Segregation refers to specialization, the idea that different parts of the brain do different things. Integration refers to communication or networking, the idea that to do something, it is necessary for those different parts of your brain to work together. These ideas are not opposites; they are complementary. They are different sides of the same coin.
A big change in brain mapping was seen at the 15th annual meeting of the Organization for Human Brain Mapping, this week in San Francisco. In past years, the focus at this meeting was on segregation, but this week there were hundreds of presentations on integration.
When imagers talk about "functional activation" they are using the language of segregation. The model is that sensory stimuli cause brain activity. When imagers talk about "functional connectivity" they are using the language of integration. The model is that there is always ongoing activity within distributed brain networks, and that stimuli modulate this ongoing intrinsic activity.
The "activation" imager asks "where?" while the "integration" imager asks "how?".
Last Wednesday, there was a satellite symposium at Stanford University, dedicated to functional imaging of the resting state. This makes no sense at all to activation imagers - how can you image activation in the absence of stimuli? But to integration imagers, it is, perhaps, perfection itself, to image the resting/restless brain, while the participant does nothing. Just as Seinfeld was a (wonderful) show "about nothing", this was a (wonderful) symposium "about nothing".
Blaise Pascal (1623-1662) wrote that "All of man's troubles stem from his inability to sit quietly in a room alone." But the restless brain reveals its functional organization, even when we try to sit quietly.