[Posted to the neuroscience group on MySpace.com]

>I've heard it put like this many times: Your brain is a muscle. Use it or lose it.

Yes, but it is important to note that the types of mental exercises that are effective are very different than those that develop muscles.

Specifically, when you want to build your muscles, you to repetitive movements with heavy objects. By contrast, the mental exercises that seem to be the most useful are ones in which your brain is constantly challenged by *novel* stimuli, not rote tasks (in other words, the point "dp" is making above). You want to constantly be engaged and kept on your toes.

Memorizing lists of vocabulary words or Latin that you will never use is actually counter-productive. That knowledge becomes compartmentalized instead of generalized. The broader a piece of knowledge is applied, the broader (and more plentiful) the physical connections are in the brain... thus, neurodegenerative diseases have to break more connections to have as dramatic an effect on cognition.

You have to consider what is meant by "use" in "use it or lose it" before you go on a mental workout that might just lead to stagnation.

[Posted to the neuroscience group on MySpace.com]

A few more thoughts on this topic...

Like I said above, the "use it or lose it" aphorism is ambiguous and misleading. Use what? And use it how?

The term inactivity is much better to describe what to avoid. You wouldn't say someone reading a book was mentally inactive, but you might view someone doing the same repetitive task over and over (admittedly, this could include reading the same formulaic romance novels and Mack Bolan adventures).

This doesn't conclusively address this either way, but no one has mentioned the studies of enriched vs. impoverished environments regarding this issue. While rats would obviously be more mentally active if they had some stimulus vs. none at all, it seems that the effect of the enrichment wasn't just to promote activity but rather generalized activity. Rats moved from one task to another, and the result was that their brains were rich in synaptic branching. These dense connections seem to be what prevented the nuns in the famous (and on-going) study from developing Alzheimer’s.

By having a more generalized (i.e., cross-connected) brain, individuals with AD are less likely to present the symptoms until the disease has progressed further. The simple reason for this is thought to be that they have multiple paths to finding the same information. In other words, they can do the same thing in a number of different ways.

The opposite of this would be to become very good at only a narrow set of behaviors. Those brains would have strong connections only in the relatively few pathways associated with the heavily-practiced tasks. If any of those connections is severed, the knowledge and/or abilities conferred are lost.

A parallel to this in the educational community is a reaction to "teaching the test." Standardized testing has proliferated (e.g., SATs, ACTs, state-mandated tests, etc.) such that the style has been to teach to the test. In other words, the emphasis has shifted from the content to the format of the test and how to figure it out.

In reality, the best approach is to cover the material in such a way that students can generalize their knowledge. If they encounter something on a standardized exam that throws them, they would do better to use genuine intelligence and a broad array of knowledge to figure it out than resorting to limited bag of tricks only applicable in a finite set of circumstances.

My childhood is an archeological expedition sometimes. I run across things all the time in movies, etc., where I go, "Oh, so that's where that came from." Seriously, I know all the words to several verses of "No one will suspect you're afraid" (I don't know if that's the actual name) from "The King & I." I found out only a few months ago that was where it was from. There are still a couple others that I'm trying to place.

[Posted to the neuroscience group on MySpace.com]

I don't know where the line is between addiction and associative learning, but there is certainly some overlap in a number of areas (for example, relapses are often triggered by environmental cues drawn from associations made while using the drugs).

As far as single association pairings, there's an example that stands out as particularly unique in associative learning, and that's food aversion learning. Whereas you typically have to have numerous bad experiences with something before you consistently avoid it, you only have to get sick after eating something *once* before you have trouble with even the smell of it for many years afterwards.

This is a biologically conveyed learning mechanism that keeps you from eating anything that your body had a bad reaction to before. Interestingly, it doesn't even matter what actually made you sick on that fateful pairing. You may very well have gotten sick from something else entirely (e.g., a stomach flu) and happened to have thrown up your most recent meal. Guess what? That's the last time you may ever eat that food again.

And it cuts both ways. This also happens to be why those colorful-but-poisonous insects are so effective in their highly visible carapaces. A bird (or other predator) only has to eat one member of the species once in its lifetime to ensure that a nasty association is created in that bird's mind. Thereafter, the rest of the species is protected.

[Posted to the neuroscience group on MySpace.com]

>...cutting up rats' visual cortex...to find that they can still process complex visual tasks.

There are a lot of parallels with memory in this and other things you mentioned. For example, in the '20s and '30s, Lashley performed a similar set of experiments (also on rats) to find the "engram" that encoded a particular memory. Guess what? He failed. Actually, what he found was that the more of the cortex you remove, the worse the rats got at the task (which in this case was running a maze they had previously learned), but they still seemed to remember the entire process. Of course, this wasn't a terribly sophisticated experiment, but it did demonstrate a similar phenomenon.

Similarly, cerebral hemispherectomies are performed on humans (almost always on children for several reasons*) which leaves them with far more than the expected 50% of function by almost any measure. One would think that they should lose half their language ability, memories, etc., but apparently this is not the case even though some deficits present themselves.

*The reasons mentioned above include the fact that this surgery is almost (always?) performed on patients with intractable and debilitating seizures. The cause of these is often due to a condition known as Rasmussen's encephalitis, which usually presents itself in children under the age of 10. Further, children have a substantially greater chance of cognitive improvement following the surgery thanks to their greater plasticity.

There's a term paper on cerebral hemispherectomies on the science page of my site if you're interested in reading more about the topic.

>I saw a video of some sort of functional imaging of the brain looking at simple shapes and the braim emitted signals/patterns that were in the very shape of the visual objects

Actually, this was very likely the case. You have to realize that there are literally maps of things on our brain. For example, we have a map in our somatotopic cortex that is arranged just like our bodies (i.e., hands next to arms, arms next to torso, etc.). You also have a similar map on your motor cortex, although it indicates which parts should move, whereas the other map indicates where we feel physical stimuli on our skin. The level of detail in these areas can, in the case of rats, for example, go down to the level of their individual whiskers. This particular example is a very popular target for a number of lines of research, in fact.

Regarding the visual cortex, rather than perceiving things externally, this area picks up the external world in front of our eyes, so it has a different map (obviously). There have been studies focusing on this area in which, for example, the cells across the cortex were labeled according to which cells were active while an animal looked at a fixed image (You can also record the activity of the cells with electrodes and get similar data). If the image happened to be something like a black and white letter "E," you would find that the image (being relatively simple) would appear in a similar form on the visual cortex.

Incidentally, I tried to track down some images to show you, but most of these are in the journals themselves, so I couldn't find any good illustrations of any of the above phenomena. Sorry. If I happen across something, I will try to remember to post a link. Or if anyone knows of a good picture, please take up my slack.

>Now, I should have known at the time...that video clip was probably a hoax

I would have to see the video in question to assess it, but it sounds legitimate. Of course, they also might have exaggerated things for effect, so it might have come across as "too good to be true."

>...that if you see a chair, your neurons are literally firing in the visual cortex a picture of the chair.

The reason why I used the "E" example above was because at higher levels of processing, you would have a lot more things going on. Specifically, you have two major parallel pathways that each process different aspects of an image (e.g., its shape, movement, color, etc.). We're still working out the details, although a lot is known so far. So, yes, at the *primary* level (i.e., in the primary visual cortex), you would "see" an image literally on the brain, but things get pretty wild after that.

Returning to the "one-to-one" question, I think what you're really interested in is not just vision, but consciousness. There is a much-debated concept usually referred to as the "Grandmother Cell" that looks at that. The idea is that all processing eventually leads back to an internal representation of a single idea. So, when you see your grandmother, all the neurons process information until they reach this single point that represents your grandmother.

Good idea? Maybe. There are ongoing debates about the merits and downfalls of such a concept, and the matter is apparently far from settled. There's a link to an interesting article on it below. Note that is was published in 1996, so there are a few things I know would have been included were it written today, but just the same, I have a feeling it's right up your alley. More than just reading about the concept of "grandmother cells," however, I think it's a good illustration of how difficult it is to get at some of these questions.

Some reflections on (or by?) grandmother cells

http://www.perceptionweb.com/perc0896/editorial.html

There was a study of the London cab drivers years ago that found that they had one section of their brains (can't remember which) that was developed to three times the size/density of the average person. The reason they chose London was because it grew up organically without much in the way of civic planning. Consequently, it's a complete mess, so drivers have to develop loads of mental maps in order to find the most reasonable routes.

I wanted to caution anyone learning about LTP that not everyone is convinced that this is the core mechanism underlying the formation memories. However, you could get the impression that the debate has been settled if you look around the literature casually. This is largely due to the prominence of Eric Kandel in the public and academic realms. He is so central to the study of the phenomenon, but he is also highly visible (e.g., Nobel Prize, his own neuroscience text, and at least three appearances on Charlie Rose that I'm aware of). That presents something of a biased perspective to outsiders when the matter is not firmly settled within the field.

Personally, I'm in Kandel's camp for a number of reasons that I won't go into here, but if you are interested in some of the arguments against LTP as the basis of learning and memory, check out Randy Gallistel's chapter in the book "Conversations in the Cognitive Neurosciences." It put a few doubts in my mind. (Note, however, the interview was conducted around 1997, so maybe his arguments have been addressed experimentally in the interim.)

[A post-car-wreck email with Dani]

>...and recall how suddenly everything happened. So suddenly, I don't even REMEMBER!!

Actually, what happened was that, by going unconscious immediately upon impact, whatever was in your short-term memory was never written to long-term memory. This also happens when people come out of a faint; they don't remember going into one, so they try to walk or continue a conversation just as if they were in the middle of something mere moments before.

There was a murder case in which the perpetrator claimed to have been knocked unconscious by an unseen assailant, but was caught in a lie when he (or she?) provided details right up to the impact of the bludgeoning. The neurologist consulted in the matter flat out told the detective, "That's impossble." The verdict: Guilty.

A few yeas back my girlfriend experienced the post-trauma metabolic depression (as evidenced by memory impairment) that is common with concussions. We were in a serious car accident and she was knocked unconscious. Fortunately, I'm indestructible and nothing happened to me. I'm not sure if she technically had a concussion or not (since I don't know the medical criteria/diagnostic definition), but as is so often the case in events such as this, she had no memory of events leading up to the accident.

Further, she had on-going memory problems for roughly a month afterwards as well. The most severe were, of course, in the days immediately following the wreck. For example, a couple days after the accident, I took her grocery shopping. Since she was still really sore, I would ask her what she wanted, tell her to wait with the cart, then go get things and bring them back to her. The first time I did that, I came back with a couple gallons of milk and she wasn't where I told her to wait. "Where did you go?" I asked. She looked at me like, "I was shopping(?)." I asked her why she moved from where I left her and she had no memory of the conversation.

Fortunately, the symptoms gradually let up, so now she has to make other excuses to ignore things I've said.

Dani refers to the accident as "rolling back to zero" or something to that effect. It happened almost a year after we started going out. Even a year afterwards, she sees us as missing that year. She always thinks we've only been together for a year less than we have been in reality.

I don't know that anyone has a clue about the information storage properties of the brain. Instead, the research into learning and memory are primarily focusing on the creation and maintenance of associations between pieces of information.

Regarding the (probably flawed) computer/brain analogy, the short-term memory issue Tim highlighted above might be thought of as an issue of consciousness being our RAM, as in the things that we're aware of that are in our short-term memory. As much as I love analogies though, we all have to be careful about reasoning by analogy. Just because "A is to B" doesn't mean that just because "C" is analogous to "A" that "C is to D."

>Do you think it is possible to scan pages and pages of material in a matter of minutes?

I would say (in most cases) no. The reason is that a lot of the things our brain does is grounded in existing wiring. Let's take perception to begin with. There are a lot of optical illusions where you don't know what you're looking at unless someone points it out or makes you look at things a different way. The information was always there, but you had to have an "in" to access it. If you happen to be looking at a page full of words, you really need to almost know what the page is trying to say before you attempt to read it. Of course, you're probably trying to read it because it's something you haven't read before, so you haven't seen it yet for this to work. Yeah, Catch 22. As a result, if you gloss over something, you aren't going to have the mental traction to make any headway through it. You have to grasp it bit by bit the way tires grip the road in order to move forward.

This is a variation on the above, but another example of prior wiring being important is that you have to understand the material to some extent already. I'm sure you've seen "The Sixth Sense" already, right? Well, the first time you saw the movie through, it was a completely different film than the second time. Without that understanding of the true situation, you don't fully pick up on all the meaning. If you read too quickly, you don't catch all the information that establishes the context, so what you're reading is not necessarily what the piece was about. You may be missing key points.

Perhaps the most important constraint is that learning actually takes time. Since learning (even in the short term) takes place by physically connecting neurons, you cannot shorten the process indefinitely. There is some lower limit ultimately, and approaching that minimized the efficacy of the learning; going below that means there is no measurable learning at all. I'm thinking in particular about a process called long-term potentiation or LTP. You might look that up in Wikipedia if you want some more background.

The only speed-reading techniques I'm aware of tend to operate not by reading faster but rather through selective omission. For example, you only read nouns and verbs or maybe you only read the first and last sentence of each paragraph. I'm not sure if the approach you're looking at works that way or not. Please enlighten me when you get a chance.

>I trying to bring up my reading time...I feel like I read so slow..I have excellent recall of the material.

These two things are related for precisely the reasons I mentioned above. There's more to it than that, of course, but that's part of it. Reading slowly is like chewing your food instead of swallowing it whole. If you just need a general awareness of something, then skimming it would be fine. If you want to be aware of every detail and/or want any sort of emotional impact from it, then you are going to want to read it at the pace that is most comfortable to you.

>I read well (I'm not sounding out the words or anyhting...the reason I read slow is because when I read I have a voice that comes on in my head...(strangely it is not my voice..or even my perception of my voice..it had a british accent..it is kind of like the narrator of the National Geographic...I can't turn it off.

That's interesting. I get that more when I write in certain styles than when I read.

This is more a tangent than a comment on the above, but one approach you might try is to use a text-to-speech program. You can paste text into it from wherever (ebooks, web pages, etc.), and it will read to you in a still far-from-perfect attempt at a human voice. However, the more you use it, the more effective it is because you gradually get used to it. Sure, it doesn't always pronounce things correctly, but your brain fills in those gaps just like when you pick up on the nuances of any language (e.g., Spanish, ASL, computer languages, visual shorthand like that found in movies and commercials, etc.). At home I have a pair of cordless headphones I can connect to my computer, so I can load a couple chapters from a book and wander around doing housework or even in my yard cutting the grass, washing the car, and so on.

If you're interested in trying out one of these programs, one of the better ones (for free anyway) is the basic version of Natural Voice Reader.

I think we all believe a lot more should be there than actually is. I mean, I think back and try to remember, say, who I ate lunch with when I was in college even, and a lot of the time I can't come up with a lot of people. However, later on I'll remember something that happened and I'll realize someone else who was there at the table. You have to come at some memories sideways, which is why therapy helps with this (But watch out for invented ones about ritual molestations by devil worshipers). Here's an exercise I did recently: I made a list of all the books I had ever read. I thought I had done pretty well on the first pass, but I ended up with maybe another 25% or more within the next few days that I had completely forgotten about, then another 10% over the next few weeks. Since then I've come up with a few more titles (about one every few months). They were part of my past, but it will take me getting on the subject of one before it dawns on me that I ever read it.

>>Did anything "imprint" with you this way?

>I thought deeply about this and I can't figure it out.

An example of this I came up with was the fact that I can remember my schedule my freshman year of college, but I can't even remember some of the classes I took just a couple years ago. The level of detail that stuck with me back then was really, really intense whereas now I hardly recall anything nonessential.

>How can electricals and chemicals lead to memories?

Although this is still far from worked out completely, in theory, it apparently isn't all that difficult. I mean, there are a number of different information technologies. Admittedly, they're usually quite a bit different than the brain, but they even surpass our abilities in terms of speed and accuracy of recall, just to pick one example.

>Thoughts? Dreams?

This is a more complicated question. Check back in a hundred years. Or maybe we'll all be stored digitally and be on-line by then. In which case, let me say hello to my digital self who will be reading this after my physical death. "Hello, Ale[x]adecimal!"

>If we ever figure out how the brain works fully, if it can ever be done, think of the amazing things that can be done. Like you can be on the internet or something and get massages, have sex, feel things in cyberspace if they knew how to stimulate those feelings.

You're getting into brain-computer interfacing and/or transhumanism. Those are both terms for essentially the same thing. You can find a lot of speculative pieces (philosophy, essays, etc.) on transhumanism, but there is also time devoted to the topic of brain-computer interfacing at the SfN conference. I haven't looked at the schedule yet this year, but there is usually a symposium with speaker after speaker presenting research and relevant engineering and/or methodology into the this area for at least a day, but I think it ran for two days the last time I was there. If you aren't there to see anything else, you can sit in the auditorium and gorge on the information until your brain explodes.

Learning = re-wiring, but the blank slate model assumes nothing is wired together. Since that is not the case, it means the brain comes with wiring that has built-in (or in-built) meaning.

Back to the index