The adult brain is hardwired and cannot generate new neurons.

For most of the twentieth century, a single sentence organized thinking about the adult brain: the neurons you were born with were the neurons you would die with. New cells might grow in the liver, the skin, the gut, but the central nervous system, once its development was complete, was fixed. Injury was permanent. The brain that entered adulthood was the brain that would age in place, its circuits laid down and unchangeable. Santiago Ramón y Cajal, the Spanish histologist whose meticulous drawings of neural anatomy earned him a Nobel Prize in 1906, put it most bluntly: "In the adult centers the nerve paths are something fixed, ended, immutable. Everything may die, nothing may be regenerated."

Ramón y Cajal's authority was enormous, and his conclusion was not unreasonable given the technology of his era. The neuron doctrine, the principle that the nervous system is composed of discrete cells, was itself a major advance. Detecting whether individual neurons were being born in living adult tissue required tools that did not exist in 1906 and barely existed in 1960. The dogma that the adult brain could not generate new neurons thus rested not on disproof but on an absence of evidence, bolstered by the authority of a founding figure and the theoretical coherence of the resulting picture.

The first serious cracks came in the 1960s, when Joseph Altman, a researcher at MIT, injected radioactively labeled thymidine, a DNA precursor that incorporates into newly divided cells, into adult rats and found evidence of new neuron formation in the hippocampus, olfactory bulb, and cerebral cortex. His findings, published in Science in 1962 and 1965, were largely dismissed or ignored. The consensus was too settled to be overturned by results that most neurologists assumed must reflect a methodological error.

The rehabilitation of Altman's insight came in stages. Michael Kaplan extended the thymidine labeling work in the 1970s and 1980s, using electron microscopy to demonstrate that the newly divided cells in the adult rat hippocampus had the structural characteristics of neurons. Fernando Nottebohm, working at Rockefeller University, showed in 1983 that the brains of canaries grew new neurons each spring as the birds learned new songs, the first unambiguous demonstration in vertebrates that adult neurogenesis was tied to learning and memory. By the mid-1990s the evidence from rodents, birds, and nonhuman primates had accumulated to the point where the dogma was untenable.

The definitive demonstration in humans came in 1998. Peter Eriksson, working at the Sahlgrenska University Hospital in Göteborg, Sweden, together with Fred Gage of the Salk Institute, used an ingenious opportunity: cancer patients who had received infusions of bromodeoxyuridine (BrdU), a DNA synthesis marker used clinically to track tumor growth, had donated their brains to science after death. BrdU incorporates permanently into the DNA of any dividing cell. In postmortem tissue from the hippocampal dentate gyrus, Eriksson and colleagues found BrdU-labeled cells that also expressed neuronal markers, cells that had divided during adulthood and differentiated into neurons. Their paper, published in Nature Medicine in November 1998, concluded: "The human hippocampus retains its ability to generate neurons throughout life." The sentence demolished a century of received wisdom in twelve words.