How do neural networks form?

[gib65]

I've always assumed neurons form networks with each other by reaching out with their dentrites and axon terminals and synapsing onto other neurons, and that these neural networks could change their connections in a similar way - that is, by one set of dendrites or axon terminals disconnecting from one location and moving over to a new location.

 

However, I've read a different method - I've read that neurons change their connections by increasing or decreasing the number of receptors in the synaptic gap. So if a connection was no longer needed, instead of removing the dentrite or axon terminal, the receptors were depleted.

 

Which one of these scenarios is more accurate - or do both processes go on?

[scicop]

I've always assumed neurons form networks with each other by reaching out with their dentrites and axon terminals and synapsing onto other neurons' date=' and that these neural networks could change their connections in a similar way - that is, by one set of dendrites or axon terminals disconnecting from one location and moving over to a new location.

 

However, I've read a different method - I've read that neurons change their connections by increasing or decreasing the number of receptors in the synaptic gap. So if a connection was no longer needed, instead of removing the dentrite or axon terminal, the receptors were depleted.

 

Which one of these scenarios is more accurate - or do both processes go on?[/quote']

 

Well you're on the right track, it depends on which stages you're looking at.

 

During development, both axons and dendrites traverse the environment and are guided by molecular cues in the enviroment (some of these cues attractants, some are repelants). The fidelity of neuron these connections work in such a with that a proper synapse is NOT formed if there is a target mismatch (based on molecular cues expresed by respective dendrite and axons). They may do a little dance with each other, but the connection will not be secured, and they will move on until the appropriate target is met. Once a connection is made or rather forged [this is both molecular cue dependant and activity (electric coupling) dependant] it persists for quite a while until the final stages of connection occur.

 

The connections are maintained through basal levels of activity. Once connections are made, neurons regulate their "plasticity" (or sensitivity to stimuli), by altering the number of "receptors" at thier synatpic/perisynatpic membrane. The more "receptors" inserted into the membrane (either through incorporation of pre-fabricated receptors, or de novo translation at the synapse) the more sensitive/stronger the connection becomes, then the ye truth of "neurons that fire together, wire together" is set forth. If a synapse is really strong, then most likely another synapse will form a nearby though a budding process on the dedritic side and a bifurcating process on the axonal side.

 

However, if a synapse becomes less used, there may be a reduction of receptors at the synaptic cleft, however basal activity will maintain the connection. Rare is a synapse really "silent". There is alot of energy expended to form a synapse, so the neurons want to maintain their connections. If it was really strong synpase, it will persist for a long time (years/decades) and this may underlie memory retention. However for the cases of really weak or damaged connections, activity will not be maintained, receptors will be downregulated and both dendrites and axons will degenerate (atrophy) They will not go searching for another synapse.

 

Hope this helps. Let me know if you need in more details, I would definately invest in the Principles of Neuroscience text book by Kandel, Jessell, and Schwartz, its like the bible for neurosci..you'll find alot of this stuff in there.

 

Also look up work by the now retired Corey Goodman (alot of pioneering neuronal connectivity work was done in drosophila at UC Berkley), Marc Tessier-Lavine (at Genentech now..after a sucessfull career at UCSF), Steven Burden (NYU), and Moo Ming Poo (UC Berkley), really awesome work showing the the level of cAMP within the growth cone can control attraction or repulsion to guidance molecules.