While I enjoy hating on fMRI studies as much as the next guy, it's a little harsh to say that this study proves nothing. Brain activity is proportional to blood

While I enjoy hating on fMRI studies as much as the next guy, it's a little harsh to say that this study proves nothing. Brain activity is proportional to blood flow/energy consumption. We can argue about the linearity of that relationship, the spatial and temporal resolution, the problems of averaging to increase signal-to-noise and other interpretational issues, but it is still a proxy for neuronal activity. To use your analogy, I'd say it's more like looking at the spatial distribution of temperature in your CPU. Even if it's just driving the screensaver, at least now you know something about which registers are involved. Likewise, the authors in this study have implicated a couple of areas as being involved in the process of daydreaming. Where I would criticize this study, is the part where they conclude that there is a connection between daydreaming and problem solving on the basis that executive function areas are active during daydreaming. It's a suggestive link, but very tenuous.