonsense.
I'd look up the studies for this if I had the time, but: the actual correlation (relationship) between calories and weight gain/loss is extremely large, with the differences between most people of similar body types less than 200 calories/day.