The idea of a “default mode network” has been around for most of the last decade. Like the abstract says, a common set of brain regions tends to decrease its metabolic activity when subjects are performing a task, rather than just lying in the scanner. This was first noted by Raichle:

http://www.ncbi.nlm.nih.gov/pubmed/11209064

It’s interesting to note that there’s a “stream of consciousness” even when a person is sitting there not doing anything in particular. The aim of zazen practice is to focus attention on one’s breathing. Sustained attention of this sort is difficult, and zazen practitioners are encouraged to simply return attention to their breathing when they notice that they’ve become distracted. If you try to sit like this for 10-20 minutes, you’ll notice a calm, quiet feeling accompanied by a subjective reduction in the frequency of verbal thoughts. More experienced practitioners are able to focus their attention for longer periods of time and recover from distractions more rapidly. There’s more to this than “woo,” as some evidence exists that meditative training really does alter attentional performance:

http://www.ncbi.nlm.nih.gov/pubmed/17672382
http://www.ncbi.nlm.nih.gov/pubmed/19181542

There is nothing wildly implausible about suggesting that spending hours training one’s attentional ability might actually work. It was reported in Science earlier this year that a mere 14 hours of training could improve working memory, a construct closely related to attention. These improvements were accompanied by reductions in cortical D1 receptor expression:

http://www.ncbi.nlm.nih.gov/pubmed/19197069

There’s nothing obscure about using the term “lexical decision task,” since it’s been around since the 1970s. You show people strings of letters. They tell you if those strings are words or not. Quite straightforward.

The brains of meditators of went back to what they were doing before each decision faster than the brains of nonmeditators, as one would predict if the claims of Zen practitioners are true.

The weak point here is equating “default mode activity” with the resting stream of consciousness in any simple way. The originator of the “default mode network” idea has also reported that functional connectivity of this same network of regions can be observed even in anesthetized subjects:

http://www.ncbi.nlm.nih.gov/pubmed/17476267

The significance of “default mode” observations to cognitive neuroscience has also been questioned:

http://www.ncbi.nlm.nih.gov/pubmed/17052921

Anyway, meditation seems to “work,” even after stripped of its religious trappings (“mindfulness based stress reduction”). Independently of its effects on cognition, meditation seems to reduce stress, albeit not magically more than relaxing in other ways:

http://www.ncbi.nlm.nih.gov/pubmed/19432513

Meditation might not reveal the ultimate nature of the universe to you and free you from the karmic wheel of birth and death, but it can make you feel and think a little bit better.

Meditation: at least it’s not homeopathy!

There are significant implications for this type of research in sleep, autism, ADHD, schizophrenia, and other disorders with a neurological component. I downloaded the paper to read more carefully.