I’m not an HVAC engineer, but I’ve dabbled in simulating building airflow with computational fluid dynamics (CFD).
US residential HVAC standards include ANSI/ASHRAE 62.2-2013: Ventilation and Acceptable Indoor Air Quality in Low-Rise Residential Buildings. Ventilation frequency is typically quantified in air changes per hour (ACPH). The exact number depends on which room (kitchen vs walk-in closet) and the number of people typically expected to be in that room.
It’s been a long time since I’ve looked at that spec, but I believe it calls out something like 6-8 air changes per hour for a living room. The catch is that this number is found simply by dividing the room volume by the fan flow rate:
ACPH = (60 * flow per min) / room volume
So there’s no guarantee that all of the air in the room is scavenged with each air change—in fact, a large fraction is quasi-stagnant. So “eight air changes per hour” doesn’t mean all the air gets changed eight times per hour.
Realistically, maybe 50% of the air gets changed 16 times per hour. Or it could be that 25% of the air gets changed 32 times per hour. It all depends on the floor plan, inlet/outlet location and how well sealed the dwelling is.
That’s a big reason why cooking smells tend to linger well beyond the first nominal air change. The only places that get fully scavenged with each air change are wind tunnels.
To answer the OP’s question, a 400 square-foot apartment might contain 4500 cubic feet of air. (I’m assuming ten-foot ceilings with some extra padding).
A typical 10,000 BTU window AC unit might blow air at 280 cubic feet per minute (CFM) or 16,800 CF/hour. Dividing 16,800 CFH by the 4500 ft[sup]3[/sup] apartment volume, we get about 3.75 air changes per hour.
I believe Joey P has a lot of practical experience with this sort of thing, so I yield to his expertise. But this covers some of the theory.