The Science of Everything Podcast - Episode 60: Memory Encoding and Recall

Episode Date: March 16, 2014

An analysis of memory, including a discussion of the different types of memory, the capacity of short term memory, how environment and depth of processing influences learning and recall, the relationship between memory and familiarity, and different types of amnesia. Complements Episode 20: The Reliability of Memory. If you enjoyed the podcast please consider supporting the show by making a paypal donation or becoming a patreon supporter. https://www.patreon.com/jamesfodor https://www.paypal.me/ScienceofEverything

You're listening to The Science of Everything podcast, episode 60, Memory Encoding and Recall. I'm your host, James Fodor. In this episode, we're going to look at memory, including a discussion of the different types of memory. So we'll look at working memory and long-term memory and also semantic procedural and the various different types. I'll talk about the length of working memory in terms of how long it lasts, how much information it can hold, and various interesting effects that influence how much we remember of things in short-term memory. I'll also talk about things that influence how much we can recall from memories, including level of processing, incidental versus intentional learning,

mnemonics, retrieval paths, and state-dependent learning. I'll also talk a bit about amnesia and the difference between familiarity and actual memory, because it turns out they're not quite the same thing. There are no prerequisites for this podcast, although the material in this podcast complements well with that in episode 20, the reliability of memory. So check that out before or after listening to this if you find this of interest. So, first of all, let's start by talking about working memory, because it's sort of a foundational idea. Working memory is the system that holds pieces of information for a short period of time. The exact length of time sort of varies depending on how you test it.

It's something on the order of like 30 seconds, maybe a minute or two, something like that. But it doesn't last very long. This is also sometimes called short-term memory, although there is, some people draw a distinction between working memory and short-term memory, and some people don't. I'm effectively just going to talk about working memory in this episode. I won't really distinguish between that and short-term memory. So working memory holds a few pieces of information for a short period of time. And as we'll see in more detail later, working memory is an active system. It must do work to actively retain the information that's held there,

otherwise it's lost very rapidly. Now, working memory has a number of subsystems or components within it that allow it to function, and the precise relationship between these hasn't been fully determined yet, so there's still different models that cognitive psychologists debate between. But one common model, which is called the Baddeley model, includes a central executive, which coordinates and shifts attention and information around between the subsystems, so kind of like the CPU, a very vague analogy. And then two other subsystems, one called the Visiospacial Sketchpad,

which stores visual information like shapes and colors, and another subsystem called the phonological loop, which can briefly store sounds, particularly speech and rehearse them. The idea of the phonological loop is that it's a sort of a partly conscious, partly subconscious sub vocalization process. Subvocalization means repeating words silently in your head. So this is probably something that you've experienced. If you're trying to remember something, you repeat it silent to yourself in your head.

Well, the idea is that is done by the phonological loop. Similarly, if you're trying to remember a visual image of someone's face or whatever, then you picture that in your mind, and that is done by the visual spatial sketchpad. The central executive, according to this model, mediates the transfer of information in and out of working memory and also between visual spatial sketchpad and the phonological loop. So the important thing to understand is that working memory is not just one box that things go in and out of. It's a complicated system that has multiple sub-components to it, which do different things, and interact in interesting ways.

And as I mentioned before, it's thought that, working memory is an active process. So to explain what that means, think about how the phonological loop is thought to work. One hypothesis is that it's not exactly a memory system per se. It doesn't store a memory like a hard drive does. Rather, what it does is it simply continually repeats a particular sound or, you know, a phrase or something. And so effectively what you're doing is you're saying that in your head and then listening to what you said and then saying it again. So in that process of listening and repeating what was said, the information is stored actively in that system of neurons. But as soon as your attention focuses to something else, that information is lost,

is the idea. So it's an interesting idea that basically it's not storage per se, it's just continual recreation of the memory, keeping it alive until you focus your attention on something else. So that's the idea of working memory. Let's talk about a few interesting phenomena and findings that are associated with working memory. One is called the serial position effect. This is a very robust finding over lots and lots of different studies. Basically, the paradigm of studies related to the serial position effect involves trying to remember a bunch of items, so it could be numbers or words or concepts, whatever. The subject is presented with a list of these items, and they have to try and remember as many of them as they can and

repeat them back afterwards. Obviously, longer lists are harder to remember, but beyond, you know, really small lists are easy to do. Everyone can remember all the items. But as the lists get longer, people begin to forget some of the items, and it turns out that there's a systematic way in which the forgetting occurs. Specifically, people are more likely to remember the first few items on the list, and also the last few items on the list,

and they're least likely to remember items from the middle of the list. You can easily test this yourself if you find a list of something to remember and find out and repeat it to yourself or get someone else to repeat it to you, test yourself, see which ones you remember. Generally, it's easier to remember

the most recent ones, and the very first one's much harder to remember the ones in the middle. Now, these two effects are called the primacy effect. That refers to the fact that items near the beginning of the list are more easy to remember. And the recency effect, which refers to the fact that items near the end of the list, are easier to remember. The primacy effect is thought to be due to the fact that there is more time for rehearsal. So, for example, usually when people are trying to remember things on a list,

you hear the first item on a list, so maybe the first item is Apple. So mentally you repeat to self. One repeats to oneself, Apple, Apple, Apple, Apple, Apple. You're trying to remember that. But then a short time later, the next item comes, Banana. So then you have to say Apple Banana, Apple Banana, Apple. And then the next item comes Pear. Apple Banana pair, Apple Banana pair.

And then that goes on and on as the list gets longer and longer. You'll see that the very first item, Apple, gets the most repetitions. The second item, Banana, gets nearly the same number of repetitions, but one less. Items near the end of the list get the fewest number of repetitions, because they haven't been included for as long. This is thought to be the reason for the primacy effect. The idea is that the items put into the list first have been repeated more times. There's more time for rehearsal, and so there's better transfer to long-term memory, or at least partial transfer. The recency effect is thought to be a result of the fact that as you get more and more items on the list, as the

capacity of your short-term memory is reached, then old items are displaced. And so, the items left over at the end of a long list will be the items still in the short-term memory, still in the working memory that haven't been displaced by newer items will of course be those items that were most recently put into the working memory, that is, the last items on the list. So that's thought to be the cause of the recency effect. And there's a reasonably good amount of evidence in favour of the distinction between these two and also the reasons for them.

So, for example, if you give someone a distractor activity that lasts for just you know, 20 or 30 seconds after the memory task, you know, a distractor activity is just anything like, you know, read this passage or whatever, just something to distract their attention. Then what happens is the recency effect goes away. They don't remember the last few items any better than the middle items, but the primacy effect remains. So that's consistent with the idea that there's a different reason for each of them. Similarly, there are a certain group of amnesics who do not display a primacy effect because they have a very poor ability to transfer the information from working memory to long-term memory.

So they don't have a primacy effect because there's no effect of increased number of repetitions on memory, but they do show a recency effect. So this is what we call a double dissociation, a very important concept in psychology. When we have two phenomena, A and B, in this case it's primacy effect and recency effect. A double dissociation exists when we can find a condition where we have A without B, so recency effect without primacy effect, and also another condition where there's B without A, so primacy effect without recency effect. effect. So when we can separate the two conditions, A without B and B without A, then we say there's a double dissociation, and that provides good support for the hypothesis that the two processes, or cognitive processes or whatever they are, are distinct and separate things. If that doesn't exist,

that doesn't necessarily mean that they, that they're the same thing, but it's harder to tell, because if they always go together, then it's more difficult to determine. Okay, so that's the serial position effect, quite an interesting phenomenon. Let me just talk a little bit about the idea of memory rehearsal. So rehearsal just sort of means repeating something to yourself, and that's a way we try and remember things. But there's sort of two different types of rehearsal that call maintenance rehearsal and elaborative rehearsal. Maintenance rehearsal is the type of rehearsal we were talking about before, where you just repeat an item over and over again to try and keep it in your short-term memory. And that can be effective in doing that. It helps us to maintain the information in working memory,

but it's a completely ineffective way of processing the information and transferring it as a long-term memory. So an example would be if you look up a phone number that you need to call and you repeat it over to yourself over and over again so it stays in your working memory, you could do that, I don't know, hundreds of times for, you know, many minutes. And then, say, you called the number and you were distracted, and then after the distraction is over, you return your attention to the phone call and you try and remember the number, and it's just gone. It doesn't matter how many times you repeated it before. It hasn't been transferred to long-term memory. Again, I'm not saying this is the case every time. people's memories work in somewhat different ways. But the basic finding is that it doesn't matter how many times you rehearse something

in a simplistic, just repetitious way. It doesn't really help. If all you are doing is maintaining the item in working memory, that does not help to transfer it to long-term memory very much at all. Elaborative rehearsal, on the other hand, involves thinking about the meaning of the information and connecting it to other concepts already stored in memory, and this is much more useful for consolidating information into long-term memory,

as we'll see later on. Now I want to talk about the notion of the size of working memory. Working memory has a very limited capacity. There's a famous study called the magic number seven plus or minus two, which, the main finding of which essentially was that the capacity of adult working memory was seven items plus or minus two. So seven is the average and some people are a bit more, some people are a bit less. The capacity of working memory is one thing that's often tested in IQ tests, for example.

it's one facet of intelligence. But the variability isn't that great. Most people, it's around seven. However, if you were listening, you will notice that I said it's seven items. And that's crucial, because what constitutes an item depends very much on the context and how the information is being processed. So this leads us to a very important concept in memory study called chunking. Chunking is the grouping of elements together into classes in order to help increase total recall. Now, a good example of this would be, if I was just memorizing digits, well, not memorizing, but keeping them in my working memory, I could probably remember maybe seven of them, four, five, eight, one, whatever, seven separate digits, because I'm treating each digit as a distinct element. I'm trying to remember them individually. But suppose that I had a sequence of digits which corresponded to years, so I had 1972 that I can think of as 1972. and then I had 2042, 2042, and then I had 1288, 128. In that case, I could potentially not think of the digits as separate numbers, but think of them as years.

And so in that case, instead of only being able to remember seven digits, I could remember seven years. And each year, of course, consists of four digits. So that would be four times seven, or 28 digits. Now, that's just one example. In order to work, though, you would have to be able to encode the information in the form of years. So this is why it's a really... Chunking is quite an idiosyncratic thing, meaning it varies a lot between people, because what can meaningfully be processed and remembered as a chunk?

It really depends on who the person is. I read about one individual who was like a horse racing fanatic or something, and he was able to encode arbitrary sets of digits as, like, finishing times or something for a horse race. And by doing it in that way, he was able to remember like 50 digits or something like that, just arbitrary digits. I don't mean memorize, as in he memorized that many digits. That's not particularly impressive because some people have very impressive ability to memorize long ways of digits.

I mean he was able to hold that many digits in his working memory, which is astounding, given that the normal limit's about seven. So in order for a person to use chunks, the chunks have to be meaningful or familiar to that person. So for me, for example, I like history, and so I often think of. sequences of four digits as being numbers, sorry, sequences of four digits as being years, but that only works if it starts with a one or maybe

a two. It doesn't work to start with a three, obviously, because, you know, we haven't gone up to that many years. And so, chunking is context-specific. Another way of remembering things could be thinking of numbers in terms of birth dates, or as parts of phone numbers,

or addresses, or whatever. Any meaningful thing that allows you to identify this sequence of information, with one single chunk of information, one item that you can recall from long-term memory, will enable chunking to take place. So, again, the basic idea is to sort of hack into some information, some contextual information stored in long-term memory

to help you remember longer sequences of information than would be possible if you just remembered each item individually. So instead of remembering seven digits, you remember seven years or seven dates or something like that, which allows you to remember a larger total number of digits than if you just memorize them or thought of them individually. And that can be done for names or for anything really. Digis is just a common example that's used in the research.

So, just to recap there, the size of working memory is about seven items, but what counts as an item is context-dependent and can be changed through chunking. And you can improve chunking, by the way, through practice. The more you practice chunking, the better you get at it, and the more things you can remember. Okay, so let's move away from working memory now and talk a little bit about long-term memory. Long-term memory refers to the, well, long-term storage of information that allows us to remember things from anything from yesterday to many years ago. So it's, I mean,

most of our memories are held in long-term memory because the capacity of working memory is so small. Almost everything that we know is in long-term memory. But there are different parts of long-term memory. Not every type of memory is stored in the same way. Generally, the three main types of long-term memory that are discussed are episodic, semantic, and implicit or procedural memory. So, let's look at all of those in turn. Episodic memory refers to memories that are effectively events or episodes that have happened, especially to us, although it could be something we've watched in a movie or something. So, you know, your first day of school or your first day at work or, you know, first date, whatever, any of those things would be episodic memories, or it doesn't

have to be at first anything. It could just be what you had for breakfast. eating breakfast yesterday, that would be an episodic memory. Semantic memory is second type of memory, and that's related to storage of facts about the world, or pieces of information. So, you know, London is the capital of England. Water is wet, or water boils at 100 degrees Celsius, or, you know, anything like that.

Any of those things are semantic memory. You see how it's different from episodic memory, because episodic memory is memory for an event, as sort of a sequence, and would include, you know, sight and smell and sound and all those things put together and a narrative of things that happened. The two are obviously related, so you could say, for example, that you have a semantic memory that your first day of school was on such and such a date in such and such a year.

That would be a semantic memory. But then if you thought back and imagined what it looked like and what you felt and whatever else, then that would be the episodic aspect of the memory. And so the two are related, but potentially distinct. You can have a semantic memory. Well, I remember that my first day of school was at this place, on this date, but I don't have any episodic memory about it because I don't remember anything about it. Or potentially vice versa. You could have a very distinct memory for a particular episode.

Like you remember being in a particular place and something happened, but you can't remember what the context is or when it happened. So in that case, you've got the episodic memory, but not the semantic memory about it. Both of those two together, semantic and episodic, are referred to collectively as explicit or declarative memory. And this is because they are available to conscious awareness. So you know what semantic memories in episodic memories you have, you can report them and talk about them. In contrast, procedural or implicit memory is not accessible to conscious awareness. So this is basically, it's often described as how a memory, how to do things, muscle memory or motor memory or something like that.

A really good example of this is how to ride a bike. People who've learned to ride a bicycle, very seldom forget, unless they've suffered severe brain damage or something like that. But if you ask someone, how do you ride a bike and ask them to explain it, it's just most people would not be able to do that. They don't have conscious access to how you do it. They could get on a bike and show you, but they couldn't really explain it. Indeed, it's quite possible for people to remember how to do things like ride a bike or play a sport or something like that, even though that they have no episodic memory of having learned how to do it, because they've forgotten, and no semantic memory, no ability to explain how to do it. Nor do they have any semantic

memory of when they learn. So they completely lack any episodic memory or semantic memory of learning it or how they learned it or how to do it, but yet they can still do it. So that's an instance where the procedural or implicit memory is in place, but the other types of declarative memory are not in place. So it's very interesting that the brainstores the information in different ways like this. So, and these three types of memory, then procedural, semantic and episodic make up the three main types of long-term memory. Now, a question you might be wondering, how, what is the capacity of long-term memory? We know that short-term memory only lasts for, you know, 30 seconds, a minute, maybe two, a very short amount

of time, and that you can only hold seven items, so very limited capacity. What about long-term memory? What's the capacity of that? How long does it last? Well, there have been some interesting, a few interesting studies about what's called very long-term memory, so memory over many decades. One way that this is tested is by conducting face recognition and name recognition of people using photos and names from college graduation yearbooks, because that's something that can easily be verified. And the accuracy of identification of names and faces diminishes over time. So even after 48 years, recognition of faces and names was, yeah, recognition of faces and names was something like 70 or 80 percent. Accurate. Free recall, which is not asking you to recognize

someone, but to generate a name in response to a face or something like that, was lower, but still considerably better than chance. So it's interesting that even after 50 years, people can still remember quite a bit. Other studies have looked at this as well. Basically, as far as we know, there is no limit in terms of the capacity or duration of long-term memory.

Now, particularly in regards to capacity, I mean, obviously there must be a limit. It's obviously not true that the human brain can store an infinite amount of information because it's a finite, it has a finite volume. So there's a theoretical maximum amount of information that it could possibly hold. But practically speaking, there's no known limit

to how much information someone can recall. In most cases, it's a problem of accessing the information rather than a problem of retaining it per se. And that's not really surprising because the way the brain stores information isn't like a computer hard drive. It's not like that there's 60 gigabytes, and when that's used up, it's used up, and then it's finished.

The brain stores information, as far as we know, in the form of patterns of connectivity and strengths of connectivity between neurons. And without going to too many details that would be a bit distracting from our main focus here, the way you store information in that is a lot more flexible than the way you store information on a hard drive. It's not a simple linear thing. You can keep packing more information in at the cost of a slight degradation in the quality and ease of access of existing information. So there aren't sort of hard and fast limits on the amount of information you can store in neural networks

in the same way as there are, say, on a computer hard drive. So, yeah, as far as we know, there's no real meaningful limit on the capacity or duration of long-term memory. People can remember a lot of things for a very long time. Okay, now I want to talk a bit about some interesting phenomena relating to how people remember and that affect the quality of learning and memory, and also of recall. So one is, one interesting finding is called the Levels of Processing Effect. basically the main finding here is that the amount you remember and the ability to recall the information is largely dependent upon how deeply or how systematically that information was processed.

So what I mean by that is sort of how much you've thought about the information. So a particularly common finding is that if you ask people to conduct shallow processing of information, so that means like asking them what does this word start with or what does it sound like? them, you know, phonetic questions or orthographic questions that don't require very deep processing, it just require you to look at the word or think about how it sounds. That doesn't, that type of encoding does not lead to very strong memories. Conversely, if you ask people to, uh, questions that involve semantic, deeper levels processing, you know, what does this mean or leaking concepts between, uh, across different ideas or different words, that results in more durable, more

reliable memory formation. So basically, the more you think about something, the more connections you make with other pieces of information, the better you'll remember it. And this applies for a wide variety of different types of information and different types of learning. So it's a fairly robust finding. When people process things according to superficial qualities like position or sound of the word or whatever like that, they don't remember it very well. When they process it more deeply based on meaning and connections with other concepts, they remember it much better. And the effect isn't small either. The likelihood of a correct memory is like three or four times higher.

with deeply processed material than shallowly processed material. So it's a very sizable effect. And there's some very interesting studies of this that have been... A common one that's been replicated a number of times is asking people about the exact positioning and wording of the portraits on their currency. So I think the first study that was done was regarding American pennies. And they showed Americans a number of...

of pennies, except only one of them was actually a correct penny, and the others had various elements rearranged, like the wording was, in God we trust, that little slogan thing was moved around, or the head was facing in the wrong direction, or the text was the wrong size or something like that, and very few people could recognize the correct one, because all they remembered was sort of that it had roughly these elements in it, the penny had roughly these elements, but they didn't remember exactly how they fit together. And this is, despite the fact that people, you know, Americans deal with, well, most Americans deal with pennies and money very regularly. And so they've seen it probably hundreds or thousands of times. But the point is that

there's no real requirement or need or motivation for them to do any deep processing of the image of the penny. And so they never really think about it. And so they never really remember it, despite being exposed to it hundreds or thousands of times. And this has been, this basic paradigm has been replicated a number of times with different countries in different currencies and other things. Basically, people do not have very good knowledge or detailed knowledge of everyday objects, even things that they interact with on a daily basis, because they are because they're not required to think about that particular aspect of it, say, the appearance of a penny in any deep way. They're not required to process it meaningfully. So this is a fairly

clear evidence that mere exposure to something doesn't help memory very much. You need to process it and really think about it. So this completely undermines the notion that, for example, if you're trying to learn language, for example, I think this is a common misconception, if you just play the audio file while

you're doing things, or even while you're trying to fall asleep or while you are sleeping, even more foolishly, that this information will just sort of somehow seep into the brain and be remembered. And really the information, sorry, the research mediates against that. Only if you actually pay attention and process what's being said, will you remember it.

If it's just there, it doesn't matter how many times you play it. If you're not paying proper attention, if you're not really processing it, you won't remember it. It just mere repetition does not seem to help memory very much at all. Okay, and there's another somewhat depressing finding, which is that, which broadly called the difference between incidental and intentional learning. And the essential idea here is intentional learning occurs when you deliberately try to memory, remember something, you think, okay, I need to remember this and you try and remember it. Incidental learning occurs when you just remember something even though you sort of didn't really

intend to, which you just sort of happen to transfer that into long-term memory and remember it. The basic finding here, which again is somewhat depressing, is that it doesn't matter if you're trying to remember something or not. It makes no difference by itself. The only thing that matters is how deeply you process the information, as I mentioned before. Now, so obviously, if you're trying to remember something, you can deliberately process it more deeply. So you can think about it and try and remember it that way. So, I mean, in that sense, intentional intentionality does help. But if you get people to engage in deep level processing of something, then they're more likely to remember it, even if that wasn't their intention. So in other words, it's the level of processing

that matters. It's not whether the learning was done intentionally or incidentally. And again, they've done studies which tease out the difference of these effects. Similarly, If you get people to process something only shallowly, shallowly, what have you say that, it doesn't matter if they're intending to remember it or they have no particular motivation to remember it. It makes no difference. They still don't remember it either way.

So, again, the main finding here is that what matters is how much you process the information and not whether you actually intend to remember that information. So this notion that if you just really want to remember something and expose yourself to it heaps and heaps of times, you'll really be able to remember it, is unfortunately not supported by the evidence. You really have to think about it and process it in a deep way. You make connections with other concepts.

That's how you remember things. One quote that I quite liked was, we memorize well when we discover order in the material. So you understand it. You discover patterns and order. You connect it to things you already understand. So, you know, for example, if you're remembering dates in history, if you connect things up in a narrative and think,

oh well, in this year, I remember these other things happen as well, so I can draw the connection there. And, well, it makes sense because, you know, this event needed to happen in quick succession, whereas this other event, well, that was more disconnected, so it happened a longer time later, so, you know, that makes sense, I can

understand that, there's an order there. You're understanding orders and patterns, you're connecting things up. You'll be able to remember it much better than if you just try and repeat the names and dates over and over and over again. Okay, another interesting finding relates to what are called retrieval paths.

Merely storing information in memory is insufficient unless you actually have some way of retrieving that information later on when you need it. So it's not good enough just to be there, as anyone who knows that they have some file or document or whatever that they need somewhere in this massive pile of other papers. Merely having it there is not sufficient. You have to be able to actually find it as well. And it's the same thing with information in the brain. Often when people say they've forgotten something, really they haven't forgotten.

that they just are unable to retrieve it at that time. So this is what's called a failure of retrieval. Retrieval paths refer to basically sequences of memories or associations in memory, connections of concepts that allow you to sort of hook onto the memory that you're trying to access. And basically, the more retrieval paths you have and the more diverse they are, so the more ways there are of getting to a particular memory, the easier it will be to recall that memory. and therefore the better you will remember it.

Or, well, it doesn't really change how much you remember it per se, but it does change your ability to recall it, which sort of comes to the same thing. So that's another reason why it's important to process information as much as you can, to try and make as many connections between different concepts, to think about something in different ways, because the more you do that, the more connections you'll make, and the more retrieval paths you'll set up.

And so the easier it will be for you to recall that memory later on. Again, this is very consistent with the idea of the brain, is a big neural, distributed neural network, because literally it's the way you recall things is having some neurons activate, which then transfer their activation to other neurons, which then transfer their activation into further neurons that they're connected to, and, you know, the action potentials propagate, and there's a sort of transmission of activation through the network. So this is sometimes used sort of figuratively speaking when we're talking about memory in one concept activating another concept. But in a very literal sense, that's exactly what

does happen in the brain through the neurons. So it sort of makes sense that this is how our brains would work. Again, this is very different to a computer. When a computer needs to remember something, it gets the address of where the memory is, where that particular information is stored, and it just accesses that address. There's no need to go through one concept to another concept to another concept to get a retrieval path. So it's a very different notion. Again, so this this broad point has come up a number of times, so I'll just emphasize it, that there's a tendency to think about memory in terms of like a computer hard drive or a filing cabinet or something like that, but it's not like that at all. It's very different. The way that information is processed,

the number of retrieval paths you have very much affects what is remembered and how it's retrieved. There are different types of memory which are stored in somewhat different ways, and you can forget some with that remembering the other. So in many ways, the way we remember is very, very different to how computers remember things or how files are stored in a filing cabinet or anything like that. So don't take that analogy sort of too seriously. I want to briefly mention mnemonics. Now, if you want to

Google this term and you haven't heard of it before, be aware that it starts with M. It's spelled M-N-E-M-O-N-I-C-S. Nymonics. And it just refers to any technique that allows, that aids in information retention. So it helps you to remember things, basically. There are

many, many, many different mnemonics. So, one One that most people probably know is a mnemonic to remember the order of the cardinal directions in the compass. So, well, actually, there are many different. The one I remember for the cardinal directions north-southeast and west is never eat salty worms. There are many others as well. But that's just a trivial example of a mnemonic.

It's something that helps you remember it. There are heaps and heaps of mnemonics for all sorts of things. Often they involve acronyms, so the first letter from each word of whatever it is you're trying to remember stands for something and you form an acrionics. him for them. Anyway, that's a very effective way of aiding memory because basically it's a way of setting up more retrieval paths. The first item reminds you of the second item reminds you of the third item. Another very interesting finding is the, what's called state-dependent learning. The basic idea here is that it turns out that you remember things better or you are able to recall information more easily when you are doing the retrieval in the same context than

in the same context which prevailed when you encode the information in the first place. So, for example, if you study at your desk, you'll remember that information best if you're sitting at your desk. There have been some rather bizarre studies done about this, for example, if people remember, memorize words or other things, when they're scuba diving at the bottom of a lake or something, then they remember that information better if they're asked to recall it later when they're also at the bottom of the lake, and not when they're up on the surface somewhere, and vice versa. That's sort of an extreme example, but again, this basic effect has been replicated many times for all sorts of different contexts.

The interesting thing, though, is that it doesn't really matter what the actual physical environment is. It's not actually where you are. It's not being underwater itself that helps people's memory in that case, or being at your desk itself that matters. What matters is the psychological context. So it turns out that if you ask people to imagine they were in a particular environment, imagine they were underwater or imagine you were sitting at your desk, then they do almost as well, or just as well, depending on exactly the experiment,

but their recall is improved as if they were actually in that environment. So it's not about where they actually are. It's not about where your body is. It's about your mental state, the psychological context, because it could be that particular clues from your environment serve as hooks. Remember retrieval paths. They help you to remember things. And it doesn't need to, sometimes it could be.

explicit. You know, the lamp reminds me of some particular concept and you know, that helps me remember things. But I think for the most part, it's probably implicit. There's nothing particular about the environment that seems to be necessary to help you remember. But it, however, our brain is processing that information,

is encoding that information, it does so in such a way that having the cues of the environment in which encoding occurred available to us, whether it be through the directness of the senses or just through imagining that you were there, having those available helps in the recall of the information that was learned there.

So this is context-specific learning. Context reinstatement refers to the specific phenomenon whereby memory performance improves if the original context during learning is reinstated. So, you know, if the divers are put back underwater, their memory improved. So that's context reinstatement. And again, it doesn't have to be physical. It can just be psychological, if you're asked to imagine that environment. And there's a very interesting application of this where it turns out that if you

encode information while intoxicated, it's actually easier to retrieve that information if you are also intoxicated compared to when you're trying to recall it while sober. So this is, this finding is often popular among students because the argument is, well, if I was drinking while studying, then clearly I should drink while taking my exam because I'll perform better.

So there is some support for that notion, although it is certainly true that alcohol in general impedes, memory, so the best approach is just not to drink at all while studying and while taking the test, but I suppose if you were inubriated while studying, it may improve your performance to be slightly inebriated while taking the test as well. But don't treat that as encouragement on my part. I'm just saying what the research seems to show. There is another interesting phenomenon related to memory recall, which is called source amnesia and also a related idea misattribution. Source amnesia is the inability to recall where or how something was learned,

even though you can remember the information itself. So this is very, very common, and often what people do is they misattribute a particular piece of information. So you remember the information, but you think you got it from somewhere, where in fact you got it from somewhere else. And this actually can be really problematic. It can lead to, well, it can lead to a number of somewhat disturbing phenomena. One is called the illusionary truth effect. This is the tendency you to believe information to be correct simply because we have been exposed to it. So people, when presented with some statement or piece of information, if they've been exposed to it in the past, it seems familiar to them. They can recall it, they recall being exposed to it

before, they have a memory of it, but they don't necessarily remember where that came from or what the context was. And it turns out people are much more likely to say that the statement is true when they have that familiarity, when they can remember it. essentially because of misattribution. They're thinking that they got that the statement was made in some sort of factual context or that it's at least likely to have been, even if it was in fact, was not. So this is how, for example, misperceptions can spread because studies have been done where even if you have like a newspaper headline which raises a sort of a leading question,

is politician ex-corrupt or whatever, people will remember seeing that, but they'll forget the source where it came from. And they'll just remember, what they'll remember in their heads is, such and such is corrupt. They'll remember that as being true, or at least they'll rate it as being more likely to be true, even though the source they got it from was completely unreliable. This even happens if you tell people that the information is unreliable or coming from an unreliable source. So if you say, you can flat out tell people, this is a false piece of information, and then you tell them the piece of information, and you reinforce them, this is not true,

and then, you know, you come back a couple of days later and ask them if that information was true, they're much more likely to rate it as being true than if you hadn't told them, if you hadn't told them before. Again, it's because they don't remember where they heard it before. It just seems familiar to them, and so they're more likely to rate it as being true. This is misattribution. And this has led to some really tragic occurrences, including at least one case, probably more of a curbed, at least one I know of,

where a female rape victim falsely accused a memory doctor of being her rapist. And the reason was because she'd actually seen the doctor on television shortly before her attack, and she misattributed the doctor's face with that of her attacker. So she confused the memories and got them mixed up. And she was very confident about the identification she made because it was very familiar. She remembered the face, but she confused the source through misattribution. And there's an interesting related phenomenon called cryptelamnesia.

This is where a forgotten memory returns, but without it being recognized as a memory by the subject, and who therefore believes that it's something new and original. So this can lead to a phenomenon called unconscious plagiarism, which I think I've mentioned in a previous episode, where people plagiarise elements of stories or songs or whatever else without realizing it, because effectively they have, as a result of cryptoamnesia, basically. They heard of it but forgot it.

And then the information resurfaced, but they didn't remember that it was a memory. They thought it was something that they thought up themselves. And as I've been hinting at, there is a big difference between being familiar with something and having specific memory of that. Or to put it differently, it's possible that you can have a memory of something, without any sense of familiarity, although that's relatively rare, and also, as probably everyone is experienced, you can have a sense of familiarity even without any specific memory

of that item or event or whatever. So, Deja Vu is a particularly well-known example of this. This is essentially a sense of familiarity of an event without any particular explicit memory of that event having occurred before. But there are other examples as well. So another example would be that you can recognize a face. You know you've seen that person before, or you know you've heard the name before. is very familiar. It feels special, but you have no conscious memory of seeing it before or where

it came from. At least at the time, maybe you'll remember it later. But still, you know, when you're in that situation of not being able to remember, you have familiarity without memory. There's an interesting effect here as well. We call the mere exposure effect. People tend to develop a preference for things merely because they are familiar with them. Again, it's thought to be related to this idea that we perceive things as being special or when we recognize them, when we're familiar with them. And sort of part of that in some sense bleeds over or interferes with our evaluation of that item, and we prefer it more. Another related idea is that when we are familiar with something, be it a word or a face or a concept or whatever,

it is easier for us to process that. So there's a greater degree of what's called perceptual fluency. It's easy for us to read the word or recognize the face. There's less effort required to process it. There's less sort of disruption of the processing. of the stimuli, this perceptual fluency in turn increases positive affect. So that means that you feel more highly disposed towards the subject. I mean, and this sort of makes sense.

If something's difficult or hard to understand, you get frustrated, you get annoyed, you don't like it as much. And so then if you even see the thing again, you feel a little bit off, or at least maybe you don't actively feel angry, but, you know, positive affect is decreased. Well, then, you know, the reverse happens if we have something that's familiar, that's something that we have familiarity with, we've processed many times before. So that's the relationship of familiarity and memory. One final point would be just to emphasize that the fact that one is familiar with something

that it seems special doesn't... shouldn't be taken as very strong evidence that you remember that because of the problems of crypto amnesia and misattribution and false memories and other things as well, which I've talked about in this and other episodes. So there is a sort of dissociation between those things. A final few words on amnesia, which you probably heard of before. amnesia just refers to some deficit in memory. It's often caused by brain damage, disease, or psychological trauma.

It's usually temporary, although particularly if it's due to brain damage, a legion in particular region of the brain, it can be permanent. There are two main types of amnesia. Antiragrade amnesia refers to the inability to create new memories. Subjects with anterior grade amnesia are unable to form new long-term memories. They still have intact working memory most of the time, so they can retain information. You can have a conversation with someone who has an anterior-grade amnesia.

A patient HM is a famous example used in this because he had a particularly severe case of this as a result of a bilateral hippocondectomy as a result of an attempt to alleviate his severe seizures. The hippocampus is the portion of the brain that's largely responsible for transferring information from working memory to long-term memory. He had a bilateral removal of his hippocampus, so he was unable to form new memories. He had anterior grade amnesia. But he remembered perfectly well everything that happened prior to his operation. and he was still able to recall information for a short period of time, so because of working memory. So you could talk to him, and he could recall what was being said in a conversation. But if you left the room and came back, he didn't know you before the operation,

then he would treat you as if you were a total stranger because he'd never seen you before. Interestingly, though, H.M. was still able to learn new procedural information, so he was taught, I think, to play the piano or tennis or something like that, a number of things like that. So he was able to, he got better and better at those things over time, even though he had no memory of ever having done it before, which would be kind of bizarre if you think about it, because you ask him, can you play the piano?

No, and then he sits down and he starts playing, and he discovers he's very good at playing the piano. I mean, this was literally how it was, and it just shows how our memories are very strange and that we can remember some types of things, but not others. He also displayed priming effects. So that priming effect is when you are more quick to, you are quicker to respond to some stimuli

when you've been exposed to it before. So, for example, let's see, if I'd just been reading a list of lots of different types of food and then someone asked me to recognize, to a word recognition task, you know, saying, is this a word or it's not a word, I'd be quicker to respond if the word that I was looking at was a food-related word

than if it was, you know, a car or, you know, poker dots or something like, or an unrelated word. That's an example of a priming effect. So HM exhibited priming effects as well, even though we had no memory of the previous list that he had read that generated the prime effect. Again, because both of those things, the priming and the...

Sorry, the priming and the procedural memories are implicit, so they're not... He didn't have conscious access to them. And it's thought that different parts of the brain are responsible for mediating those types of memories, but the details aren't really very well understood. So that's anterior grade amnesia. Retrograde amnesia is the one people are more familiar with, probably from Hollywood or whatever. This refers to the inability to recall information before the onset of the amnesia.

So you can form new memories, but you can't remember what happened before the amnesia began. Retrograde amnesia is often caused by a head. trauma and is often temporary, although not always. And often episodic memory is more likely to be affected by semantic memory. So this is somewhat accurate in the movies where you see people are still able to talk and drive a car and remember facts about the world, at least sometimes, but not remember episodes from their life. That's because they have semantic memory partly intact and procedural memory largely intact, but episodic memory largely not intact. Also, recent memories are less likely to be recovered than all the memories, and this is thought

to be related to the strengthening of memories over time through recall and processing. So it's particularly common that if there's some sort of traumatic event leading to amnesia, events immediately preceding the event and immediately following the event are least likely to be recalled because their processing and encoding and consolidation has been most disrupted. There's also another condition called Korsakov syndrome, which is caused by lack of vitamin B1 in the brain. And it's mostly related to chronic alcohol abuse and also severe malnutrition. And it leads to a number of memory effects, including, I think, anterior-grade amnesia,

inability to form new memories. And also, it's associated with a number of types of confabulation. So people tend to report things that are manifestly false. Like, for example, they're able to form new procedural memories or implicit memories of various sorts. But then if you ask them how they knew that or where they got that, then they just make up things essentially. This is called confabulation. It's not conscious. They think it's true. I talked about this when I discussed the introspection illusion, the notion of confabulation, but Korsakov Syndrome seems to be particularly associated with that.

Okay, that's all I wanted to talk about for this episode. Hope that it was interesting. If you liked the podcast, please jump onto the Facebook page. Just go to Facebook, type in The Science of Everything podcast and give our page a like. Also, you can go to iTunes and post a review. I much appreciate that because reviews help to increase the visibility of the podcast. If you would like to ask specific questions, things that you'd like to know more about, or a topic you'd like to see in a future episode or whatever, send me an email. My address is Fods12 at gmail.com, F-O-D-S-1-2 at gmail.com. Even if you don't have a specific question, I'd just like to hear generic feedback,

let me know your listening, criticisms, anything like that. Love to hear from my listeners. So, thanks for listening, and I'll talk to you next time. Thank you.