I might be wrong, but I don't think the battery consumption is related to the bit/sampling rate you use. Even more when there are no mechanical parts involved.
They are more related to headphone current demands, phantom power and LEDs on.
Does the manual specify anything on that point?
Once again: we badly need someone who do serious tests on the H4. with measured data and all.
I feel you have a point about recording mode, indicators, and phantom power having effect on battery life, likely ONLY phantom power being used or not and for how much power the connected mic is drawing is significant in making real difference in length of recording time.
No manual so far seems to offer any information on specific power demands with deck modes.
While I have yet to do a review of the new Zoom H4, while doing the bench tests for the reviews of MT and R-04 on my site, I didn't notice significant power changes with recording modes or indications.
The H4 is quite different from those two decks in having an extensive 'effects' function that may change power in a more significant way, but as mentioned, somebody needs to measure this and post the results.