The way they refute it is interesting. > Those authors observed that fluorescent dyes injected into the cerebrospinal fluid (CSF) via the cisterna magna penetr

The way they refute it is interesting. > Those authors observed that fluorescent dyes injected into the cerebrospinal fluid (CSF) via the cisterna magna penetrated further into the cortex during sleep and anesthesia. They interpreted this as showing that molecular movement into the cortex must be faster during these states. However, the concentration of dye in any brain region will always be the difference between its rate of arrival and its rate of departure and so increased dye penetration in sleep and anesthesia can be equally well explained by a reduced rate of clearance rather than an increased rate of entry. So the previous paper assumed that the presence of dye further into the brain meant more clearance, but these authors point out that the presence of more dye could mean less clearance (because if the fluid isn't being replaced as quickly the dye can simply diffuse further). The previous authors also injected dye into the CSF, which complicates matters as the CSF is moving both into and out of the brain all the time. However: > Our data in Figs. 2 and 3 show that, averaged across the brain, clearance is reduced by both sleep and anesthesia ... if you just focus on average dye concentration in the brain, and you inject the dye directly into the brain (the parenchyma) rather than into the CSF, the mean concentration is higher after a period of sleep than it is after a period of wakefulness. Which suggests less clearance, rather than more clearance, during sleep.