Frontiers in Human Neuroscience has a great two page article that nicely summarises the thinking about how blood flow measured by brain scans relates to the workings of the neurons.
No one with common sense would believe that in a house, water movements in pipes could tell you how many lamps are on and how much fuel is used for heating. Surprisingly most neuroscientists are convinced that in the brain monitoring local cerebral blood flow (CBF) what I call plumbing, is a reliable surrogate method to localize electrical neuronal activity and monitor metabolic events.
The piece is by neuroscientist Jean Rossier, and he discusses the two main theories of how blood flow relates to what the neurons are doing.
The ‘metabolic hypothesis’ assumes there is a causal link between how much energy the neurons need, in the form of glucose, and the subsequent regulation of blood flow in the brain. In other words, the neurons signal the need for energy, which is delivered later.
The ‘neurogenic hypothesis’ hypothesis, suggest that blood flow can be ‘pre-ordered’, in anticipation of neural activity.
Needless to say, it’s important to understand the exact relation between the operation of the neurons and blood flow, because brain scanning studies typically measure blood flow to infer the working of neurons and hence the relationship to cognitive or mental processing.
The Frontiers in Human Neuroscience article is a concise piece which discuss the neuroscience of this link, and covers some of the most recent studies which have attempted to make sense of what brain scans tell us when we’re doing psychology experiments.
Link to article ‘Wiring and plumbing in the brain’.
Link to PubMed entry for same.
Mind Hacks 
Thanks for this. Terrific metaphor which will hopefully deflate some of the more sensationalist claims being made.
Actually, this strikes me as a particularly stupid metaphor. Given that water flow in a house, in the vast majority of cases, has little to do with power generation, the analogy is very poor. Since blood flow does indeed deliver glucose which, in the end, neurons need to fire the comparison is, at best, irrelevant.
A more apt metaphor along these lines can be seen in, for example, my house which is heated by steam heat. If I were to monitor the water flow in a roughly 3m x 3m x 3m cube in my basement I could estimate my heating requirements quite accurately.
Now, the hypothesis may certainly be suspect but this argument doesn’t really advance its cause much. This: “Needless to say, it’s important to understand the exact relation between the operation of the neurons and blood flow, because brain scanning studies typically measure blood flow to infer the working of neurons and hence the relationship to cognitive or mental processing.”
on the other hand, is correct (needless to say indeed!).