There are plenty of sources on the number of synapses the average neuron in some region of the brain has. However, its become clear that there is some degree of redundancy in these connections, where a pair of neurons have multiple synapses between them. I'm aware that the multiple synapses aren't truly redundant, they serve a purpose. I just want to know if there are any estimates on how many other neurons a single neuron receives from on average.
3
-
$\begingroup$ Off the top of my head here is an example estimate of the reverse question (how many neurons does a neuron target) to get you started: physoc.onlinelibrary.wiley.com/doi/full/10.1113/…. Most pairs of neurons that are connected are expected to have one synapse with each other, although cases of multiple contacts are easier to detect: ncbi.nlm.nih.gov/pmc/articles/PMC1159394. However, the cortex, especially L2/3, is build to maximise potential connectivity, so the real question is how many other neurons can a neuron connect to with a synapse. $\endgroup$– vkehayasCommented Oct 12, 2022 at 9:45
-
2$\begingroup$ @vkehayas "Most pairs of neurons that are connected are expected to have one synapse with each other" is different from my understanding for neocortex, and conflicts with the paper you link, as well: "The number of potential synaptic contacts formed by a presynaptic neurone on a target neurone varied between four and eight" $\endgroup$– Bryan Krause ♦Commented Oct 12, 2022 at 14:25
-
$\begingroup$ Yeah, I know! But there are known limitations of the method. Weak connections are harder to detect, even if located on the soma. It becomes more challenging because most connections are in distant dendrites, due to a phenomenon called "space clamp". For a more unbiased measure, see Table 1 of Lee et al. Large-scale models (e.g. Braitenberg & Schuz) have predicted this fact based on basic features of the cortical architecture. It can still be, though, that there is some space left for specificity in connectivity (Lee et al., Fig. 2). $\endgroup$– vkehayasCommented Oct 13, 2022 at 8:30
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
2
For biological neural network in the brain unlike an artificial one, neurons are not strictly layered and one particular neuron may be connected to a large amount of other neurons which is obviously hard to correctly measure even for one brain, not to mention for a representative sample. For some rough idea this 2019 arXiv submitted paper estimates each of the neurons has 10,000 connections with other neurons.
On average, the human brain contains about 100 billion neurons and many more neuroglia which serve to support and protect the neurons. Each neuron may be connected to up to 10,000 other neurons, passing signals to each other via as many as 1,000 trillion synapses.
This 10,000 number is also confirmed by the academic webpage titled "Introduction to Neurons and Neuronal Networks" authored by John H. Byrne, Ph.D., Department of Neurobiology and Anatomy, McGovern Medical School.
it has been estimated that one neuron can receive contacts from up to 10,000 other cells. Consequently, the potential complexity of the networks is vast. Similarly, any one neuron can contact up to 10,000 postsynaptic cells.
-
$\begingroup$ 1,000 trillion synapses, and 10,000 other neurons would be 100 billion synapses between each neuron on average. That seems excessive, does that number of synapses include ones that don't connect directly to other cells? $\endgroup$ Commented Feb 17, 2023 at 16:11
-
$\begingroup$ The reliable and important part is "Each neuron may be connected to up to 10,000 other neurons" addressing your titular question, as for the distribution of number of synapses of a connection pair of neurons as reflected in above comments they're different in various brain regions and also is not simply a multiplication, should be combinatoric. $\endgroup$– cinchCommented Feb 17, 2023 at 21:08