brain science beyond predictive coding | reaching old memories | infantile amnesia | llms
How is it possible not to remember something for so many years, then after seeing something, in a kind of way, it suddenly brings that thing to mind?
How does the mind decide at that point that it is the turn of that old memory?
Memories are theorized to be functions of the mind. Their distributions are theorized to be qualifiers. Simply, memories, feelings, emotions, modulations — as functions — have numerous qualifiers, making determinations for where they go, when, how, why and so on.
This distribution qualification for memories makes it possible to theorize that infantile or childhood amnesia, where things between the ages of 0–2, are almost never remembered as adults, may also be because the distribution qualification that those memories had could not be accessed.
This means that hypothetically, if adults were to briefly revert to being babies, those memories would be accessed because the distributions would carry out reaches, to some experiences of those years.
How does the mind reach old memories, or how does distribution happen in the human mind?
It is postulated that in a cluster of neurons, impulses form a set or sets. It is within these sets that they have formations or configurations to carry out functions. Simply, all functions of the mind are mechanized by configurations of electrical and chemical impulses in sets.
Formations are made by rations of respective chemical impulses. This means that different chemical impulses contribute, for functions. This is why most neurotransmitters are implicated across functions, not just a few, since they contribute to rations, within sets, including those that send in fibers from elsewhere.
All configurations of impulses are never to full capacity for the ration of their impulses. This means that the total volumes that the formation can have are hardly filled, leaving spaces where the functions can be qualified.
Some qualifications for functions are attention or prioritization, awareness or pre-prioritization, intent or free will, self or subjective experience, distribution and so on.
Distribution [reach or shares] is what might make a memory result in another, or a memory to an emotion, and so forth. It is what also makes old memories come around. Mental disorders disrupt distributions as well as mind altering substances.
Since one neuron has thousands of synapses with other neurons, it is theorized that there are active and non-active synapses in sets of neurons. It is some of these non-active synapses, between active synapses that determine some qualifications, including distribution.
Often, formations are made as thin as possible, for unique features, so that whatever is common between one or more, are provided in some thick sets.
For instance, a unique door and common features — of kinds — of doors. The thick sets may not be in the same locations, but distributions appear to make it appear they are, or might be.
Thin sets also take in minimums for the function they are made of, so that in distributing, there are parts that can be amplified. Their thinness allows them to be qualified better, which may allow for parts to set off for unpredictable reaches, at times.
This means that while it is common to see doors, windows, vehicles, highways, poles and so forth, it is possible to see one, against a certain background that would result in bringing back an old memory.
It can also be theorized that another reason for infantile amnesia may be that many of the formations are thick instead of thin, so by the time they become thin, with age, linking back is almost impossible, with reach, even if elements remain.
Also, for those that were thin, the qualifier for distribution may have overlapped with others, making it more difficult to have some of the memories of infanthood get distributed, as they were or would have.
How does the brain do what seems like predictive coding and predictive processing?
Predictive coding theory says that the brain generates predictions of the external world by using models of the internal. The problem is that how exactly the brain or mind does has not been made clear, neither is the way the brain actually takes up the internal model.
The observation that appears as prediction may be accurate, but the brain is not predicting. For instance, in remembering something that has been forgotten by seeing something else, the mind first interprets, with distributions and then distributes further, especially if it has to evoke something.
In brain science, it is established that electrical impulses leap from node to node, over myelinated axons, in what is called saltatory conduction.
It is theorized that in a set of impulses, in a cluster of neurons, some electrical impulses split from an incoming bundle to go ahead of others, to interact with chemical impulses like before, within that set or elsewhere.
It is this go-before that is observed as prediction.
The go-before option is sometimes just one thing, not more. It is often linked, not just likelihood, like in llms.
If the input matches with the go-before, the rest of the incoming follow those that went ahead. If not, the incoming one goes in another direction. This is termed prediction error correction, along with predictive processing.
For instance, hearing a sound, getting that it might be something [which is the early split], then listening closely, and finding that it is not, means that the rest of the incoming went in another direction. Assuming it matched, other distributions would have continued.
The mind is not predicting like llms, which cannot seem to correct their error when they hallucinate or confabulate, showing that brain science has to leap from the predictive coding hold.
Visit us at DataDrivenInvestor.com
Subscribe to DDIntel here.
Have a unique story to share? Submit to DDIntel here.
Join our creator ecosystem here.
DDIntel captures the more notable pieces from our main site and our popular DDI Medium publication. Check us out for more insightful work from our community.
DDI Official Telegram Channel: https://t.me/+tafUp6ecEys4YjQ1
