"... his finding also suggested some of the papers were actually reporting false positives, in other words noise that looked like data. He urged researchers to

"... his finding also suggested some of the papers were actually reporting false positives, in other words noise that looked like data. He urged researchers to boost the power of their studies by increasing the number of subjects in their experiments." Conclusions based on low sample rates, should be seen as poor technique, an indicator of potential bias and suggest a lot more validation is required before acceptance.