Matt_Simpson comments on Error detection bias in research