Testing by Simulations

We could do an awful lot of testing by simulations, and even by running real hardware against a simulation of the rest of the system. For example, the Boeing 777 was certified without an extensive program of real-life flight, largely on the basis of simulations.

This is why there are so many math models in the "things you can do" list. And somewhere in there might be a real microwave oven and a control panel for the communications system, being tested for its function in the integrated system, while the ground support center practices receiving data from remote sites (We fake it with geosynch comsats, but the remote stations are real.).

But none of that will tell us how the whole vehicle, plus the crew, will perform in zero gravity. The underlying question is: how confident are we of our ability to imagine what it will be like for the crew? A lot of us have put hours of skull sweat into trying to imagine what it would be like, and to communicate those thoughts in stories, but is imagination good enough?

To argue the opposite point for a moment, though: do we have any evidence from past space flights that imagination was not good enough?