Self experimentation is pretty common in psychophysics experiments. I think a big part of it is that the experiments are long and boring, so the scientists themselves are the only people likely to pay attention and perform the task accurately the whole time.