Friday, January 30, 2009

Neuroscience gets a right and left hook

First a "leaked" study in press in Perspectives on Psychological Science confirmed suspicion focused on the use of inappropriately adjusted correlations in many studies using fMRI. Edward Vul and colleagues surveyed 54 articles that do not contain sufficient information in the method section to confirm how the analyses were produced and find over half of the studies to contain "grossly" inflated correlations.

The damning conclusion was that "a disturbingly large, and quite prominent, segment of social neuroscience research is using seriously defective research methods and producing a profusion of numbers that should not be believed."

The article is freely available here: Voodoo Correlations in Social Neuroscience

The week in neuroscience then went from voodoo correlations based on questionable research methodology to a questioning of the very foundations of the core technique of neuroscience: functional brain imaging. Blood flow changes picked up by fMRI scanners may not reflect the metabolic demand produced by local neural activity is the message that comes across strongly in a letter to Nature by Yevgeniy B. Sirotin and colleagues this week. The authors monitored neural activity and blood flow in two monkeys (they only need one intact one to show fMRI inaccuracies) whilst performing visual tasks. The key finding was that between stimuli there was virtually no neural activity but increased blood flow was observed suggesting that anticipation of expected tasks rather than neural activity may be a driving force behind blood flow changes.

Commentry: Another shock for brain imaging research

1 comment:

Justin Marley said...


Happy New Year! Any chance of some constructive feedback on a video I put together on the above study?