Monday, April 2, 2018

Why the Instantaneous Recall of Old Memories Should Be Impossible for a Brain

One of the most powerful arguments against the claim that minds are made by brains is what I may call the navigation argument. This argument can be simply stated like this: a long-term memory cannot be stored in some particular part of the brain, because there could be no way in which your brain could ever instantly find the exact location where such a memory was stored.

Let's consider a simple case. You hear the name of a movie star. You then instantly recall what that person looks like, and see a faint image of that person in your “mind's eye.” But how could this ever happen, if the memory of that person is stored in some particular part of your brain? In such a case, you would need to know or find the exact place in the brain where that memory was stored. But there would be no way for your brain to do such a thing. It would be like trying to find one particular needle in a skyscraper-sized stack of needles.

Let's try to imagine some ways of getting around this difficulty, and consider whether they are viable. One possibility is that a memory hypothetically created in the brain might have some type of unique positional identifier, something rather than a GPS coordinate or a longitude/latitude coordinate (although it might be something like a 3D coordinate). When a visual memory was created, the brain might associate this coordinate with some memory cue (such as the person's name). So, for example, when you hear the phrase “White House” maybe your brain subconsciously retrieves some 3D coordinate allowing you to find where you have stored in your memory an image of the White House in Washington, D.C. The visual below illustrates the idea. 

But there are three reasons why this doesn't work as an explanation. The first is that there is no way that the brain could ever know what the 3D coordinate was for some particular spot where a new memory was created. For example, if a brain is creating a new memory at a location with an X coordinate of 23.23342, a Y coordinate of 45.34245, and a Z coordinate of 33.3293, the brain would have no way of knowing that the memory was being created at that exact location. A second reason is that even if there was some brain storage area storing the location coordinates for particular memories, there would still be the question of: how could the brain instantly find the correct spot in such a cue storage area to find where these coordinates were? Since your brain can store millions of different memories, we must imagine that this cue storage area would also have millions of items. There is no reason we can think of why you would be instantly be able to find the exact part of this cue storage area that would have the coordinate needed to locate the appropriate memory. Third, there is the difficulty that even if the brain had the physical coordinates at which the memory stored, it would have no way of navigating to such a coordinate.

To try to make things a little better, let's get rid of this idea of a coordinate system. Let's imagine that there is some kind of cue memory area with direct neural connections leading to the spots where memories are formed. So then if you hear a particular name, your brain merely finds that name in the cue storage area, and then follows this little neural connection (rather like a wire or telephone line) to where the memory is stored. The visual below illustrates the idea.

But this still does not give us a plausible answer as to how you could recall a memory stored in a particular spot in your brain. For one thing, there would be the difficulty of explaining how this wiring-up was occurring. It doesn't seem plausible to maintain that each time you memorize something, you are adding an entry in two different storage areas of your brain, and also instantly creating a neural wire or line connecting only these two. That seems like too much work and coordination to be occurring so quickly. We have no evidence that coordinated changes occur in two different areas of the brain when a memory is stored. While we know that connections can gradually form between neurons, this isn't something that can instantaneously occur to link separate areas of the brain when a memory is created.

You would also still have the previously mentioned problem of how your brain could instantly find the correct spot in this cue storage area where these cues are located. Since your brain can store millions of different memories, we must imagine that this cue storage area would also have millions of items. There is no reason we can think of why you would be instantly able to find the exact part of this cue storage area that would have the direct wire or neural connection needed to locate the appropriate memory in the visual storage area.

So we seem to be getting nowhere trying to imagine how the brain could allow you to instantly recall memories. Let's try looking at how computers are able to quickly retrieve data. Perhaps that might offer some clue.

One basic technique computers use to speed retrieval access is physical sorting. The same technique is used by a simple file cabinet in which the files are alphabetically sorted. But we cannot believe that the brain uses physical sorting. A mass of microscopic neurons (rather like a city-sized blob of tangled spaghetti that has been shrunk) is not something in which physical sorting is possible. All the different brain connections make physical sorting impossible, just like you can't sort a giant ball of tangled string into different parts (unless you disassemble and reassemble). A brain structured totally different from the human brain might be able to use physical sorting, but not the human brain. We have zero evidence that neurons are physically sorted.

Another basic technique computers use to speed retrieval access is physical grouping, such as when different types of information are placed in different computer files. The same technique is used by a simple file cabinet in which different types of information are put into different manila files. But we cannot believe that the brain uses physical grouping. A mass of neurons filled with connections between neurons is not something in which physical grouping is possible. A brain structured totally different from the human brain might be able to use physical grouping, but not the human brain. When we look at the brain, we see no evidence that a physical grouping of neurons is occurring. Neurons are not arranged into little clusters like stars are arranged into galaxies.  Neurons exist in incredibly tangled arrangements, and when everything is so tangled, physical grouping is not possible.

In short, it seems physically impossible that a brain structured such as ours could ever instantly retrieve memories, if such memories were stored in particular parts of the brain. There is no physical mechanism that can explain how this could occur in the human brain.

memory retrieval

One way to try to resist such an argument is to claim that a particular memory is not stored in a particular place in the brain, but throughout the brain. If that were true, there would be no problem of the brain trying to find just one particular spot where a memory is stored. But a human mind has many thousands or millions of memories. We can hardly believe that each of these is stored throughout the brain. If my memory of my first kiss was stored throughout my brain, that would leave no space for the thousands or millions of other memories I have. We may also note how absurd it is to imagine that, for example, that I have stored throughout all the neurons of my brain some trivial image such as the image of SpongeBob Squarepants.

It is sometimes maintained that a specific memory is stored not in one little spot in the brain, or throughout the entire brain, but is scattered across several different places. This does nothing to make the retrieval of a memory from a brain more plausible. You do not lessen the difficulty of “how could the brain know where to find one specific spot where a particular memory is stored” if we assume that the memory is read from five or ten different specific places, because there is still the problem of how the brain could find those specific locations. Similarly, there would be no way for someone to find some information very quickly if it is in some book among many thousands of books in a library that all had no title on their cover; and it would be even harder to get such information quickly if the information was scattered across five or ten such books in such a library. Lacking any coordinate system, a brain storing memories would be like such a library.

In a blog post, a neurologist has replied to the type of argument being made here (after it had been made by another writer). The neurologist writes: “You don’t have to know (nor does your brain) where in your physical brain a memory is located, because you can access that memory simply because it is integrated with so many other memories.”

This reasoning is fallacious. Using similar reasoning, I could argue that I don't have to know the phone number of someone in California to telephone that person, because all of the telephones are integrated with each other in a telephone network. But that's erroneous. You do have to know someone's phone number to instantly access that person's phone. To instantly access a memory such as occurs when you recall a face after seeing a name (or vice versa), the brain would absolutely need to know where that memory is stored (or have some mechanism for instantly locating that precise brain location), if the memory was stored in one particular location in the brain. But no one can give any explanation of how the brain could know such a thing, or how the brain could instantly find the correct memory. I may note that when you recall a face instantly after seeing it, it is not at all a case of one memory leading you to find another. We absolutely cannot use “integration of memories” to explain such a thing.

If you say the name of a famous person such as John Kennedy, I do not have some mental experience of searching through memory like someone flipping the pages of an encyclopedia. I do not see in my mind's eye many images flashing until I finally reach the image of John Kennedy. Instead, you suddenly say "John Kennedy," and I instantly retrieve nothing at all except the image of John Kennedy. How could that work so quickly, if the image is stored in, say, brain location #834,342,430, and I have no way of remembering that the image is stored in that location, in a brain that has no coordinate or addressing system allowing exact brain addresses?

The complete lack of any workable theory for how memory recall can occur so quickly is admitted by neuroscientist David Eagleman, who states:

Memory retrieval is even more mysterious than storage. When I ask if you know Alex Ritchie, the answer is immediately obvious to you, and there is no good theory to explain how memory retrieval can happen so quickly.

Another difficulty is that there is nothing in the brain that seems to have the job of reading. Consider the question: how does the reading of information occur when someone reads a book or when a computer reads a file? In the case of someone reading a book, we know how it works. A person's eyes can focus on a particular spot on a page. So when you read a book, your hands open to a particular page, and then your vision focuses on one particular line. Similarly, we know how reading occurs when your computer reads a file on your hard disk. The disk is a circular spinning surface. Something called a read-write head can move to any spot on the disk, to read from that location. But in the brain there seems to be no comparable item that might allow the brain to focus on one particular spot where a memory is stored.

A brain has absolutely nothing like a vision system that can focus on one particular spot. Nor is there anything in the brain remotely similar to a read-write head. So how could “reading from one particular spot” in the brain even occur?

The reasoning above suggests that our neurologists will never be able to solve the problem of how the mind is able to recall things so quickly. The human brain is simply not structured the way a physical system would need to be structured in order for an instantaneous recall of detailed complex memory information to occur from one particular storage location in the brain. The impossibility of explaining how instant memory recall occurs is a powerful reason for believing that long-term memories are not stored in your brain. We must postulate that the human mind is part of some reality that transcends the human nervous system. Call it a soul reality, or call it a spiritual reality – it is something that must go beyond the human brain.

We can compare the idea that brains store our memories to the theory of Santa Claus, the idea that there is someone who goes around the world on Christmas Eve delivering toys to all the children. If we give just a moment's thought, neither idea seems untenable. But if we try to construct a detailed hypothetical account of how either of the things might work, we begin to realize neither idea can hold up to scrutiny. Once we start trying to construct a detailed theory of Santa Claus, we start to realize things such as that no Santa Claus sled could store millions of toys (enough for all the world's children) in a sled, and that no Santa Claus would have time to deliver toys to even a tenth of the world's children in a single night. Similarly, once we try to construct a detailed theory of memory storage and retrieval using neurons, we realize that because of rapid protein turnover there is no stable place for a brain to store memories lasting 50 years, and that there is no way a brain could instantly recall a particular memory from some very specific spot where just that memory was stored, given the complete lack of an indexing system or coordinate system in the brain, no means by which some very precise location in the brain could be identified. Nor can we think of any realistic theory of how something like episodic memories or learned abstract concepts could ever be physically represented by something like synapse arrangements or neuron arrangements.

Given the complete lack of any coordinate system in the brain by which the exact locations of neurons can be specified, the brain can be compared to these things:
(1) the US phone system if no one's phone number had ever been published;
(2) a vast post office with countless post office boxes, none of them numbered;
(3) a city in which none of the streets were named, none of the buildings had an outside identifier, none of the apartments had apartment numbers, and none of the houses had street numbers.
(4) a vast library in which none of the books have titles on their covers, and none of the chapters have chapter titles.

Imagine how hard it would be in any of such things to navigate to a precise location -- a particular post office box, a particular phone, a particular chapter of a particular book, or a particular apartment. That's the kind of situation that should exist in a brain storing abundant memories, because there is no coordinate system in a brain, and neurons don't have neuron numbers or something like a brain longitude and latitude. Instantaneous recall of obscure memories should be impossible if our memories are stored in brains. The fact that we routinely perform such instantaneous recalls is strong evidence our memories are not mainly stored in brains.

Nor do we get out of these difficulties that assuming that perhaps a particular memory item is scattered across multiple spots in the brain. If I have to access, say, ten particular neuron locations in my brain to retrieve the mind's eye image of Abraham Lincoln when I hear his name, like someone reading from ten particular books in a vast library where none of the books have titles, that is simply ten times harder to explain than if I have to get that image stored in one particular brain location (like getting information from one particular book in such a library). The same difficulty is involved, that there is no way for me or my brain to know where those exact locations were, to be able to instantly retrieve the information. 

No comments:

Post a Comment