David Hume famously argued that when he went looking for the self by introspecting his own first-person experience he was unable to find it. More recently Kai Vogeley and I argued that neuroscientists have a similar problem.

When they go looking for the self, using advanced imaging technology, they find the self both everywhere and nowhere in the brain. That is, so many areas of the brain activate under different conditions involving self-reference or self-related tasks that no one area can be defined as self specific.

Some theorists point to the default network which activates as a resting state when we are not focused on some task in the world, but rather may be self-ruminating. In fact, however, numerous experiments have shown that the default network, overlapping with cortical midline structures, which already covers a quite large set of brain regions, activates for self-other differentiation, which means it involves others and not just self. These areas are involved in social cognition (theory of mind) tasks when we attempt to understand the actions and minds of others. They may also be involved in various kinds of reflective judgment tasks.