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

Neurons, DNA, Memory and Learning

  1. Jul 12, 2015 #1

    I am not an expert in this field but I am really hoping to understand as much as I can about the concepts described in my questions below. I might be using some improper jargon and expressions, so I apologize if some things are incorrect or confusing.

    #1 Do all the neurons have the same expression of the DNA that they contain? (DNA is the same in all of them, right?) Do the neurons cells get instinct behaviors directly from the DNA that they contain?

    #2 I suppose the behaviors that are learned (rather than instinctive) come from environmental stimulus and are expressed in combination with the pre-programmed genetic instinct. Also, I imagine that the learned behavior needs to happen a bunch of times through a given batch of nerve cells in order for the behavior to be remembered by the nerve cells' pathway that is being created. Is there a certain amount of times before this pathway becomes permanent? Does the pathway always stay more or less the same (since the same nerve cells would need to be stimulated, right?)? And is there a limit to how many different "stimuli"/"pieces of information" a give nerve cell can "transmit"/"hold" (I read somewhere it can be anywhere from 1000 to 10,000!)?

    #3 Once a nerve cell "adopts" a learned behavior, and it becomes "permanent", can it ever "forget"/"discharge" it? Or does the "charge"/"information" stay in that nerve cell for the rest of the person's life? Or does it slowly "fade" until it totally perhaps eventually dissipates?

    #4 I read somewhere that a new "batch" of nerve cells need to override an old, habituated, unwanted behavior, which actually never really "goes away", and merely stay "overridden", and can in fact return back if the new habituated behavior doesn't "hold-up". Is this true? And if it is, is this new "information" "stored" within the same batch of neurons, along the same pathway that held the "old" memory or is this a whole new batch of neurons, with a totally new pathway, with the "new" information somehow "overriding" the other neurons that hold/transmit the "old" behavior? (or both?)

    Many many thanks!
    Last edited: Jul 12, 2015
  2. jcsd
  3. Jul 12, 2015 #2
    Every cell in the body has the same dna.
  4. Jul 12, 2015 #3


    User Avatar
    Science Advisor
    2017 Award

    While all cells in the body contain largely the same DNA, if certain mutations happen during development (i.e. after the initial cell division when the embryo splits into the two cell stage), then these mutations will only occur in a subset of the cells in the body. Such mutations cause a phenomenon called mosaicism. Interestingly, mosaicism within the brain caused by the insertion of retrotransposon sequences within the DNA of neurons has been postulated as a http://www.nature.com/nrn/journal/v15/n8/full/nrn3730.html.

    As for the general questions about learned vs instinctual behaviors, here's an old PF thread that you might find interesting: https://www.physicsforums.com/threads/instinct-encoded-in-dna.602902/
    Last edited by a moderator: May 7, 2017
  5. Jul 13, 2015 #4
    Thanks for your response Ygggdrasil,
    I didn't realize that DNA can mutate within the body in its lifetime. And that it happens in neuron cells. Is this something that happens as a rule of thumb or more of a rare occurrence with neurons?
    Thanks for the link to the old PF thread, that thread is in fact how I stumbled upon this website when I googled about this topic. It didn't quite answer the questions that I had so I thought I'd post a new one here.
  6. Jul 13, 2015 #5
    All neurons contain the same DNA, but the "expression" of that DNA depends on what type of nerve cell is being produced. Obviously different genes are expressed depending on whether you're a mitral cell, amacrine cell, pyramidal cell, or a GABA-ergic inhibitiory interneuron.

    That's an oversimplistic concept of the way neurons and neuron networks work. The short answer is no.

    I think neurons in vitro can be artificially stimulated to up to 1000 pulses per second; there is a refractory period that places a limit there. However, these extreme frequencies are almost never seen in in vivo populations of neurons. The typical neuron in the mammalian brain pulses at less than 10 per second, it's the collective activity of 100's of millions of neurons all pulsing at about 3-4 spikes per second that drive brain dynamics and learning.

    Nothing is "permanent" in brain dynamics. And you're looking at it in the wrong way, something like an atomistic perspective of brain function. Learned behaviors are the the product of modified "Hebbian cell-assemblies;" they are not a product of the biology and genetics of single neurons. Rewarded behaviors strengthen the synaptic connections between neurons which facilitate the future expression of those behaviors, and unrewarded behaviors lead to the extinction of those behaviors through atrophy of those connections.



    The brain doesn't work like a computer, there are no separate sectors and hard drives to store separate information. Everything is stored, superimposed, on the same medium. Everything is constantly being written and overwritten, so practically, no, no memory is ever completely lost, but it can be so weakened that it is effectively lost, and given the right conditions an effectively lost memory can be regained or "reignited," so to speak.
  7. Jul 13, 2015 #6


    User Avatar
    Science Advisor

    Neurons might have different DNA, but if one wants a very solidly established case, one can look at VDJ recombination.
  8. Jul 13, 2015 #7


    User Avatar
    Science Advisor

    It depends on what you mean by permanent. There are almost certainly no organisms that are immortal, because of the accelerated expansion of the universe http://arxiv.org/abs/hep-th/0106109 :P

    More seriously, let's say you define some idea of "permanent". Then there are clearly behaviours or memories that can be learnt in one shot, while others need many repetitions.

    Here is an example of learning from one time: http://www.ncbi.nlm.nih.gov/pubmed/11572949 (free link to article at the top right).

    In a dish and in theory yes, so this probably happens, but off the top of my head, I don't know a solid demonstration that this is true for a behaviour in an animal. http://thebrain.mcgill.ca/flash/i/i_07/i_07_m/i_07_m_oub/i_07_m_oub.html

    Yes, this can happen, eg. in the experiment of Packard and McGaugh: http://www.ncbi.nlm.nih.gov/pubmed/8673408.

    They trained rats to do something. After 8 days the animal did it using method A. After 16 days the animals did it using method B. Was it because method B had erased method A, or was method A still there and method B was preferred? Packard and McGaugh silenced a part of the brain. The silencing made the rats revert to method A, indicating that method A was still stored in the brain.
    Last edited: Jul 13, 2015
  9. Jul 13, 2015 #8


    User Avatar
    Science Advisor
    2017 Award

    It seems like retrotransposon elements may be more active in the brain than in other tissues. Scientists are still working out the best techniques to detect these retrotransposition events, so there are widely varying estimates as to how common these events are in neurons.

    In any event, even cells with identical DNA can have different patterns of gene expression. These are though to be due to differences in the way the genes are packaged. Cells can mark specific regions of the genome with certain chemical tags on the DNA or on the proteins that package the DNA which can tell the cell to turn the genes on or turn the genes off. These epigenetic changes probably underlie most of the differences in gene expression between different individual neurons.

    Neurons can also have different levels of gene expression in different parts of the neuron. Neurons traffic mRNAs and ribosomes to their various axons and dendrites, and local neuronal activity at specific synaptic spines or terminals may change which genes are made at that synapse and only that synapse. So gene expression can not only vary from cell-to-cell, but in neurons at least, gene expression can also vary within different regions of the cell.
  10. Jul 13, 2015 #9
    You should look up the theory of allosteric modulators...which is absolutely fascinating in my opinion. Allosteric modulators, in theory, have a built in cap for the limit of the their activity, which is quite different than agonists or antagonists (the more you give, the more response you usually see). One idea that people have is that if you could develop allosteric modulators for targeting various things in the brain, you could give massive doses of a drug and have a built in safety mechanism for toxicity, but the real power of allosterism is the fact that you might be able to obtain a uniform response from all neurons. That's something you can not achieve with our current regimen of drugs that target the brain, and it could address how you treat brain diseases when each neuron might be behaving differently due the uniqueness of cell physiology profiles at any given moment in time (some neurons might die at a high dose of a drug that is needed to treat diseased neurons, but if you had an allosteric modulator, you could cap the max response and achieve a uniform treatment amongst all cells).
  11. Jul 13, 2015 #10
    Thanks so much for your response DiracPool, it clarified so many things!

    - And what about different types of information that they can process? Is there a limit to the variety of tasks it can process collectively?

    - I am guessing that atrophy here means that connections really never "die" but just get weakened or unused (until they might get stimulated again sometime in the future as you pointed out in an answer further in the thread)
  12. Jul 13, 2015 #11
    Thanks atyy!
    Great point about learning happening quick or slowly depending on the circumstances.
    Also, your http://thebrain.mcgill.ca/ link had another "procedural memory" reference to different types of learnings were defined, very helpful!
    And thanks for the Packard and McGaugh, exactly an answer I was looking for!
  13. Jul 13, 2015 #12


    User Avatar
    Science Advisor

    That's one of the most celebrated and important discoveries about memory - made by Brenda Milner's observations on H.M. The other major discovery she made with Scofield, the surgeon, is that the medial temporal lobe stores long-term memory, but not the super-long-term memory. See 10:05 for the discovery of procedural memory.

    What Brenda Milner looks like in real life.

    Another related phenomenon is http://www.ncbi.nlm.nih.gov/pubmed/12464700.

    I should mention this very interesting paper about the concepts involved in extinction as new learning versus unlearning: http://www.ncbi.nlm.nih.gov/pubmed/15466310.
  14. Jul 14, 2015 #13
    These are incredible resources! Answered so much! Thanks again atyy!
  15. Jul 14, 2015 #14
    I am still curious about the relationship between innate and learning behaviors and the influence of DNA on our behavior. I have found this article below and I especially like the image by the end that displays the percentual relationship between innate and learned behaviors in different animals. I was wondering if this is an accurate explanation of how the two different types of learnings develop.
    http://www.cals.ncsu.edu/course/ent425/tutorial/Behavior/ [Broken]
    Last edited by a moderator: May 7, 2017
  16. Jul 14, 2015 #15


    User Avatar
    Science Advisor

    I wouldn't take the percentages too seriously. Nonetheless, there are behaviours that are more "innate" and others that are more "learnt". But what is learning? There is no universal agreement, but a simple one is a change in behaviour as a result of experience.

    This definition is very broad, and if one thinks about it, it makes development and learning a continuum. http://bioteaching.com/phenotypic-plasticity-2/
    Last edited by a moderator: May 7, 2017
  17. Jul 14, 2015 #16
    Here are some of understandings that I've gained so far (I must admit I was slightly struggling with understanding the explanations of memory extension and spontaneous recovery paper(s)):

    - When a brain learns a fear response to a stimulus, the new stimulus that is not fear induced is a new memory. I get that. Does the experiment show that both behaviors can continue to exist independently, depending on wether they are stimulated or not? So at times when the stimulus is fear induced, the original memory will be reinforced, and similarly if the stimulus is not fear induced, the other "stored" memory will be induced?
    It seems like any memory will become habituated as long as there is a stimulus to reinforce it enough times, and once it is habituated, then the behavior can propagate, as long as it is not stopped by another behavior? And if the reinforcement stops (or some other behavior overrides it), the habituated behavior will slowly start to "fade".

    Furthermore, I understand that neurons (and groups of neurons) can have a certain amount pulses per second, but I am still unclear about whether a certain neural path can "hold" more than one learning? The way I get it so far is that learning can build on top of each other, using one learning to learn more. For example, if I meet someone who has a name similar to a famous actor, I will associate that person to all the movies that I saw with that actor. So new neural pathways are now built on top of (or in "collaboration" with) the previous pathways that held previous knowledge. I wonder if these same neural pathways can also be in charge of knowing other names, or some other types of information. Maybe certain cells can only take on only explicit or only implicit behaviors? And if so, would there be a limit if they can take on more behaviors?
  18. Jul 15, 2015 #17


    User Avatar
    Science Advisor

    Brain Cells for Grandmother
    R. Quian Quiroga, I. Fried, C. Koch
    Scientific American 308: 30-35, 2013

    The numb
    er of memories depends on how the neurons are connected, how the information is encoded and read out etc. Here is a relevant quote from the above article:

    "Using statistical methods, Stephen Waydo, at the time a doctoral student with one of us (Koch) at Caltech, estimated that a particular
    concept triggers the firing of no more than a million or so neurons, out of about a billion in the medial temporal lobe. But because we use pictures of things that are very familiar to the patients in our research—which tend to trigger more responses—this number should be taken strictly as an upper bound; the number of cells representing a concept may be 10 or 100 times as small, perhaps close to Lettvin’s guess of 18,000 neurons per concept."
  19. Jul 15, 2015 #18
    Fantastic article! And fantastic references to each of my questions atyy! Thanks again.
  20. Jul 28, 2015 #19
    Here is another question on the general topic above:
    Are there neurons in the brain that haven't been "fired up" yet and are waiting until a specific learning is to be in line? Or neurons always firing up, but not necessarily with any useful information to retain? How does this work?
    For example, when a baby is born, what "state" are all the neurons in? Are they are ALL active from the get go and start building neural pathways through the whole brain? Or is it more that certain parts of the brain are "dormant", if that's the right word.
    For example, considering that explicit memories don't start forming after a couple of years after birth, does that mean that, for example hippocampus is inactive during all that time? (I am assuming that hippocampus is only in charge of explicit memory).
  21. Jul 28, 2015 #20
    Yet another question I have is about the triune brain. (pardon my ignorance if I got lot of this wrong)
    Do reptiles have any form of a limbic system or is that only present in mammals? Similarly, do only primates have a neocortex, or do the rest of the mammals have a somewhat developed neocortex?
    Do fish, amphibians (and so on) have a reptilian brain too? Or some form of it? Do they have anything at all similar to a limbic system or neocortex?
    Now, If reptiles (or some other similar species group) only have a reptilian brain but no limbic system or neocortex, would that mean that they can't think and cannot feel? And does that mean that they (and all animals apart from mammals) don't think and don't have feelings and purely function based on survival, "reptilian" instinct?
    And how does this affect memory and learning in other species? Is implicit memory stored in the "lower", reptilian and limbic structures and is explicit memory present in only the neocortex? And what about learning by imitation? Which parts of the brain can learn through imitation, and as a result, are there certain species that can learn through imitation and some can't?
    Last edited: Jul 28, 2015
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Threads - Neurons Memory Learning Date
Replacing a neuron by an artificial connection Jan 15, 2018
Medical Peripheral nerve anatomy Jan 3, 2018
Neuron lengths, sizes and quantity Nov 23, 2017
How realistic are modern neuron models? Oct 17, 2017
Neurons and Memory Jan 7, 2004