prase comments on Error detection bias in research