Ordinary discourse is notably self-centered, to judge by the range of hyphenates now in everyone’s vocabulary: self-interest, self-esteem, self-expression … The list in my notes reached some two dozen items, with no end in sight, before I abandoned the effort as (let’s get this over with) self-indulgent.
It’s an awful lot of emphasis to put on something so difficult to define and even harder to locate. The most common and experiential notion of the self pictures it as a sort of ghostly occupant of the space in one’s head that —according to studies from the past decade or so—sometimes extends into the upper chest. Gregory Berns, a professor of psychology at Emory University, uses MRI scans to conduct a more thorough search in The Self Delusion: The New Neuroscience of How We Invent―and Reinvent―Our Identities (Basic Books).
The title, while snappy, misdirects the reader’s attention somewhat, insofar as the author very quickly identifies three selves (none of them necessarily delusional) living as roommates in one’s cranium. These are the past-self, the present-self and the future-self. They operate in various parts of the brain—though it would be more apt to call each the result of an array of activities taking place across the brain. Remembering an event (or reading, or weighing a decision) is a process distributed throughout different regions.
That much will be reasonably familiar to many readers, and Berns shifts attention instead to more challenging concerns. “When memory research began to take off,” he writes, “most researchers focused on how and where the brain stored different types of memories, paying relatively little attention to how memories get sewn together.” The past-self may have access to childhood experiences, knowledge of the multiplication table and the ability to play the guitar. Each involves memory, but of unrelated sorts: encoded in different ways, stored in distinct areas of the brain and retrieved with what can be very uneven degrees of accuracy.
What Berns call the present-self experiences itself as persisting from moment to moment in the midst of its surroundings. (I think, therefore here I am.) But its awareness and durability are limited. At one point the author suggests the present-self spans about two seconds’ worth of consciousness, on a generous estimate, with the contents of experience continuously downloading into the memory system. Some of it is stored in long-term memory, but in compressed form, “using algorithms similar to those for streaming audio and video.” Meanwhile, elsewhere in the gray matter, other operations are underway to anticipate what the organism will experience or need to do next.
The future-self’s monitoring and prepping goes on continuously whether the present-self is thinking ahead or not—while the latter is, most of the time, responding to material served up by the past-self. One feature all of these systems share is that they work with incomplete information. Storing experience through compression involves deleting some of it, then filling in the gaps upon recall. “Even our most vivid memories of key events in our lives contain a high degree of confabulation,” Berns writes. “Memories are like blind spots. The brain fills in what isn’t there with a simulation.”
Likewise, the operations of the present- and future-selves involve working around lacunae. The various processes feed into each other. A given experience “occurs only once,” Berns writes, “and then it passes into memory. As we’ve learned, every time it is remembered, it must be reconstructed, like a movie editor stitching together frames. Every time it is reconstructed, it gets corrupted a little bit by whatever is happening at that instant.” Since the brain “tries to stay one step ahead of the world by constantly predicting what’s going to happen next”—based in part on previous experience—strange feedback loops can emerge.
Like memory, a short article on The Self Delusion can only be the product of omission through compression. The losses are more than granular. Berns popularizes the findings of his own research as well as that of his colleagues, including work from outside neuroscience. The reviewer might best compensate for necessary omissions by emphasizing a point of overlap especially dense with thought-provoking implications. Such is the case with Berns’s emphasis on narrative. It brings together both the question of how the individual’s triune self is bound together and the fact that a number of such complex selves can hold shared memories and feel a common identity. As with any other ability, the power to condense perceptions into so many zip files and reconstitute them later as memory takes time to develop. Object permanence comes first, and even that is not sufficient.
As the brain continues to develop, the range of experience expands, along with the capacity to store it. And some of that experience is vicarious. As Berns writes, a child “begin[s] to incorporate contemporaneous events that happened to other people in their life—particularly family members. By hearing and telling stories to parents and extended family, the child begins to assimilate a version of the past that is based not only on firsthand experience but also on shared knowledge.”
The tales are not just filed away but become models, both for present-self behavior and for how memories are organized and activated. “Because these stories create templates for all stories that follow,” the author points out, “they implicitly bias the perception of each new bit of information. The significance of incoming events may not be judged by the objective truth of the events but by how well they fit an ongoing narrative.” (Many people later develop a more complex perspective on reality, though it seems clear that potential can atrophy.)
It might not be too wide of a stretch to say that the self is a story the brain tells about its own contents. But the most intriguing aspect of the author’s research concerns the possible impact of narrative on the fully grown individual’s brain while at rest. This involves functional magnetic resonance imaging (fMRI), which generates an extremely detailed picture of brain activity over the period it is being scanned.
A continuous scan of the brain of someone asked to relax for 10 minutes will show “coordinated activity in disparate regions,” the author says, “where the measured signals fluctuate up and down in synchrony. These are called resting-state networks or, sometimes, default mode networks, because they represent the default mode of activity when a person isn’t doing anything.” In what Berns calls “the most complicated experiment [he] had ever designed,” he gathered about 7.5 minutes’ worth of resting-state fMRI data from 19 subjects.
The first five days created the baseline for each subject. Then he had them read a historical novel, Pompeii by Robert Harris, in nine installments over as many days, with tests to ensure they’d done the reading and questionnaires to determine how engaging they found it. After finishing, they underwent another five days with fMRI sessions.
The whole experiment unfolded over 19 consecutive days, generating a group profile of changes in the resting brains during, and for a period following, the reading of a novel. Berns expected to find changes in emotion-related areas. What emerged instead was “a network of regions organized in a spoke-and-hub pattern, with the hub centered on an area of the left temporal lobe called the angular gyrus … known to be critically involved in language comprehension.”
An increased connectivity between these parts of the brain developed during the period of novel reading; it faded afterward. A longer-term pattern persisted throughout and following the reading, occurring within “the bank of folds … where tactile impulses enter the cortex and motor impulses leave.”
Berns suggests Harris’s fictional depiction of “Roman feasts and orgies and the showering of molten ash on the inhabitants of the city” stimulated that portion of the brain. Stimulated it—and, to whatever degree, changed it, along perhaps with some element of the readers’ selves.