semi-spontaneous silence as time we should get started. Welcome back to week nine everybody. This is the second penultimate lecture of content.

So this week and next week before we're then on to the revision week. For those of you who'd like to join us on Mentimeter, please do so.

A couple of housekeeping things first.

Today we're going to be focusing on the development of memory. So we're going to look at how we can measure memory in infancy and childhood and what kind of evidence we have for the structure and existence of memory in infancy and childhood.

As I say this is the penultimate class so we are doing memory development today and then that links nicely to what we'll be doing next week which is looking at metacognition.

First of all let's check in there.

How are we all feeling this week? All right yeah I'm definitely a seven as well so please bear with me today.

A good mix. I hope the person that's feeling number four is okay and everybody else as well and congratulations to you who are feeling more number nine. That's weird. It's good to see a mixture of different vibes in the class.

Okay a couple of housekeeping things. So for those of you who are level three you'll have seen at your grades your coursework are now available. Your MSc coursework grades will be released shortly hopefully later today if not first thing tomorrow. For all students we have feedback drop-in sessions this Thursday the 14th of March so coming up and that's in either Josephine's office or my office between 11 and 12 and three and four. We'll be there ready and waiting.

I've put the two office numbers there so Josephine's in scrimgier 2.30 35.

and I'm in scrimgier 2. We have both of us so all the grading was split between the two of us and obviously there was moderation of things between us but what you can do when you're looking at your grade is check who actually marks yours and then that's the person to go and speak to if you have any queries or questions clarifications that you'd like about your grade and we're very happy to see and talk to anybody so please do feel free to come in but saying that will not be asking or answering questions in class because that's what Thursday is dedicated for. Okay we will leave that open there for when we come back later and we will get going with today's class.

So the aims of today are to look at first of all how memory is structured, how it can be defined and looked at in a childhood.

We're then going to talk about how memory has been assessed in infancy and in older childhood and into adulthood so how can we model memory and how can we measure it. We're then going to explore memory or specifically working memory as a type of executive function seeing whether it is a type of executive function or whether memory is something different from more general executive functions and we'll also be looking at how working memory abilities may relate to other cognitive skills such as IQ.

So let's start then by talking about the structure of memory and some of the key models that have been proposed to describe memory as well as the distinctions between different types of components of memory and particularly as I say focusing on working memory.

We're going to start this section by covering some key ideas that have been around for quite a long time now in regards to memory and the importance of memory. Now I'm not going to give too much detail on these they're just kind of an overview into why memory is important to study hopefully we all have a kind of common sense feel for that but just to put that in more concrete terms.

You can find that much more detail by looking at papers but also chapter seven in the Thornton textbook the modules textbook that talks about the development of memory and the opening paragraph to these chapter seven in Thornton's textbook states this fundamental to all intelligence is the ability to learn to take in information to store it and retrieve it in other words to remember it memory plays a key role in everything we do from the deliberate learning we put in before an exam to the casual recall of things that happen to us in our daily lives without memory no learning or knowledge would be possible everything we encounter would seem new and unique and unconnected with anything else nothing could even become familiar we would know nothing about the world not even about ourselves and this quote really nicely captures the critical role sorry that memory plays in our daily lives and what we're interested in let's see if that stays okay what we're interested in then is how does memory inform how we change across development what role is memory playing in infancy in childhood adolescence and into elder adulthood what effects can changes in memory have on an individual the term memory itself has been applied to lots of things throughout history psychologically speaking people use the term memory as an umbrella term from quite early on so that is they might be aware that memory wasn't just one specific thing there's lots of different components of memory but they referred to it using just an overall title of memory back in 1949 though a researcher named hev started trying to capture the distinctions that may be present within memory itself he proposed two differences or two components of memory and they were long-term memory and short-term memory hev argued that those are different so long and short-term memory are different very conceptually in terms of what they actually are as well as on a physiological or neurological level so he argued that the brain structures underlying short-term memory and long-term memory were very different a little later on other researchers called Atkinson and Schifrin further this suggestion by arguing that not only does memory consist of these two different abilities but that there's also a particular relationship between these two capacities between the two different types of long and short-term memory and that is how they function efficiently they have to have this specific relationship between them specifically they argued that you must have short-term memory is necessary in order for information to enter long-term memory so in their view short-term memory is acting as a sort of rehearsal chamber allowing information to enter long-term memory storage whenever we learn something new we first learn it according to Atkinson and Schifrin in a short-term manner before it then becomes part of our long-term memory imprint both of these definite definitions lead themselves to a two systems model where memory consists of two main components sorry I hope the wire is loose one second you still hear me okay um two systems models so according to two system module the short and long-term memory are intrinsically related you have to have one to have the other and specifically you have to have a short-term memory process in order to enter information into long-term memory if this two systems structure is indeed reflective of how memory is structured then we should be able to gather evidence for this using neuropsychological patients that supports this suggestion so for example if we find somebody that has a brain injury such as following a stroke and they have problems with their short-term memory as a result of this brain injury then it should by default according to this this model it should mean if they don't have short-term memory they cannot form any long-term memories they should not be able to they should not be able to have a problem with their short-term memory uh if they also have intact long-term memories in contrast if an individual has problems with their long-term sorry I should put this up if an individual has problems with their long-term memory then according to the two systems model this is um they should still have short-term memory abilities so having problems with your long-term memory according to two systems model does not mean that you wouldn't have a short-term memory that does work this table highlights some key studies that have tried to examine this question directly and they use neuropsychological patients patients that have had a stroke to assess whether this two systems model of memory is true and what we can see is that this is not the case so if we highlight this study here where a patient is observed have impaired short-term memory shall I send one to reported neuropsychological patients who experience just this so they have problems with their short-term memory but they retained intact long-term memory and other cognitive functions suggesting that something is wrong with this model this model is not accurately describing how memory is structured in the case of the shall have some warrant in study despite being in possession of this um uh or despite not being in possession of this so-called rehearsal chamber the short-term memory these individuals were still able to form long-term memory traces so therefore something else may be helping them or must be helping them to form the long-term memory on the other hand in the last column badly and warrington reported patients who presented problems with their long-term memory whilst retaining intact short-term capacities so this is in line with the model they have okay intact short-term memory but they aren't able to um form long-term memories so this would support the two systems model but the full story so these findings show that unlike the two system model would predict long-term memory is relatively unaffected by changes in short-term memory or short-term memory impairments indicating that the idea of short-term memory is just being the rehearsal chamber in the two system model is far too simplistic for capturing what memory actually is it's not capturing the true underlying structure of memory this has led researchers to try and identify what further distinctions could possibly be identified within short-term memory to try and more accurately capture how memory is structured and the different roles that components of memory may play and following this one of the most influential models to have been or to have come out about memory research in the literature was proposed back in the 70s by badly and hitch uh in the um in their model of memory if you're familiar with badly and hitch's memory model which i very briefly showed you last week this is not the diagram that i showed you and is not the one that is traditionally seen in memory research today i've put this in as this is the first model they use to try and define memory and we're going to move on to their more updated one in just a bit.

Badly and hitch's argument was that within short-term memory there are different abilities associated with different types of tasks that is there are a number of sub-components which are each ascribed a specific and different function and they identified three core components the central executive which is primarily associated with the frontal lobes and the central executive's job is to control both behavior and attention so if you think back to last week we talked about executives being the kind of the ceo or in charge of things and that applies here as well so in this context the term executive is being used in a similar way to executive function where we're talking about the job being to manage and control other aspects of memory and other aspects of short-term memory there's then the visual logical loop and these are identified as further sub-components of memory that are used to manipulate different types of information the visual spatial sketch pad is primarily associated with the right hemisphere and it's a store for rehearsing memory of a specific type and in particular that is the processing of visual and spatial information this links nicely to names in a visual spatial sketch pad that kind of visual the diagrams the images that we might see when we're trying to remember them but this can also refer to things like kinesiology information such as where my body is in space how positioned I am if I need to raise my hand and I want to try and remember a particular gesture that I made or something so your body in space and time as well the phonological loop is suggested to be associated with two distinct sub-components which is the rehearsal ability and a storage ability these are intrinsically linked as they both deal with auditory information and they do not deal with visual information so the visual spatial sketch pad is the visual information whereas arguably the phonological loop is that more audio information as of any theory put forward in literature once this was published people were very quick to point out that there were problems with this model suggesting that other things needed to be accounted for and to be encompassed within the model to help actually understand what memory looks like and actually badly and hitched themselves fully recognised this and a few decades after they first proposed this model of memory they when they were writing a foreword for somebody else's book about memory they provided some reflections on their own model based on research that had happened since they first proposed it since they first published it and one of their notes was that they themselves didn't think that their description of essential executive was detailed enough in fact they stated it was underdeveloped and that they need to further our understanding of what this specific element of short-term memory may involve they also felt that the function of the rehearsal for the phonological loop that is not necessarily the storage within that but the rehearsal bit specifically seems a bit too simplistic on reflection they need to explain they feel they need to explain that a bit more and in particular there are some types of information which seem just too complex to retain if you're just rehearsing something because it may be that you don't only need to remember the information but you might also need to remember the order that the information came in so if for example somebody reads out their telephone number to you you could remember all of those numbers but they won't do anything unless you remember them in the specific sequence that you have been told them that's only what that's when they become functional so in when we're trying to remember this information that we've heard auditorily we need to remember it not just factually but also the order they or factually in terms of the audio but also the order that it comes in so Badly and Hitch argued that the phonological loop as a concept seems a bit too simplistic to be able to recall this type of information where it needs to be exactly correct and in ordered Badly further highlighted some remaining outstanding questions about the structure of memory that he confirms are not answered by the memory model that he has proposed first he highlights that we do not yet know how it is that visual and verbal codes are able to be combined to form multifaceted representations in our long-term memory in other words a lot of our memories are not simply visual or simply auditory rather they combine both elements we have a memory that involves both visual and auditory information so the question becomes how is this information combined into one memory trace so to speak and how does this work in terms of moving information from short-term memory through to long-term memory the second question is how is it we're able to recall information that clearly exceeds the capacities of our verbal or spatial rehearsal for example it's been shown that there must be some kind of function of memory that is not long term because the information is later forgotten and yet when we're engaging in this rehearsal component of short-term memory it allows us to remember a lot more information than would be predicted if we're just using visual or phonological stores as an example of what i mean here Badly and Wilson report a study looking at two patients who suffer from amnesia they were asked to remember to read through and try and remember a long passage of prose or writing the patients were documented as having severely impaired long-term memory they could not record things later on in time forgetting they ever read the passage and saying i've not seen this before however when they were then presented with long prose passages they were able to remember certain bits of information they were able to recognize which passages they'd read versus which ones they hadn't read and and it seemed to have exceeded the predicted storage capacity of the phonological loop or the visual spatial sketchpad themselves so that is to say even if they were going over and trying to keep it in their short-term memory and it doesn't go to long-term memory more information was provided than they should be able to hold in mind if they were then asked to immediately report which one they looked at they were able to do so relatively accurately despite exceeding that capacity so Badly and Wilson argued that given that each component of the memory model is supposedly limited in capacity these findings present quite a big problem of this oversimplification of the memory model these patients were able to remember long passages in their short-term memory despite the opposing predictions of the model that they built so given these findings and they identified limitations to the original proposed model this led Badly and Hitch to update their initial model of memory in this updated model the key difference was the additional component of this episodic buffer in the middle here this is the figure i showed you in last week's class which i said we'd come back to today and it reflects the memory model that remains very influential even today so this is still being widely discussed and investigated in the literature it still isn't argued to cover everything that memory does but it is a a module that has stood the test of time so far episodic buffer is a chunking system and it's proposed to be the storage system that's responsible for binding information from different sources such as visual and auditory information and it does this by binding things into chunks called episodes and this Badly argues is what allows conscious awareness of what we remember we can see here in this model that the episodic buffer appears alongside the visual spatial sketch pad and the phonological loop Badly also highlights that in this model there are two different types of system they are fluid and the central executive components which are the visual spatial sketch pad the episodic buffer and the phonological loop are all identified by Badly as fluid structures which means that to do with short-term memory things are coming and going and are active we're using our short-term memory actively to rehearse things before it then either goes into long-term memory or gets forgotten in contrast crystallized structures with this title referring in this context to long-lasting nature of crystals so the long-lasting long-term memories that we might form and long-lasting memories are defined by memories that persist over time and are not necessarily always conscious but you can bring the information to mind when you need it one of the reasons this particular model of memory is important is because Badly and Hitch highlighted the idea that memory is not just passive it can involve more than just holding information in mind indeed sometimes we need to hold information in mind in order to apply it to a problem to use it to solve a particular solution or to deal with our environment for example if we were in a corridor even in this building in the scrimmage building and there was a power cut if it's always that we've gone through many many times we might be able to find our way out without the light whereas if it's a new thing we haven't formed those long-term memory traces yet we might have a problem so we use the information that we store in our short-term and long-term memory to apply it to problems and actively use it in this way and this is why Badly and Hitch's model is referred to as working memory because it's describing this active working memory component about how we use our memories to remember and retrieve information rather than just passively recollecting it so it's not something that passes us by we have to have an active role in rehearsing and keeping information in mind at any given point so the multifaceted structure of components is that they are each directly involved in the temporary storage and manipulation of information as I say we're kind of focusing here on this active type of memory sort of working memory and this is what the overall model is pointing to but it's this active component that differentiates this from other types of memories such as long-term traces.

Bayliss and colleagues further highlighted that when we're trying to understand the structure of memory we need to keep in mind that there are different ways to interact with information you're trying to remember a key distinction for example is between processing of information versus simply storing information processing information adds this active component of manipulating the information in some way trying to engage a bit in a bit more depth and then using it in whatever way is necessary in contrast storing information simply means holding it in mind to remember things each of these two processes individually contribute to memory storage and processing are there for different parts of memory and there are different influences on on your memory abilities so how well you can recall something may depend on whether you're processing or storing that information one of the reasons for example that I was glad that you all said you're okay with the mentor meter questions is because that is where you actually have to actively recall things we've just talked about and use that to answer rather than it being just me I know it's a majority of just me talking at you but trying to store that so we're doing two different things keeping memory in mind more actively as well as just kind of hoping it goes in and stays in okay it's been argued that what makes working memory different from short-term memory is that whilst short-term memory refers to your ability to store information in mind working memory additionally involves this attentional processing mechanism in this way arguably short-term memory becomes a subset of working memory according to the definition of these concepts as we've said another way of differentiating this is the notion that working memory involves not only holding information in mind but also a manipulation of that information in some way and badly himself also argues for this suggesting that working memory is different from short-term memory because of the need to manipulate information very briefly I'm also going to go over one other type of distinction in memory just to ensure you're also aware of this proposed distinction and this is between declarative and non-declarative memory declarative memory refers to the ability to consciously recollect things usually facts or information and this is a conscious recall of information and it's often argued is representational in other words it's about your ability to represent information or or about the memories that represent something about the world declarative memories can be either true or they can be false so you could falsely remember something as a fact an example of this might be knowing truthfully that London is the capital of England but you could also be extremely confident that and say that York is the capital of England which would be a declarative memory even though it's incorrect in contrast to this non-declarative memory refers to memory processes that are non-conscious and are expressed in performance rather than explicit reporting of recollection and they don't necessarily have to be true or false in these particular scenarios another term you might see to refer to non-declarative memory is procedural memory this type of memory is not to do with recalling specific pieces of information or facts so much as it is to do with performing different types of skills so for example if you learned as a child or as an adult to ride a bike but then you don't ride a bike for a very long time when you get back on that bike you might have a procedural memory that very quickly allows you to pick up that skill again even if you don't remember exactly what you're meant to do the idea that it doesn't necessarily have to be true or false is referring to how we can all do things in different ways when it comes to procedural outcomes so people might ride bikes in different ways or for example in this image from Friends Rachel and Phoebe are both going for a jog they're doing it in very different ways but technically they have the same goal outcome they have different procedural approaches to achieve the same outcome so there's this differentiation between declarative and non-declarative memory types further evidence to support this particular distinction has come from case studies of patients such as patient HM so following a bike accident when patient HM was around seven years old um HM developed a very severe case of epilepsy by the time he was 27 years old he had become so incapacitated in his day-to-day life due to the epileptic seizures that he was unable to engage in his daily life including not being able to work to be left alone for any period of time and just to conduct himself independently and as a result of this it was a decision was made that at 27 years old he would undergo brain surgery to try and stop the seizures where they removed significant sections of his brain that the seizures were occurring unfortunately following the brain surgery patient HM suffered from severe anterior grade amnesia which essentially means he was unable to form any new memories for example he would underestimate his own age he would immediately forget the names of people as soon as he was introduced and he was described his own state as like waking from a dream where every day and alone is itself every day is alone in itself this type of amnesia contrasts to retrograde amnesia which refers to the ability to or inability to recall events prior to the injury his memory of uh events that happened prior to the operation remained relatively intact he just had this difficulty in forming any new memories further intriguingly patient HM's deficits only appeared to apply to declarative semantic knowledge and not his non-declarative knowledge in other words HM was able to learn new skills as long as they did not require him to acquire semantic knowledge and I have an example here this so HM was administered something called the mirror drawing task in this task participants are asked to trace a shape but they can only see the shape and their hand in a mirror so in this image here this is the mirror up here and there's a piece of paper with the shape that they need to trace on and then this block here so that he can't see directly to the piece of paper in front of him he needs to use guiding in the mirror in order to draw the shape it's going to actually be quite a difficult task and it is not something we're used to doing drawing in a different way and it often takes some time to get good at doing it working with patient patient HM Milner observed objectively so looking at how well he was tracing that star or whatever the image was HM did improve across successive trials even though he himself reported that each time he was doing this I've never done this task before it's a brand new task to me so he did not have a memory of performing the task previously this is particularly interesting as he was obviously acquiring and learning the mirror drawing skill his his performance was improving every time he did that without he himself having any recollection of learning that skill or of engaging in that task.

Patient HM is no longer alive he died in 2008 at the age of 82 years old but he continued to be a willing participant in lots of psychological and neuropsychological studies across his life providing hugely invaluable insights into how memory may be structured and providing details on these types of separable components of memory capacities. The findings of HM's case study of these differentiations between declarative versus non-declarative memory components has been further replicated with another patient case study by Nikelian colleagues in 1988.

They also observed separable impairments seen across these two abilities the declarative and non-declarative and taken together these findings indicate therefore that these really are two distinct components of memory non-declarative memory can be intact even if declarative memory is impaired.

As a note though I should say I don't know of any studies that have found it the other way around so I'm not sure it would be possible to know what you're meant to do but not be improving in the skills I mean actually that is that is true that's learning I think but in terms of the actual different types of memory so it could be that you have to have the declarative the non-declarative memory in order to get the declarative but you might have to have declarative without getting non-declarative so they may be completely intertwined as well as does that make sense can't have it the other way around excellent okay let's pause to go through that first section historically short-term memory processes have been dissociated from long-term memory processes treating them as two separate components of memory and this led originally to the two systems model of memory studies and particularly insights from clinical patients have highlighted that this proposed two system model of memory is simply far too simplistic a particularly influential model of memory has since been proposed by badly and colleagues and in this memory in this model they suggest that short-term memory is distinct from working memory according to this proposal working memory involves separable active components that not only rehearse information but also manipulate the information other important distinctions to note in memory mean meaning that we need to keep them in mind when we're conducting research into memory so that we can be sure about what we're specifically looking at include differentiations between explicit and declarative and semantic memory processes versus those more implicit non-declarative and procedural memory processes okay let's go to um mentimeter sorry i'm trying to remember the structural things okay let's go to mentimeter to do some questions on the structure of memory okay i'll give you a couple more minutes if anyone wants to join in so first question from this section look putting our working memory to test a key difference between short-term memory and working memory is that short-term memory is defined as holding information whereas working memory is defined as holding and manipulating information or short-term memory is uh defined as declarative information working there is non-declarative information working memory is holding information short-term memory is holding and manipulating information and working memory is declarative information short-term memory is non-declarative information or is there no difference between those that's why that was a lot of words excellent okay yes so short-term memory is holding information so just observing it and not necessarily doing anything with it whereas working memory is both holding that in mind and having to manipulate it or maneuver it or use it in whatever we're trying to solve at that point.

Excellent. Next question. Patient HM suffered from retrograde amnesia, confabulation, anterograde amnesia, or Kortikov's syndrome.

anterograde amnesia, or Kortikov's syndrome. Which one did patient HM suffer from?

did patient HM suffer from? Excellent. Yes, so he suffered from anterograde amnesia, which is the inability to form memories post-injury or post-operation in his particular case.

Whereas retrograde would be if you have a difficulty remembering things from prior to the injury operation.

Excellent. Next question.

I don't quite understand how there's 35 people on the 40 people answering, but it's okay.

A declarative memory is memory for habits, memory for events and facts, memory for skills, or memory for conversation.

Which of those is the best definition of declarative memory? Excellent.

I think that's the first one we've got 100% on.

So congratulations, everybody. Well done. Yes, the declarative memory is memory for events and facts. The next question, I think, let's see if it will come through.

The processing spatial information, dealing with auditory information, controlling behavior and attention, or binding information into chunks. Excellent.

Yes, so the visual spatial sketchpad is to deal with that more visual or spatial information, what you're seeing and how your body is in the world. The phonological loop is for that auditory information. The central executive is that controlling behavior and attention.

But the episodic buffer is taking the information from visual and audio information or streams and binding it into chunks so that you can link those two memories.

those two memories. Excellent. Okay. This is not a quiz question, but before we move on to the next section, out of interest, how old are you in your earliest memory? Trying to make sure it's a real memory, not a memory of a photo or something. How old do you think you were when you first remember something?

something? There's no right or wrong answer to this one.

It's just interesting to see. Okay. There's quite a lot of spread, which is nice to see. Some people have some really early memories. Some people have some quite late on memories, but most people falling in that four to five-year-old group, which is pretty normal and on average is what kind of age people have their earliest memory from.

Awesome. Okay. I'll leave some question box up here if anyone wants to submit questions, but we are going to go back to memories.

So in the first part of today's class, we were going through a relatively brief overview of the different types of memory and memory structures that have been established, but what I should say is they've been established from studies in adults.

In the next section, we're going to be looking more at how we might study memory in infancy and see whether by studying memory in infancy, we can gain insights into how it's structured during early development and into later childhood.

Now, this cartoon I just included because it kind of made me chuckle, but also it's kind of really important to think about when we say we want to understand memory in infancy.

So the tagline at the bottom is cartoon says, I'd love to catch up on old times. It's just we don't have any. And this is kind of true for many reasons, including infants do not necessarily have as many experiences to form memories of.

But also more formally, studying memory in infancy can be challenging and there are a number of reasons for this. One of them is simply that infants have very limited world experience and thus they don't have memories that many memories to hold and for us to assess. We need to therefore keep that in mind when we're trying to look at and assess memory in infancy. More concretely, Hayne highlights two issues with trying to gain an understanding of the development and structure of memory across infants.

Firstly, there's the fact that very young infants do not yet possess language abilities, meaning we can't simply say what do you or don't you remember. They can't answer that directly and therefore we need to find alternative ways of assessing whether they do possess memory.

If you've ever spoken with a young child or a toddler who may have started to develop language, but you asked them a question about something, there's often also a lot of imagination and kind of making up some stories, which is great as a part of development, but it also means even when we do have those verbal abilities, we can't trust them to be reflecting accurately on previous memories. A second challenge is the potential for something called childhood amnesia.

The term childhood amnesia was first defined by Freud, and this experience or this occurrence refers to defining that most people, and again not all people as we saw, but most people struggle to remember events from very early on in their childhood with any degrees of accuracy.

degrees of accuracy. Indeed, research has shown that most people don't remember memories from prior to their third or fourth birthday or around about those ages, so for the average person the earliest memory we're able to accurately recall is from around the age of three, and this is what the phenomenon is referred to as childhood amnesia. Why don't we remember those first three years?

those first three years? Different researchers have argued that this experience of childhood amnesia may be due to a number of different things.

It could be related to developmental changes such as the acquisition of a self-concept where there are or there is a change in how we perceive oneself and other people, and that taps into the kind of theory of mind changes as well that we've seen and that you've spoken about with Josephine.

It could be related to the emergence of qualitatively change our understanding of memory and our own minds and therefore how we process things, so maybe we can form maybe a stronger memory trace because we have language to describe the event that's just happened and that we've just experienced, or it may be due to the presence of neurological changes.

Flow maturation of the brain means that memory storage is perhaps just not developed enough until around three years old to actually hold on to information.

There have been although the definitive answer is not as yet known. What we do know is that it's present for the majority of individuals and therefore seems to be a robust phenomenon. This information means that relying on things such as verbal reports of memory from children and infants even once they have verbal abilities would mean that we were unable to test or at least we would be unable to test accurately what memory abilities young children under the age of three years old are in possession of. If we waited until they could actually tell us what they remember we would miss out on those first few years.

Alternatively it would mean that we can only start to research as well research memory abilities in older children when we can trust that they know how to tell the truth and that they're not just going to be answering using imagination and that we know for sure they have some type of memory structure present. Again this would mean that we are missing the emergence and development of memory before it gets to that more fully developed stage and so that's not very helpful if we want to understand the developmental trajectory of memory abilities.

Thankfully there are other ways and other studies that have been designed to remove the verbal element required in most memory tasks and it's these types of tasks that we can look at to see whether we can assess memory abilities in infancy. These non-verbal tasks are what we're going to focus on in this next section including operant conditioning tasks, visual recognition tasks, imitation tasks and event related potentials which we'll come back to a little later on. So let's have a look at what is involved in these types of methodologies and how they can therefore provide us insight into whether infants have memory or not.

So we'll start with the operant conditioning tasks. This is a similar condition to a more classic conditioning tasks and you may recall having seen some of these examples before if you're a level three student. In an operant conditioning type of task a scenario is created in which an infant is conditioned or taught to associate a particular behavior that they produce with a particular outcome as a response.

For instance let's take this example here. So if an infant is provided with a bottle top like this one and if they suck rapidly on the bottle top a light starts flashing above them for them to look at. This seems to be something that infants enjoy again if you've been around children at Christmas time with the fairy lights and things that seems quite rewarding stimulus for them. So you condition a child to know that when they're sucking on the bottle the light starts flashing and it gives them this reward-based feedback. Once the infant has been familiarized or conditioned to this relationship memory in this scenario is then assessed by seeing how long the infant will persist with the sucking of the bottle top when the light stops coming on in response to their action. In other words we're measuring how long it takes for the conditioned response to disappear or to return to baseline.

The reasoning here is that this length of time how long it takes them to stop sucking on the bottle top and return to baseline may tell us something about the infant's expectation of a particular outcome.

If the infant is expecting the light to start flashing the conditioned outcome this would indicate that they have a memory of this habit happening previously.

To expect something to happen again to repeat you must remember that it's happened before. So according to this argument in a task like this the infant should therefore only continue with the sucking response if they were presented with a bottle top. If we change the object to be a rattle or another toy that they haven't seen they should not have that conditioned response and they should not increase the sucking response to a new object as there's no association between that object and the light flashing. Another example of this type of operant condition is studies that use mobile conjugate reinforcement as illustrated in this image.

So the motivation here is once again that the child learns to associate performance of a particular action that they are producing with something in the environment that responds to in a rewarding or interesting manner when that action is performed and this association is created through conditioning.

When the association is then removed as in the previous example how long the child perseveres with the initial behavioural action in the absence of that stimulus feedback response is then recorded.

In this particular version of an infant's foot and the other end of that string is linked to a mobile above their head which they can see.

When the infant then kicks the toy mobile moves or spins around and has this rewarding feedback.

After this conditioning has been achieved the researcher then removes the string.

Infants are assessed both immediately after the string removal to see how much more they keep kicking their foot to try and make the mobile move but also they are tested after a few days weeks and even months after this initial training or conditioning period.

At each of these time points it's measured how often the infant kicks in response to seeing the toy mobile above them in comparison to a preconditioned baseline.

It's argued that if an infant is found to persevere in their response more than they would at baseline or when a new toy is presented and without the preconditioned association this would indicate that they're remembering the particular toy mobile that they've seen above them. That is they've retained a memory of the first time they saw the toy in which there was this reinforcement and so measures are focused on whether a long-lasting memory has been created as indicated by persevering with clicking behaviours rather than assessing relearning at each time point.

So I could only find a really vintage video of this particular task but I think it's useful to see it in action so I will play this for you. So recent experiments with three month olds have revealed a surprisingly complex mind already at work.

I think that most people who work with human adults think that babies are like rats until they're eight or nine months of age and I think that what we've shown is that babies are very human for the moment they're born. Infancy researcher Carolyn Rovie Collier and assistant Peggy Borson make a house call. John B.

Watson's work helped shape the attitudes of an earlier generation. This new research helps document a new vision of the infant. We're going to be teaching you a game Martin.

Yes he says well it's one of these things you know you can lead a baby to a mobile but you can't keep going to play.

Martin's foot is attached by a ribbon to a separate stand not the mobile itself.

Peggy counts his kids.

This gives a comparison for the experiment which follows.

Peggy connects the ribbon to the mobile. Now Martin can see and hear the consequences of his action. His rate of kicking doubles. The infant is now an active player in his world. But that's only one part of what these experiments really. Next it was on to visit Michael's.

One of a pair of twins whose parents stay around for Rovie Collier's explanations.

Michael explores the ribbon although it is not yet attached to the blocks. But still he gives it what Carolyn Rovie Collier calls the good old college tribe. Three-month-olds cannot crawl or walk and research has shown some parents, especially fathers, think infants only become people when they begin to speak. But sometimes actions speak louder than words. So what they're trying to show in that study is whether the infant is one, recognizing that moving their foot is having this direct impact on the mobile but also looking at whether they remember that at different time windows.

Opal and conditioning tasks are important and are widely used to allow these assessments of implicit forms of memory in very young infants. They provide a way of estimating whether an infant is able to remember something as long as the behaviour is utilized to assess this remembrance are ones that the infants can actually physically achieve such as the sucking or kicking or doing something. So when we were looking at imitation a few weeks ago or you were looking at it with Josephine in the workshop classes, we were talking about how if their actions are beyond the repertoire of an infant then it would be hard to actually assess these things. Okay I'm going to run a few minutes just to finish this section and then we'll take a break so just a few more slides in this bit. Another way of assessing memory in very young infants is the use of visual recognition tasks assessing things such as looking times.

Visual recognition tasks tend to utilize novelty preferences seen in young children and infants.

So to do this infants are habituated or familiarized with a particular object images of a particular object and once they've been habituated they are presented with a pair of stimuli one that's old and a new stimulus at the same time. The amount of time spent looking at each object is then recorded. If the infant shows a novelty preference looking more at the new image or the new object than the old boring one that they've seen before then this is taken to indicate that they must be remembering the original one so they have or they are in possession of a memory of that original one which is why it's now considered boring and they can distinguish it from the novel one. These visual recognition tasks are also very popular in infancy research and are used to assess whether infants are remembering things that they've seen or whether they are not. Imitation tasks are another way of trying to gain insight into infants and children's memory abilities and in imitation tasks a child is shown as a particular action usually a slightly unusual action that they may not have seen before they shouldn't have come across regularly before so for example they might be introduced to these two characters here and let's say the duck puppet rings a bell and the cow puppet picks up a rattle and shakes it.

Following this there is then a gap of different periods of time whether that's minutes hours or even days weeks and months later before a child is invited back into the research lab and at these later time points the child is then provided with the particular toys that they saw perform these actions at the earlier time point and it aims to see whether they can reproduce the target actions that were shown at the original time point so can they get the duck to I think I said ring a bell and the the cow to shake a rattle.

What researchers have found reliably is that results often vary with these types of imitation measures depending on how the task is presented and structured.

For example if when you're first showing the actions to the children you say this duck is ringing a bell this cow is shaking a rattle children have better long-term retention of those actions at those later time points compared to children who observe the same actions but they don't have that verbal commentary alongside it so if you're providing that verbal commentary then even six weeks after the original event we see improved performance for those who had that commentary versus who didn't.

Finally the last measure I'm going to talk about in this section is the use of and this kind of complements nicely the looking times of imitation researchers have also used event-related potentials.

ERPs, event-related potentials, measure an infant's physiological changes in response to different types of stimuli.

ERPs are used to measure patterns of electrical activity in the brain that are constantly happening so they use measured using these electrodes such as in this image here they're entirely harmless to infants adults or whoever's across the brain which tells us things like which areas of the brain we're seeing have a particular spike of activity the speed in which the brain is responding to different things different types of stimulus and the characteristics of the brain waves that are generated by the different stimuli.

So in this way you can use ERPs to assess whether at an implicit level when showing an infant something they've seen before do we see a different pattern of brain activity compared to a novel object and that would tell us on this implicit level whether whether there is this memory trace for individuals for infants.

As we've seen then there are a lot of different tasks and different approaches to studying memory in young infants this table provides an overview of different tasks and the ages of which children on average pass these tasks of memory using these measures.

A note of caution as always is that these tasks use implicit measures of memory essentially because that's what's required if we want to study these earlier ages these infants as we mentioned earlier.

What this does mean though is that we again have to be very careful in our interpretation and how much we can say that these are telling us exactly what we think they're telling us. Whenever we have an implicit measure of something or an indirect measure of something and we can't get direct access to what the infants are actually thinking we have to make sure that we are cautious in our interpretation of results. It could be that we're seeing evidence of memory it could be that we're seeing evidence of something else. I'm not going to go into detail on all of these different tasks here but just to flag those of relevance if you're looking into this.

Okay I think let's pause there for a 10-minute break and we will come back in a second for the second half.

Okay let's carry on with this section so having talked about different ways we might be able to measure memory and infancy let's move on to looking at some of the observed changes in memory abilities within infancy.

Hain has argued that from the data we do have we in regards to infant memory development there are some things that we can reasonably conclude. Throughout infancy for example we can observe improvements in an infant's memory specifically memory becomes faster and easier for infants to create so less time is taken to form a memory trace.

With an advancing age infants are able to keep these memories in mind for longer periods of time recording them from further and further time points and they can do this with less specific cues so they are becoming better at remembering things separate from the initial memory trace formation.

For example they might be able to recall how to solve a particular puzzle if they're presented on one that's very similar but not exactly the same as the one that they've seen before.

So it seems then that developments in memory across early infancy may reflect improved efficiency of processing memory traces.

Adele Diamond who we've mentioned a few times over the past few weeks argues that there must be some form of working memory present during the first year of life.

In previous lectures we've talked about how working memory is suggested by Diamond particularly in terms of the A0B search task. In this task Diamond sites it must involve working memory because at least to some degree and working memory must therefore be present in young infants when they start passing this task at around eight months old. And again this is because Diamond argues that the A0B search task is an assessment of both inhibitory control and working memory abilities where infants require some sort of memory trace of both the object that they're looking for and also the current location of that object if they're going to know where to correctly reach for a particular object.

We know that infants begin to pass the A0B search task from around eight months old if the gap between the hiding event and the search opportunity is very small.

Results from a number of studies including those from Adele Diamond have shown that between eight to twelve months old improvements in the task can be seen where infants are able to keep that memory of where the current location of the object is for longer and longer periods of time before they make A0B search errors. So that is the gap between witnessing the hiding event and the opportunity to search can get longer as they get older between eight to twelve months old.

And this evidence from the A0B task suggests that infant's development with age matches the predictions made by Hayne which is that over time infants can hold the trace of where a particular toy is hidden in their minds for longer periods of time longer gaps between the hiding and searching events.

A more recent paper on the A0B search task has been conducted by Halford and colleagues.

In this paper they argue that infants are doing the task in the, sorry, infants when they're doing when they're doing the A0B search task are using their working memory to remember both what the object that they're looking for is so they're motivated but also where it's currently located in space. The infant therefore needs to realize that the object itself is not tied to any particular location or particular position in space other than where it has been put last.

Their memory traces of the new location need to be strong enough to overcome the previously habituated or familiarised search location of location A. To do this they need to remember where an object has been hidden in amongst potentially various hiding locations it could have been placed. So we talked about A0B in two locations but you can make this more difficult by having multiple locations and they need to form a link between the object that they're looking for and the hiding place that they observed and this is a form of dynamic binding. To do this Halford and colleagues argue, infants need to use working memory abilities where they're actively manipulating the information to keep in mind where the object is and where it's currently located in space even when out of sight and this they suggest is why performance on the A0B task changes across infancy as working memory abilities gradually strengthen and develops. They get better and better at using this dynamic binding and remembering things across longer periods of time. It does seem therefore that there is some form of working memory present in the first year of life according to results from studies such as the A0B task, operant conditioning tasks and visual recognition tasks.

These results match research by Carla Humber regarding development of inhibitory control abilities which we discussed last week where she also argues that there's at least some form of inhibitory control present in the first year of life.

I mentioned at the start of today how important memory abilities are for daily life and thus our ability to understand the development of memory and factors that may affect it are also really important and this is further illustrated by studies that have found that maternal reports of an infant's temperament at eight months old along with EEG data recorded when the eight-month-old infants complete the working memory task alongside an inhibitory control task positively predict the child's working memory performance at the age of four and a half years old.

So that is to say how well an eight-month-old or how well the EEG signals pick up on working memory in an eight months old brainwaves predicts how well they'll be able to perform on working memory tasks at four and a half years old.

So even these early infant abilities are predicting what a young child is going to be acting like how they're going to be utilizing their working memory abilities at later ages and this really highlights that it's worthwhile trying to further our understanding of these capacities at early infancy or during early infancy because if there is a problem we can maybe intervene at earlier stages to try and help improve those abilities as we get to older ages.

Again though something to note because it's always important to have a note of caution with these and Karan and colleagues highlight that much like when we talked about inhibitory control last week with working memory tasks we assess working memory differently in infants as opposed to four-year-olds and later childhood and into adolescence and so the evidence in that EEG study for example eight-month-olds were using EEG at four and a half years old were using more explicit measure of working memory capacities it could be that we're actually measuring two different things that happen to correlate and so we really need to make sure we're assessing things in as matched way as possible to make sure we can draw those links comprehensively between the two events. This doesn't mean that the implicit measures in young infancy are worthless they're not they're may or may not be possessed at these early ages but we do need to once again be cautious in applying our findings from one young age group and using it to predict older ages and and vice versus.

Okay so assessing memory abilities in young infants can be challenging and we need to be careful in our interpretation of implicit measures however there is enough conversion evidence to suggest that there is some form of memory abilities in children as young as eight months old or at least within that first year of life.

In particular it's been suggested there is a form of working memory present from around the second half of the first year or eight months onwards and it's important and interesting to note that assessments of working memory at this young age seem to be able to predict memory performance later in life. Working memory abilities recorded around the age of eight months old using EEG can predict working memory abilities in early childhood suggesting these capacities may be relatively stable although as I said a second ago we have to be cautious in our interpretation of these links and these findings. Okay so that was a section on memory in infancy in the next section we're going to move on to looking at memory abilities in childhood.

Given everything we've talked about so far today it seems that there is relatively substantial evidence that there's some form of working memory within that first year of life and this can be assessed by various different tasks including implicit measures EEG and the A0B search task. Further according to work by Wolf and colleagues it seems that performance on these implicit measures in infancy is meaningful it can predict abilities in later life and if we take all the data from infancy together it indicates that yes working memory is present in these early ages but we're not clear then in what form working memory and memory abilities are taking in childhood.

memory and memory abilities are taking in childhood. So what about older children does anything change between infancy and childhood or do we are we born with memory abilities that pretty much look like what we have in childhood and adulthood?

and adulthood? Whilst we may be able to identify precursors to working memory and infancy these insights are not enough for us to determine whether working memory is present in its adult form and structure in early childhood.

For that we need to look directly at older children as a starting point to see how their memory functions. I mentioned Aloe and Gavricol last week when I was talking about memory and executive functions and they are leading researchers in the field of memory and particularly in childhood memory and they conducted a study to investigate the structure of working memory in preschool age children.

They used two types of factor analysis to examine children's performance across a number of different memory tasks.

I briefly touched on factor analysis in last week's lecture and factor analysis is a statistics technique in which you take an individual scores from across an array of different tasks and you then see if their performance can be explained by different models or different factors and these models are used to describe different abilities or capabilities depending on the model that you're building.

So in Aloe and Colleague studies they use both exploratory and confirmatory factor analysis in their approach to the data. Exploratory factor analysis is a bottom-up data-driven modelling approach and what that means is that after running a set of tasks or questionnaires whatever measures it may be that you're using they then look at the data and ask a statistics program to draw out patterns and relationships that seem to exist in the data. The researcher then tries to determine what these patterns may be indicating and which variables they may be capturing.

In other words based on what the data looks like you extract likely candidate variables from the data and say that most of the variance seems to be explained by this which you then label as variable x. This exploratory group approach can be really really good in some ways because it's data driven meaning you are not just coming with a lot of preconceived theories of how the data should look or what you think the results should show you're literally just looking at what does the data show me and how can I interpret that.

The problem is however it can be difficult when exploring when using this approach to know with any certainty what the actual variable being measured is in your data set specifically because it's always being built and described after the fact.

So when using an exploratory factor analysis we're not doing that classic scientific process of having a theory, building a hypothesis and then testing that hypothesis which is what most of you will do in your dissertations. Instead we're starting with the data and extracting something from it. There are strengths and weaknesses to both but I just wanted to kind of highlight that what they're trying to do is let the data speak and see whether it fits their theory as opposed to and analyzing things based on a theory.

Confirmatory factor analysis in contrast is when you do a very similar technique of doing lots of tasks lots of questionnaires or measures and then you're looking at the relationships in the data but instead of this time letting the data speak you have a predefined model based on previous theories and based on your predictions and you then ask the stats program to tell you whether the data fits your predefined theory.

How closely does the data fit in line with this? And this is what Mayaki and colleagues did in their study investigating the structure of x-electric functions in adults as we looked at last week and I showed you this figure then.

So here they knew what they thought. They thought we had three core executive functions. They did a number of different executive function tasks and then they asked their stats program using confirmatory factor analysis does the data fit this and they found that it did where different tasks loaded onto three different components the shifting updating and inhibition so cognitive flexibility working memory and inhibitory control and they found that using this analysis they could confirm that there were these three main aspects of executive function.

So in contrast to exploratory factor analysis confirmatory factor analysis is when you are fitting a predefined model to data and you're trying to confirm that the model is correct using the data you've collected.

Okay so back to Aloe and colleagues in their study. In their study they aim to use the two different types of factor analysis exploratory and confirmatory to investigate the structure of working memory in children. Aside from the usual components of working memory defined in prior research so that's the visual spatial sketch pad the episodic buffer and the phonological loop they proposed a fourth component which they referred to as phonological awareness. They argue that the phonological loop as previously established is an important part of working memory but phonological awareness is a different scale entirely.

In their view the phonological loop is involved in short-term storage and rehearsal of auditory information whereas phonological awareness is the is involved in the ability to encode access and manipulate the sound units of language which are specifically called phonemes.

So a phoneme which you may have heard in your language lectures is the smallest unit of sound that distinguishes one word from another in language such as the k sound in car or k sound in car.

Aloe and colleagues suggested that the phonological loop and the phonological awareness components are related but they do not believe they are identical so they wanted to investigate in children whether the phonological loop was identical to the skill of having phonological awareness or whether they do indeed reflect separate capacities as they predicted. When doing factor analysis large samples are required to ensure there's enough power behind the analysis and in this study they had 635 children who took part between the ages of four and five years old.

Each of the children completed a battery of different memory tasks including recall tasks, span tasks, sentence repetition, phonological awareness tasks but also some non-verbal measures as well to make sure you're looking not just at the verbal aspect of these memory abilities.

They use this particular battery of tasks because they tap into different aspects of memory and so they should therefore each be related to different types of cognitive abilities.

Now again as a side note despite this being quite a comprehensive and good task battery it's noticeable that they don't have anything that directly measures the core limitation of their study but they may have had good reason even if it's just that they have a limited amount of time with the children so you had to select your tasks but that's quite a noticeable absence for the particular study design which we'll see when we look at their results.

Their aim was to use the outcome measures for each of these different tasks to test different models of memory in childhood using factor analysis.

In doing so results would speak as to whether these are truly separate skills as the model predicts or whether they're actually all just measuring the same thing so is it if I'm good at one then I'm exactly the same level of good on another one or can I be differentially good and are they measuring different things. So the question they were asking is are there different subtypes of memory in childhood or is memory just one overarching thing? So using confirmatory factor analysis to find out which data was the best or which model was the best fit for their data they produced a model which was in line with their expectations so the model showed the data to be clustered around five separate factors and these were the five factors that the authors had expected to find as the basis of the structure in memory of child or in child of memory in childhood.

The five factors that they identified were an overall global working memory factor within which there were three related but separable components so the phonological loop the central executive and the episodic buffer and they also had two other factors present in their data which the data suggested were not the same as working memory and this is as predicted the phonological awareness indicating this is indeed separate from the phonological loop and also non-verbal ability.

So what these results indicate and this is their figure is that there is a fit between what they wanted the task to measure and what the tasks actually seem to measure.

In other words they chose their task to assess whether these five different types of memory in the circular ones here factor one two three four and five exist in children age four and five years do they have five different types of memory.

The confirmatory factor analysis where you essentially tell the computer what you think the pattern of data is going to be the computer checks that this is true and then in this case the results indicated that this was true the data loaded onto these five factors as predicted and this suggests that this five factor model of memory in childhood is an accurate representation of the structure of memory in childhood. So in this figure the rectangular shapes are indicating the tasks that infants were that children were doing the four and five year old children were doing the circulars are the factors that are being proposed this would be the episodic buffer the phonological loop etc which we'll zoom in on in a second and then all the lines are telling us the strength of the relationship between each one.

Now again there's a lot going on here in this figure so we're going to zoom in on a little bit to the bits that we're actually interested in so we know from this that the tasks do load onto the different factors what we're really interested in is the actual factors. So these are the five factors that on the previous slide are labeled as factor one two three four and five but I've put in the actual words here because it helps by a grain.

Their figure is a bit overwhelming when I'm trying to process all the things are going on but as I say what we're interested in is that what we know from these particular results is that the data shows these five separate factors do indeed exist.

The authors argue that the data collected in their study is best illustrated then by a five-factor model.

The ovals in blue are highlighting the three different components of the working memory while the ovals in green are showing two separate but associated constructs.

The model matches on quite well with the idea that working memory has multiple components here it has a central executive the phonological loop and the episodic buffer and that these are present in young children. It also tells us that working memory is separate from but related to capacities such as non-verbal intelligence and phonological awareness as the authors predicted.

As a note remember that they did not include any specific measures of the visual spatial memory so visual spatial sketchpad and so if they had we might expect to see one more blue circle here with visual spatial sketchpad as part of the model they didn't measure it though so we don't have that as one of our factors here.

The main findings or takeaway messages from this study are that according to these results children's working memory is already quite adult-like in structure by the age of four years old.

They also establish that like with executive functions which we talked about last week the different components of memory examined in this study seem to be separate but related.

The results also highlighted that phonological awareness is not the same thing as the phonological loop which is what they predicted these two passages reflect different things.

The phonological loop acts as a short-term auditory store and rehearsal mechanism whereas the phonological awareness is referring to the ability to encode and manipulate the sound units of language so those individual sound the smallest unit of sound.

As I mentioned at the beginning of this study they used both confirmatory and exploratory analysis and both types of analysis they found that they had this five-factor model but the confirmatory factor analysis with a five nodes of memory that they had already outlined was the best fit of the data so this is the one that they say is the model best fitting model of memory in children age four and five years old. Okay so moving on to a different study although this one is also by Aloe and different colleagues this time or Aloe and Gabbo I should say and Pickering this time.

So in this study they again wanted to examine working memory looking at whether there are different types of working memory resources for different types of information.

They suggested that working memory could work or be structured in different ways.

The first option could be that there is general working memory ability which gets called upon to help remember information regardless of what type of information that is so it could be verbal it could be visual spatial and either way there's one resource that's utilised and used for both of those. The second option is that there are distinct working memory resources for acquiring verbal versus visual spatial information so that is on the one hand you have on the left hand side the general working memory resource a finite amount of working memory resources that can be applied to different things at the same time or there's a domain specific working memory where your ability to remember things that are verbal is separate from your ability to remember things that are visual or spatial and in this domain specific working memory model there are two distinct abilities.

Each of these models would lead to different predictions about what should happen in a task if they're a true reflection of working memory abilities. For example if working memory is a domain general process with both types of information using the same finite working memory resources then performance on tasks measuring each of these abilities should be strongly and highly related across verbal and visual spatial information. If however working memory resources are domain specific separate independently working from each other then performance on the verbal working memory tasks should not be as strongly related to performance on visual spatial working memory tasks as they're reflecting separable processes. So to assess which of these models is correct AlloA and colleagues used again a large sample of children 708 children this time between the ages of four and 11 years old and so what they were also looking at is how working memory may change and evolve across this period of time between four and 11 years old so looking at development of working memory.

Children again completed a task battery assessing both short-term and working memory again of both verbal and visual spatial processing and data was then analyzed to assess whether children's performance on these tasks was predicted by domain general or domain specific working memory abilities.

Confirmatory factor analysis was used then to assess which model best fit the data in this study and they found that according to the data the best fitting model was the one where children have a domain general working memory component with separable but associated short-term memory components for verbal information and visual spatial information.

In other words the results indicate that there is one domain general working memory resource used for the different tasks whether verbal or visual spatial information is being processed. So this is a figure illustrating their factor analysis and as part of their factor analysis Aloe and colleagues also wanted to look at this development of this working memory ability.

So they wanted to compare the fit of this three component model across different age groups within their sample to see whether this continued to be the best model for each of these age groups and does it describe the data best across all three ages. So again last week we were talking about executive functions and there seemed to be a unitary structure to a more separable structure from childhood to adulthood in executive functions and they wanted to see whether the same was true for memory abilities here but actually what they found when we were looking at memory abilities is that between children aged four to six seven to eight and nine to eleven this was the best fitting model across all of their different age groups and this suggests there are limited changes in memory structures between the ages of four to eleven years old.

So what does this actually tell us? Well the results indicate that the cognitive structure found in adults of the central executive and then the separate short-term memory stores for the verbal and visual spatial information seems to be present in the young children by the age of four years old.

This further confirms and again in line with other studies that adults and young children have the same structure for their work and memory.

What does differ however is the capacity and efficiency of these abilities rather than the actual structure.

So that is to say they also report that working memory has stronger links with short-term visual spatial tasks in younger children compared to older children which suggests that in younger ages they're spending more time having to manipulate the information and really keep in mind creating a stronger memory trace so there's more effort involved when you're trying to remember these things at younger ages compared older ages where we're more efficient at simply remembering things without having to manipulate that information.

If you're interested in this paper I recommend looking at it in more detail they find some really interesting outcomes and there's a lot of little things to kind of get your head around as always it's on the reference list if you're interested.

Okay the last study I'm going to talk about in this section is by Conkling and colleagues who argue that working memory as an executive function is strongly related to frontal lobes which differs from other parts of memory such as short-term memory which is more strongly linked to the hippocampus. As we covered last week when we were talking about the structure of executive functions and changes being from unitary to specialized in the three-factor model from childhood to adulthood we know that the frontal lobes and the prefrontal cortex continue to mature through our adolescence and right up into your third decade of life they're changing developing across this time span. This could mean then that given the structure supporting working memory is also maturing and becoming more efficient during this time that this could be reflected in changes in working memory abilities also across this time period including across adolescence and into adulthood. So these researchers argue there are different parts of the frontal lobes and the prefrontal cortex that support different types of activities and tasks.

For example the ventral lateral prefrontal cortex is involved in maintenance tasks only and develops relatively earlier on in adolescence. In contrast they argue that the dorsolateral prefrontal cortex is involved in maintenance and additionally processing tasks.

This arguably would indicate that tasks requiring the use of the dorsolateral prefrontal cortex are going to be more difficult to reach peak levels in and consequently the abilities that require this brain region may not develop until later in adolescence when we're actually reaching peak maturity of these particular brain regions. To investigate whether this was actually the case or not Conklin and Colley set out to assess whether different frontal lobe areas do indeed develop at different ages and how performance of this may actually vary sorry performance may change according to the age on different types of tasks of memory.

So they had a sample of, excuse me, 117 children and adolescents between the ages of 9 to 17 years old and each of the participants again completed a battery of working memory tasks to see how performance was going to change and be observed across adolescent years.

As they expected results showed that maintenance only tasks were found to be much easier for participants to complete.

They were more easily accomplished by younger adolescents and maintenance only tasks would involve things like being able to just read a list of numbers repeating them back as exactly as you heard them so no manipulation of the data that you're trying to remember is required and children quite young were able to do this extremely efficiently.

However in tasks that required maintenance and processing so tasks that involved remembering information and also manipulating it for example a sequence of numbers that you have to then order put in a particular order.

Participants were found to struggle with this much more until older ages this improved across adolescents suggesting that these are much more difficult tasks to do and it was something changing that allows children and adolescents to get better and better at this as they mature.

The study also tested recognition abilities that is tasks that did not require working memory they were just a test of whether a participant remembered something or not and results showed that performance on these recognition tasks did not change across adolescents with younger adolescents doing just as well as the older adolescents on these tasks indicating that this task was not assessing working memory abilities like the other tasks and as they would predict.

Paid together then these results indicate that working memory abilities specifically continue to improve across adolescents in line with the neurological maturation occurring in the frontal lobes at the same time.

Again this further supports the ideas of working memory as an executive function located in the frontal lobes working memory being distinct then from short-term and long-term memory. Okay let's pause here again to reflect on what we've talked about in this section. Using confirmatory factory analysis Badly and Hitch's model of working memory can be applied to children as young as four years old suggesting there is a similar structure at four years old as an adulthood of how working memory is structured.

The different components are related but separate which means we're not saying that they are completely individual things they are very much highly correlated but they seem to be individually traced so I can do I might be really good at remembering visual information but lesser audio information. There's evidence that working memory continues to develop throughout adolescence particularly in terms of capacity and efficiency so although working memory is present at earlier ages our ability to use those working memory abilities is becoming better in capacity and efficiency as we get older and this seems to be related to development of frontal lobes so frontal lobe regions are also changing across these time periods and that may be what's driving these changes in working memory. Okay let's switch over to Mentimeter for a few minutes and we have a few more quiz questions for this bit okay so thinking about what we've just gone through in the last couple of sections which of these has not been cited as a cause of childhood amnesia the development of a sense of self neurological development blocking of unacceptable impulses or development of language which of those has not been cited as a cause of childhood amnesia excellent yes the blocking of unacceptable impulses would be more about inhibitory control and things but both development of a sense of self or theory of mind those kind of too intertwined neurological developments of changes in the brain and development of language have all been suggested that they could help overcome childhood amnesia as we get older and that they may be able to explain while we experience childhood amnesia. The A not B search task indicates what that working memory is present in the first year of life that working memory does not develop until after the first year of life that inhibition is present in the first year of life or that working memory and inhibition are present in the first year of life.

Excellent again a slightly trick question because today we have been biased towards working memory and therefore we're talking about that but Adele Diamond we said this in a few classes now she argues that the A not B search task involves both inhibitory control I need to stop myself reaching to location A because the object's been moved to location B but also working memory because I need to remember that the object has been moved to location B and so trick question because of the priming of working memory but the true answer here is that both working memory and inhibitory control are indicated in the A not B task and again I put this question in we talked last week about making sure to read the question but also to highlight the trying to the links between some of these classes so although today we focused on memory and particularly working memory it's very much complements executive function so if I don't know just as a hint if you were answering a question of executive function you might have more than one lecture to refer to when you were trying to answer that. Okay next question which of these procedures has not been used to test memory development in infants operant conditioning gaze following habituation procedures or deferred imitation has not been used to test memory development in infants excellent yes so operant conditioning habituation procedures so habituation is getting them used to something so for example teaching them that moving their foot makes the mobile above the cock move and then once they've habituated to that removing that and seeing removing the connection and seeing if they continue doing the movement and deferred imitation is where for example you play with the puppet ring the bell shake the rattle and then you see over time whether they still can remember that and whether they can imitate that gaze following is not one that is used for memory gaze following is used as a test in infants but not for memory okay last question imitation tasks have indicated that infant memory can be enhanced by what using verbal cues using familiar objects using novel objects or they cannot be enhanced they just are what they are okay excellent yes so by saying i'm going to get this puppet to ring the bell or shake the rattle that seems to help improve performance even across different times okay let's have a look at the leaderboard changing excellent congratulations okay so any questions at the moment whether in sense so far okay i have a proposal for you all because it is almost 10 2 we have one more section to go through i propose that i upload a video or bonus material for you and we finish five minutes early does that sound reasonable to everyone okay it's first of all a question is explicit memory the same as declarative memory uh yes so explicit memory is something you can consciously bring to mind and you can comment on and you know it's a fact but it could be true or false so remembering something that isn't true like york is the capital of england so i could know it's a declarative fact it happens to be wrong but it's explicit knowledge because i know that i know the capital even though i happen to be wrong in that case so yes explicit um and then implicit would be that non-declarative or more procedural type of memory okay i put the this i'll put the slides up and i'll put the video up with the bonus section exam preparation uh the next week we'll be back talking about metacognition in childhood otherwise thank you all for your attention and i'll see you next week or on thursday for people that want to drop in please

Transcript paused due to scroll

Lecture 9: Development of Memory