Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Neural Connections

  1. Jun 6, 2005 #1
    Research has shown that memory and learning is based on the values of weights that the neural synapses have. Learning involves the adjustment of the weights (strengths) of these synapses so that the reactions given to stimuli are adjusted as well.

    But are all the correct values learned during one's life, or are they also inherited? How are the weights initialized when a baby is born?
  2. jcsd
  3. Jun 6, 2005 #2
    I am not an expert in neural connections, but I know that a spider has certain neural connections it has had genetically "wired" into its brain. A spider web is a very complicated structure, some strands are much thicker than others etc. But anyway it seems that the answer to your question is that everyone starts off with a similar but not identical neural connections. Inheritance of "intelligence" is obvious, we as humans pass on this ability to reason. However the approaches we take to reasoning can be learned and are obiviously very different then something essential like making a spider web, and so everyone starts off with almost the same chance.
    Last edited: Jun 6, 2005
  4. Jun 6, 2005 #3
    "But are all the correct values learned during one's life, or are they also inherited? How are the weights initialized when a baby is born?" values?

    synapses are all about inhibition/excitation/"all or none"firing ...the weight values that you speak of is a math model refereing to [1]number of connections between two neurons...which can be many [2] amount of myelination. If a neuron is myelinated(white coating) then conductance is faster down an axon.

    AT birth it is said that an infants brain is FC-fully connected and unmyelinated
    and during the first 5 years of life pruning/cell death occurs to excise unused or weak synapses and myelination for those of strong firing. Pruning stops at the age of about 5 and myelination i don't know if it stops but i read that its said to be about 15 to hone ones skills...but adults can learn so i don't know if thats do to the network setup but somethings gotta change for them to learn. Also note that "terrible twos" is names as so because of the constant change at a young age and its really cool that if you damgage your visual system or audition at a young age that the other area can grow or adapt to handle some processes of the damage neural substrates.

    "But are all the correct values learned during one's life or inherited" the majority of them are learned or experienced...I can't quote on inheritance because its an ongoing debate...there may be some neural structures that are geneticallly wired but for the most its grown. Food is argued to influence the brain.

    "How are the weights initialized when a baby is born?" should be requestioned to say how are they reconnected...and well this all starts when the brain is not even a brain.
    But nutrition probably has alot to do with it. Genetic Code also prolly says hook up these neurons to these for the time being.
  5. Jun 6, 2005 #4


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    What you're referring to (I think) is one model of how synaptic plasticity occurs (it's not the only one, and right now, it's still just a hypothetical model, not a known fact). I haven't looked into this in depth, though we've somewhat recently hired on someone who works in this field, so I've attended his seminars as well as those of other candidates working in that area. Anyway, my understanding is that neurons are spontaneously (? - there'd be some argument if it's spontaneous or directed) producing outgrowths (spines) all the time, but the right signals (again, vague here...complementary transmitters and receptors or surface adhesion molecules) have to be present for that outgrowth to form a bona fide synapse. If there is strong signalling at that synapse, it is reinforced and more likely to become long-lasting, but if it is only weak, or weaker than other nearby synapses, the neuronal outgrowth retracts.

    Of course, during development, some of this has to be genetically determined. For example, neural adhesion molecules provide paths for neurons to follow as they migrate into their adult locations. There are developmental stages during which some molecules are turned on or off that help guide different stages of neuronal development.

    Much of the "weighting" your talking about, as I've heard it, refers more to after the neurons are in position and there is juxtapositioning of axons and dendrites. Once the major structural components are in place, the neurite outgrowth is where most adult synaptic plasticity takes place.

    While my area of research is not in learning/memory, I am interested in neuronal remodeling/synaptic plasticity in a different context. So, I'll likely return to this discussion later with a little better explanations. I'm just posting quickly while gulping down coffee before heading back into the lab, hence the rather vague, incomplete, and not fully thought out explanation above. (Okay, I'm just posting so you'll stick around to continue the discussion later. :biggrin:)
  6. Jun 6, 2005 #5
    ah yes being a modeller i tend to forget the drugs/neurotransmitters influence on teh brain.
  7. Jun 8, 2005 #6


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    :uhh: What exactly are you modeling then? :biggrin:
  8. Jun 9, 2005 #7
    mmmmm the network connections =] the software not the hardware i'm a strong AI supporter
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook