While there may be an argument to make for more risk-taking in science, with space travel itself being among the likeliest candidates, I don't see how it applies to this situation.
Remember that every one of these simulated missions provides hundreds or thousands of examples for indidividual human interactions each. That's an indication of the amount of data needed to draw applicable conclusions. Allowing people to die could never approach those levels. Any conclusions would also be tainted by the nature of these experiments as simulations far more than any other events–because "is the result worth the price" would be the dominating question.
While there may be an argument to make for more risk-taking in science, with space travel itself being among the likeliest candidates, I don't see how it applies to this situation.
Remember that every one of these simulated missions provides hundreds or thousands of examples for indidividual human interactions each. That's an indication of the amount of data needed to draw applicable conclusions. Allowing people to die could never approach those levels. Any conclusions would also be tainted by the nature of these experiments as simulations far more than any other events–because "is the result worth the price" would be the dominating question.