[trentlott]

They probably don't have much choice for where they did it. The animals likely couldn't see it, and the stress levels were probably minimal. It's not likely to be enough to harm the experiment.

While, yes, everything can be a variable, the effect wouldn't be big. I imagine the person running the lab had done this before and has gotten good results: rats are expensive and a hassle to keep. If his experiments had been 'screwed over' because of practices like that, his PI probably wouldn't be able to gather enough useful data to afford funding for animal use.