genetic control of inter-neuron signals
rsn_ at _comcast.net
Thu Apr 1 15:55:53 EST 2004
On 1 Apr 2004 11:32:27 -0800, QXUXBTVOTSAO at spammotel.com (Sandy
>Few people desire sex with a sibling. It seems well established that
>this is an evolved adaptation, one of a great many cases of behavior
>controlled by our genes. But given our current understanding of how
>brains work, it is difficult to understand the mechanism (at the
>molecular level) by which genes control behavior.
>When you see your sister and recognize her, then either some "sister
>sue" neuron is activated, or more likely, there is a spatio-temporal
>pattern of activation of certain neurons. But when these neurons
>were laid down during fetal development, they did not have particular
>meanings - it was learning that produced the association of these
>neurons with your sister. Thus from the neuron's point of view, it
>does not "know" that it is recognizing your sister. It only knows
>that it responds to the neurons that connect with it - it has no way
>of knowing what those other neurons are.
>How then, can a gene direct this neuron to behave in a way, that
>produces the result of not being aroused by one's siblings?
>Suppose the "sister sue" neuron sends out a dendrite, that forms a
>synapse with a "sex arousal" neuron. If a positive link is formed,
>then thinking about your sister will be arousing. This does not
>happen, but how does a gene prevent it? The "sister sue" neuron has
>no way of "knowing" that the neuron it has reached, is involved in
>There seems to be a disconnect between theories coming from
>evolutionary psychology, and theories coming from cognitive neurology.
> I think this disconnect is worth noting. I think both sciences
>could benefit from considering the problem.
> * * *
>I do also have a wild theory that, if true (and I don't think it
>likely) could explain how genes produce behavior. Suppose that there
>is some small protein that is exchanged at synapses. The synapse has
>a receptor for the protein, and the protein must be recognized to be
>absorbed (like any receptor). But suppose there is a segment of this
>protein (perhaps on the inside) that the receptor does not look at
>when recognizing the protein. The action of the receptor is merely to
>extract that segment, and release it (perhaps tagged in some way) into
>the cell. Thus neurons are able to communicate with each other by
>exchanging short protein "words." If the stretch of protein
>exchanged was no more than ten amino acids in length, many million
>distinct messages could be exchanged.
>Suppose one particular protein word has the meaning "sibling". Then
>the "sex arousal" neuron would have this word floating about, tagged
>as a message from across a synapse. The "sex arousal" neuron was not
>associated with sexual arousal when it was laid down in neural
>development, but it is connected with other sex cells, so a protein
>message can be passed. Thus the "sexual arousal" neuron has the
>protein words for sibling and for sex floating about. It is quite
>easy for a gene to be such that it is activated only when both these
>proteins are present, and the gene could cause newly-formed synapses
>to die, or make them inhibitory.
>* * *
>Well, that's my theory. I don't think it likely that I've hit on the
>correct mechanism by guesswork, but -
>1) It could be true. There are almost surely many neural
> peptides that have not yet been discovered.
>2) If not this exact mechanism, then by some other way,
> neurons could exchange messages that have a lot of
> content, more like words. It would be so easy to do
> that it is hard to see why it wouldn't happen.
>3) A great deal, that is hard to explain otherwise, and
> not just evolutionary psychology, could be explained if
> neurons can exchange high-content messages.
Leaving aside your ideas on incest taboos, your idea for neurons to
exchange complex information sounds interesting from a theoretical
perspective. Unfortunately it lacks support from the biological
mechanism end. It is conceivable that digital information can be
encoded into an amino acid sequence, just as it could into a
nucleotide sequence in DNA/RNA or into an action potential sequence.
However there is no experimental evidence to support such sequences
being used that way in any case that has been examined nor is there
any mechanism known that would allow such an interpretation.
First, protein signaling agents are not taken into the receiving cell,
but act by stimulating receptors on the membrane. These act by
activating a second messenger system which exist in relatively limited
number inside the cell.
Second, if the protein did enter the cell, there is the decoding
machinery to deal with. For now, the only known way a complex message
like you envision could be decoded is for it to bind to a specific DNA
sequence and so activate a specific gene. Although some lipid or lipid
soluble hormones do enter the cell and activate genes, and some neural
transmitters can influence gene expression through gene activation,
these all work using proteins that originate inside the cell, not be
proteins taken into the cell.
Third, nothing we now know about the genome or the processes of
protein/DNA interaction suggest that information could be transmitted
the way you suggest. Even if genes could be activated, there are only
so many ways a cell could react by changing its activity.
Cells do interact with each other in many ways beyond one cell simply
producing a synaptic potential on another and these produce complex
patterns of interaction using, as you suggest, proteins as
intermediary messengers. Still, coding information in the particular
sequence of amino acids is a little far fetched for what we now know
>From my perspective, the inability of real biological cells to do all
the incredible things we can think up using mechanisms that seem to be
relatively simple is good evidence for the absence of intelligent
design. Any really intelligent designer would have done all these
things! But cells don't seem to work that way.
More information about the Neur-sci