I have been reading that, with exceptions for a few parts of the human brain, no neurogenesis occurs, so no new neurons are created. I have also heard that no new axons are formed through the white matter of the brain after birth as well. I also know that the axons are myelinated, minimizing crosstalk and other interesting structural effects that might occur.

Between those, does that imply that the structure of the white-matter of the brain is fixed at a very young age, and never changes through our life? It seems like a natural conclusion from the data, but it feels counterintuitive when contrasted with the continuous creation and destruction of synapses which happens so close nearby.


No, white matter structure is not fixed at a very early age. Although a person is born with most of the neurons they will ever have, those neurons must form connections between each other, via axons and synapses, and many of the axons become myelinated during this period. Although the number of neurons remains relatively constant, these processes lead to a substantial increase in overall brain volume. After a few years, there is then a period of synaptic pruning (https://en.wikipedia.org/wiki/Synaptic_pruning), in which many of those connections are rationalised. These processes explain many of the characteristics of human development.

For example, mammalian species can be characterised as precocial or altricial. A precocial species is one in which much of the brain's development has been completed in utero. This allows for a relatively sophisticated behavioural repertoire to be immediately available (think precocial = 'precocious'). For example, a foal is born and almost immediately is able to walk, follow its mother, and so on. Meanwhile, an altricial species like a human is born in a relatively helpless state, incapable of independent locomotion and with even very limited perceptual processes. It is the development of axonal and white matter pathways, with associated synapses, that allow learning and development of new skills in the post-natal period.

During this process, the brain is extremely flexible. For example, a baby can initially distinguish between the phonemes of any language. As its white matter connections develop in response to a lot of exposure to the language(s) in its home environment, the brain becomes more specialised. Eventually, past the pre-teen period, it becomes much harder to perceive and articulate subtle phoneme distinctions from other languages, and the acquisition of new grammars shifts from effortless and rapid, to slow and requiring formalised instruction rather than being spontaneous.

This shift to relative inflexibility reflects the second phase of white matter development (i.e. large-scale synaptic pruning). That is, in the first phase, there is widespread growth of connections between neurons via axons and synapses, along with myelination. In the second phase, the flexibility of this system is traded for the efficiency and focus of a brain that has learned what specific abilities it needs to have. This continues up until about age 10, at which stage 50% of the synapses present at age 2 have been lost (https://link.springer.com/referenceworkentry/10.1007%2F978-0-387-79061-9_2856). Learning is then no longer so rapid, but with specialisation and development, an adult is able to focus attention explicitly and maintain discipline towards achieving goals, where a child might get more easily distracted and seek novelty.

Synaptic formation and loss then continues throughout life, as the underpinning of learning processes, but no longer on the mass structural scale seen in earlier development.

  • $\begingroup$ Is it only the synapses that are changing, or are the long axions also growing or changing, such that they might terminate somewhere else in the brain? $\endgroup$ – Cort Ammon Oct 12 '18 at 20:22
  • $\begingroup$ It is probably the competitive selection and destruction of axons that is the dominant process. eg in the adult macaque there are 50 million axons in the corpus callosum, selected from the 200 million available at birth. Check out chapter 6 of this textbook by Ackerman for a good introduction: ncbi.nlm.nih.gov/books/NBK234146/?report=reader#ddd00050 $\endgroup$ – Michael MacAskill Oct 13 '18 at 0:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.