If a single neuron can have thousands of synapses with other neurons, how do each neuron "knows" to which further neuron to send transmitters? i.e. selectivity


When a neuron fires an action potential, it doesn't "select" a target. It releases neurotransmitter (somewhat probabilistically, as vesicle release has a probability <1 at most synapses) to every cell it synapses on, all of the thousands you reference.

What is important for selectivity is the development of synapses, which are shaped by axon guidance cues and later by synaptic plasticity. All of this is laid down well before an action potential causes neurotransmitter release.

However, for almost all connections in the brain, a post-synaptic neuron won't fire due to the activity of a single pre-synaptic cell. Therefore, the selectivity in which neurons will fire an action potential is driven by the population of neurons that are previously active, not any one cell. Neurons that fire in the near future can be said to be determined by the combination of all the neurons firing in the recent past, including both excitatory and inhibitory neurons. This gives brains an incredible diversity of possible states. If we are to simplify neurons as "firing" or "not firing" in a given time window that would equal 2^(number of neurons) possible states, each leading to a different subsequent state.

  • $\begingroup$ Thanks for the detailed response! I didn't quite understand the second part about the synapses. I understand that they get stronger/weaker and that there are cues that direct neurons to their target neurons, but how is that related to selectivity of action potential if we say that they always release neurotransmitters to all their synapses? Also, can you explain the part about "neurons that fire in the future are determined by past fired neurons"? How is the fact that some sequence of neurons fired previously determine which future synapses will be activated? $\endgroup$ – Penguin Dec 18 '20 at 17:09
  • $\begingroup$ @Penguin 1) A neuron's selectively is solely by which cells it is connected to. Neurons don't connect to every other neuron, so every time they fire they are selective in which cells they impinge on even though it is the same cells with every firing of that specific cell. 2) The set of neurons that fired in the recent past determines which neurons fire in the near future. Those neurons that fire will release transmitter at their synapses; the neurons that do not fire will not release neurotransmitter. Therefore, only some synapses will be activated (the ones from the cells that fire). $\endgroup$ – Bryan Krause Dec 18 '20 at 17:13
  • $\begingroup$ Oh I see!! Okay I think I almost got it. Is the fact that some sequence of neurons ("N") fired in the past and released transmitters relate to the action potential? Like, that sequence "N" fired and hence out of a group of possible neurons (e.g A,B,C) only A is activated since only its action potential was reached? Possibly because "N" is also connected to A? Or is it something completely different? $\endgroup$ – Penguin Dec 18 '20 at 17:27
  • $\begingroup$ @Penguin Yes I think you've got it; we call that "threshold": cells only fire an action potential if their "threshold" is reached. A simple equation expressed in programming logic might be "if (ExcitationToA - InhibitionToA > ThresholdOfA) then (FireActionPotentialA)". FireActionPotentialA would mean that all neurons that A synapses on receive the input from A in the next "iteration" (of course, biology doesn't do iterations, everything is continuous, but it helps as a thought experiment). $\endgroup$ – Bryan Krause Dec 18 '20 at 18:06
  • $\begingroup$ Yay! Thank you so much!!! I'm not sure if I should start a new question or if this is related, but what would make a new neuron connect to an existing group? Is it the guidance cues you mentioned? Like, I understand that if neuron A is activated very often with neuron B at some point it's very likely that they will have a connection (Hebbian theory). But how would a new neuron that we never had before (new knowledge) be suddenly activated? I assume that if that new neuron is activated with existing neuron A then they'll be connected, but I'm not sure about how that new neuron will be activated $\endgroup$ – Penguin Dec 18 '20 at 20:03

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.