This question is about terminology:

By which established and catchy terms are the sets of neurons with respect to a given neuron $N$ called which

  1. directly affect $N$ synaptically

  2. are directly affected by $N$ synaptically?

These terms came to my mind, but I am not sure which ones are most commonly used:

  • the presynaptic/postsynaptic neurons of $N$
  • the afferent/efferent neighbours of $N$
  • the afferents/efferents of $N$
  • the incoming/outgoing neurons of $N$
  • the neurons projecting on/to $N$ vs. the neurons projected on/to by $N$

Or is there no such established term, and anything goes, depending on the context?


At the simplest level, the terms 'pre-synaptic'/'post-synaptic' neurons should cover what you wish to describe. 'Afferent'/'efferent' are usually used to refer to axonal projections, usually between functionally distinct areas, or nerves. However, I wouldn't be too surprised if someone used these terms to refer to synaptically connected populations although I would consider it unusual. Simply, 'connected neurons' is also a term frequently used.

I will now go into a tangent to give you some background that should help you formulate an informed view about the vocabulary being used:

In areas that display a high degree of local connectivity, like the cerebral cortex and the hippocampus, any given neuron is connected with many different neurons. Hebb's theory of cell assemblies posits that a group of highly connected neurons may form over time in order to subsist a specific function. Given the high degree of connectivity, any given neuron will be able to participate in many different assemblies. It is often thought that these assemblies constitute the substrate of engrams, the physiological correlates of memory, at the level of the neuronal network. Hence, the concept of cell assemblies is prominent in our thinking about groups of synaptically connected neurons.

Setting aside the questions of the definition of the engram, recent efforts have claimed to identify engram cells under this or that experimental paradigm, although the connectivity of those cells has not been established. While we know that functionally similar neurons display a higher degree of connectivity and that groups of cells with correlated activity patterns can be formed by artifically-induced synchrony, the direct link between increased connectivity and behaviourally relevance has not been established.

On the other hand, only a handful of studies have directly measured the connectivity between more than two neurons (Song et al., Perin et al.), revealing the intricate properties of neuronal network structure. Depending on the level of description they may refer to distributed cell assemblies, synaptic clusters of neurons, $n$-vertex cliques, and motifs.

All this is to say that as soon as you start considering more than a pair of neurons, it makes little sense to talk about neurons that are exclusively connected with each other. As every neuron is connected to almost all other neurons in its immediate neighbourhood we often focus on the functional implications of their connectivity. Under this light, the concept of cell assemblies is central in modern neuroscience.

| improve this answer | |
  • $\begingroup$ I added "directly" to modify "affect" in the original question. $\endgroup$ – Hans-Peter Stricker Oct 5 '17 at 14:10
  • $\begingroup$ Sorry for the inconvenience. $\endgroup$ – Hans-Peter Stricker Oct 5 '17 at 16:22

I think the only way to reasonably answer this question is to say: "read some papers and follow their conventions" - it's much too hard to develop an exhaustive description of how all of those terms are used without creating an answer that's much too broad. I'll give some comments, though:

Presynaptic/postsynaptic is most often used in the context of synaptic transmission/plasticity/etc where the subject is a synapse. You would not typically say presynaptic neurons to N when you are really just talking about a network from A to N.

Afferent/efferent 'neighbors' I have never heard used, won't claim it never is but it seems completely wrong to me.

As for afferents/efferents these refer to the fibers primarily. Those terms are most appropriate when you are talking about something like the spinal cord, where afferent projections head toward the CNS/brain and efferent projections head toward the periphery; in the brain I find them often confusing or incorrectly used. This is because the terms afferent and efferent don't actually refer to inputs and outputs, but rather to inward versus outward direction. Defining "direction" in the brain gets pretty ambiguous if you aren't talking specifically about primary sensory or motor areas.

Incoming/outgoing I rarely if ever hear.

In the context of "projecting" usually the noun form is used: projections, as in "thalamocortical projections" or "projections from V1 to V2." I'd say this is the least ambiguous/best terminology if you are talking about connections between different brain areas. Within a local brain area (i.e., a given nucleus or a region of neocortex) you would be less likely to use the term "projection" though it wouldn't necessarily be wrong.

| improve this answer | |
  • $\begingroup$ But isn't the concept of "the set of neurons which directly affect neuron N synaptically" a very clear one, and would deserve a name? (And I did read more than just some papers, and didn't come up with a convention. Maybe because these sets aren't interesting and/or dealt with?) $\endgroup$ – Hans-Peter Stricker Oct 5 '17 at 17:04
  • $\begingroup$ It depends on context, but there's also no strict rules. If you are talking about just one cell in a computational fashion, you would probably just say "the inputs to cell X" but people would understand if you used some other terminology. I'd say this is typical of English in general, and maybe that gets carried over into scientific fields where English is the main lingua franca. $\endgroup$ – Bryan Krause Oct 5 '17 at 18:47
  • $\begingroup$ "Inputs to cell X" sounds good. Does "outputs from cell X" sounds good to you, too? $\endgroup$ – Hans-Peter Stricker Oct 6 '17 at 7:19
  • $\begingroup$ That's fine if you are talking about the outputs themselves; you could also use "targets of cell A" to describe the population that receives inputs from cell A. $\endgroup$ – Bryan Krause Oct 6 '17 at 15:06
  • $\begingroup$ You have to admit: terminology and wording is not very controlled and unequivocal in neurosciences. $\endgroup$ – Hans-Peter Stricker Nov 4 '17 at 14:31

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.