The 919 uses absolute air pressure sensing and temperature sensing within the airbox to derive an air density value.
Then the RPMs and Throttle Position inputs get added in to the mix to net a mass air flow rate.
The coolant temp input is a bit of a kicker.
It will enrichen at lower temps to account for the fact that cold engines like some added richness.
Then it leans things out as the coolant comes up to normal temp.
BUT when the temp gets higher than normal, it does not lean things out as one might expect, instead, it starts enrichening as a power (heat) reducing and quenching effect (which would easily evidenced by increased unburned HCs and increased CO).
This is why controlling coolant temp on 919 dyno runs is so crucial in order not to end up with a map unwittingly leaned out to correct for the over temp over richness the ECU will cause.
919s are high volume X low radiant heat area cooling systems. They are not designed for adequate heat rejection rates for extended periods of full power output, noting that very dry lighter high altitude air will reveal this sooner than humid sea level air.
For more, and very authoritative insight into the coolant temp richness protocol, check out 919.org and look for the two part dyno paper. (Part 1 by Dan Kyle, Part 2 by LDH.)