Much of what makes us human can be traced to the cerebral cortex, a wrinkled sheet of gray tissue about an eighth of an inch deep that forms the outermost layer of the brain. It is the source of our language ability, home of our capacity to reason, and may be the seat of our consciousness. But how does the human brain become so convoluted, while those of many other mammals are perfectly smooth? What dictates where and how the cortex folds? David Van Essen, a neuroscientist at Washington University in St. Louis, offers a simple explanation. A tug-of-war among developing nerve cells, he says, pulls the brain into shape.
Without its convolutions, the cortex would take up three times the area it does in its wrinkled state. For something with the surface area of a very large pizza to fit inside the skull, it has to crumple up, says Van Essen. But researchers have never been sure what makes the cortex crumple. They do know that the folding starts in the sixth month of development; before that, the embryonic brain is smooth. Researchers have suggested that some cells are genetically programmed to grow faster than others and that this uneven growth causes cortical tissue to buckle and fold. But Van Essen doesn’t think that specific genetic instructions for such uneven growth are necessary to explain the folds. The brain’s wrinkles, he argues, could be the natural by-product of its attempt to wire itself as efficiently as possible.
Van Essen has studied the visual cortex and has been intrigued by a curious phenomenon. Two of the primary visual centers in the cortex, called V1 and V2, communicate via an enormous number of axons--long, stringy, signal-carrying fibers. In the developing embryo, V1 and V2 are relatively far apart. But by the sixth month of development, as axons from nerve cells in V1 connect to cells in V2, a fold appears between the two areas, bringing them close together.
That fold, Van Essen now thinks, is the natural pucker resulting from a great number of strings tugging at two regions on the cortical sheet. When I chatted with my colleagues about this, he says, I assumed that someone else had come up with such a fairly obvious notion and had explored it at some point in the last century. But as I examined the literature, it seemed no one had.
Many neuroscientists believe the brain is organized in such a way that the total amount of neuronal wiring is minimized. The brain’s many folds are created, Van Essen believes, by large numbers of developing axons trying to minimize the distance between different sites in the brain.
Areas with many axons between them, like V1 and V2, will be pulled together by tension, reducing the total yardage of nerve fibers. But areas not so intimately connected lose out in this tug-of-war; the axons that run between them are not numerous or strong enough to dictate the pattern of folding, so weakly connected areas may actually end up farther apart than they would if no folding occurred. These axons may be long, but they are few in number--so the sum total of axon lengths is still minimized.
Smaller-brained mammals, like rats and mice, have smooth cortices. Why doesn’t the tension in their brains produce folds? For one thing, they have relatively puny cortices. In a lowly tree shrew, says Van Essen, less than 15 percent of the cerebrum is cortex. Their cortex wraps tightly around the internal brain structures, like the skin of a balloon. Even though axons tug on the cortex, the tendency to buckle is countered by pressure from the internal structures. They push out against the cortex just as air pushes out against the wall of a balloon.
Van Essen is now developing a computer model to test his tension theory. A colleague has already shown that axons can generate the equivalent of 1 percent of that generated by mammalian skeletal muscle. That might seem negligible, but it’s probably enough to contort the brain, says Van Essen. Given the pliable nature of embryonic tissue, I suspect that minute forces would be sufficient.