Intermittent fasting or caloric restriction?

Intermittent fasting and caloric restriction are associated with many of the same benefits (Mattson, 2005). In trying to sort out the differences between these protocols, people will often treat IF and CR as if they are completely different things. It's not clear to me that they are entirely dissociable. I've previously mentioned that it is common for intermittently fasted rodents to eat slightly less than ad libitum controls. This varies from study to study and with the duration of dietary restriction (as does the degree of bodyweight difference between IF and control animals). The data in the figure below are from rats (Leveille, 1972) and mice (Anson et al., 2003), respectively.

It's worth noting that the C57BL/6 strain of mice (right panel) are intermediate (relative to other mouse strains) in terms of how much less IF mice eat than controls (Goodrick et al., 1990). And while Anson et al. (2003) claim a dissociation between IF and CR, the truth is that their animals were calorically restricted for much of the study. The question in my mind is whether both dietary protocols tap into the same underlying mechanism? Or can IF and CR interact to produce better than predicted results? The problem is that IF involves some CR and CR involves some IF. The following quote is from Leveille's paper, and is worth reproducing. Note that the term "meal-fed" refers to an intermittent diet where access to food is provided for only 2 hours each day and "nibbling" refers to ad libitum food access.

The results of the longevity study (experiment A) are difficult to interpret. Although the meal-fed animals had a significantly longer life-span than the nibbling rats the difference cannot be ascribed to meal-eating, per se. In addition to the metabolic differences between the meal-fed and nibbling rats there existed a significant difference in body weight and presumably in body fat. Thus these results can be interpreted as supportive of the classic experiments of McCay et al. (21, 22) showing that rats whose food intake was restricted had a significantly longer life-span. However, it is important to note that in McCay's studies as well as other similar studies (23,24) the restricted animals were very likely "meal-fed." It has recently been noted that restricting rats to 80% of ad libitum intake quickly results in a shift in eating pattern in which the animals ingest their daily food allowance within a 2-hour period (25). Thus one cannot determine unequivocally whether the increase in life span is due to the reduced food intake or to the meal pattern.

Masoro (2004) makes a similar point, and while his earlier study (Masoro et al., 1995) attempts to tease apart IF and CR, even he admits that the evidence does not conclusively eliminate the role of meal patterning. It seems that studying the meal patterning of individual animals within days may help to clarify this issue. This may also bear on the confusing results from human studies, since meal patterning is often not reported and maybe not even recorded.