Cognitive, LTM, WEEK 8 Flashcards
Why does LTM need a taxonomy?
- Taxonomy = classification of things
- We need a taxonomy because LTM is not just one thing, we have different kinds of LTM for different activities (procedural, semantic, episodic etc..)
Taxonomy of LTM
• LTM is first split into declarative (explicit memories) and nondeclarative (implicit memories) which have further sub-divisions
- Declarative (explicit) divides into facts and events
- Non-declarative (implicit) divides into procedural memory, priming, associative learning (classical cond) + non-associative learning (habituation + sensitisation)
• Semantic = world knowledge, Episodic = important events (e.g. birthdays)
• Procedural = how to do things (tie shoelaces), Conditioning = Pavlov’s dogs
Two key categories of LTM
- Non-declarative memory: ability to recall a movement sequence or how to perform some kind of action/behaviour
- This is unconscious memory: subjects can demonstrate knowledge, such as a skill, conditioned response, or recalling events on prompting, but cannot explicitly retrieve the information (e.g. say how they did something like ride their bicycle or tie their shoes)
- Declarative memory: refers to ability to recall what you know, be able to give details about time, place and circumstances of the event etc..
- This is conscious memory: subjects can retrieve an item and indicate that they know that the retrieved item is the correct item
- Explicit/implicit distinction is not very different to the declarative/non-declarative distinction > d/n-d are most commonly used > these kinds of memory can be referred to in different ways (see table 14-1) > different terms but refer to the same thing
- Main distinction between declarative + non-declarative sub-divisions can be seen below
Procedural memory
• implicit, non-declarative types of memory
• Procedural memory is used in reference to skills like how to perform certain actions > e.g. typing on your keyboard without looking at it, riding your bicycle, a pianist playing the piano
- Procedural memory occurs very slowly and gradually > in the lab we assess this using a serial reaction time task, this is where you have many trials where people learn a complex pattern (e.g. pattern on a keyboard) and see how their reaction time changes across the trials > usually the more trials (aka practice) you get the quicker your reaction time at doing the pattern > in am implicit way have learnt the pattern
Priming
• implicit, non-declarative type of memory
• Priming refers to when the processing of a stimulus is influenced by a prior encounter with the same or a related stimulus (certain features of the first stimulus could be similar to the other)
• Priming occurs rapidly > e.g. cat (prime) followed by picture of cat (target) then measure how quick ppts name the picture or e.g. cat (prime) followed by reading the word dog (target), dog and cat are not the same but are semantically related (both animals, both typical pets) so ppts would still respond fairly fast as opposed to if the prime was something semantically unrelated like “pen”
• The repeated presentation of a stimulus means it can be processed more efficiently, using fewer resources
• Priming is seen not only with RT’s, can be seen with the brain
Priming effect in the brain is called a repetition suppression effect. E.g. in the case of reading words, the brain areas responsible for reading words need fewer resources the second time the word is presented, so there is a suppression effect in these brain areas. > the brain areas will light up less so the second time (less active) because there is a repetition suppression effect only in the necessary brain areas related to the task being completed.
What makes explicit and implicit memories different?
Encoding
• Implicit memories are processed in a “bottom-up” or data-driven manner > essentially information is encoded in the same way it is received so it is easier to encode
• Explicit information is processed in a “top-down” or a conceptually/person driven manner > have to recognise the information + recognise you are learning the information
• In implicit tasks the person therefore has a more passive role (follow the data) whilst in explicit tasks the person has an active role (person-driven)
Both priming and procedural memory are implicit memory: information that is learned implicitly, without intention to learn, or awareness that the information is learned > you don’t even need to have any awareness that you are learning the information but you can still learn it
What makes explicit and implicit memories different? Retrieval
• Same information can be retrieved in more than one way depending on the task > could retrieve the same information in an explicit way vs in an implicit way
• E.g. in an experiment, if ppts are exposed to many word pairs (like car-school) then are at the end asked what word was paired with car? This would be using your EXPLICIT memory because you are actively thinking of what you saw with the word car. However, if the experimenter asks what other word comes to mind when you think of the word car? You are more likely to say the word school than another word due to your implicit memory (you saw car-school together), this is unconscious + may not realise why you are saying school > may not explicitly remember that you saw “car-school” but this is implicitly remembered without conscious awareness
• Another example of implicit retrieval is if you ask people to rate what word pairs you like (exposed to + unexposed) > so e.g. how much do you like car-school vs car-table? > more likely to say car-school because your implicit memory registered this + not car-table
The results differ depending on HOW people retrieve from memory > context of memories is the same but can be recollected differently (explicit retrieval = conscious recollection of prior experience, implicit retrieval = no conscious recollection, prior experience is still used but it doesn’t require conscious awareness of this)
How are explicit and implicit memories studied in lab settings?
• Typically, experiment would show a list of words > e.g. “remember list: plant, leopard, breakfast”
• Next you give the ppt a test (many different kinds) > the type of test differs + will assess if explicit or implicit memory is being used depending on the type of question you ask in the test
• Free recall = recall all words in any order (explicit)
• Cued recall = what word did you study with …. > the word (e.g. breakfast) is the cue to help recall other words in the same list (explicit)
• Forced-choice recognition = given two options like below + have to remember what they saw (explicit)
• Yes/no recognition = similar to the above just answer yes/no if they saw or not (explicit)
• Lexical decision = have to make a decision whether something is a lexical item or not (is it a word or not?) > faster to do this for words which WERE in the list (using implicit memory)
• Word fragment completion = If given a word with missing letters + asked to make a word, ppts are more likely to fill in the word which was seen in the list because you have been primed + are using implicit memory
Word stem completion = Missing letters for the end of the word + asked to make any word > more likely to finish with letters which make up what was in the list (implicit)
What does amnesia tell us about the structure of memory?
• Amnesia = term for loss of memory (“being without memory”)
• Amnesia can be caused by many different things such as
○ Acute virus infection that effects the brain
○ Parkinson’s disease
○ Brain resection as treatment for epilepsy
○ Physical accident
○ Korsakov syndrome
Even psychological (psychogenic memory disorders)
Patient HM
• Most famous amnesia patient + contributed to our understanding
• Dr Scoville performed a bilateral medial-temporal-lobe resection > Seizures originated in the region including the amygdala, hippocampus + hippocampal formation, and associated subcortical structures, so Scoville removed them bilaterally
• Breakthrough in understanding memory from memory loss in 1957
• H.M. had epileptic seizures after an accident, which became more severe with age (10 to 27) > In 1953 Dr. William Scoville performed a bilateral medial temporal lobe resection > helped learn about how memory is organised
• Removal of the medial-temporal lobe results in severe + selective impairments of episodic memory
The dark gaps in the image show the areas which were removed from HM’s MTL > can see all the areas which had to be removed as they are all very close together + are crucial for memory
Disconnecting explicit memory - HM
• Following HM’s surgery, he suffered from severe amnesia + couldn’t remember anything from after the surgery > lost his explicit memory from after the surgery (knowledge facts)
• Despite this deficit, HM still had an above average IQ, performed well on perceptual tasks + could recall events from way before the surgery (e.g. childhood)
• HM’s performance on implicit memory tasks remained intact > big discovery showing that implicit + explicit memory MUST be quite distinct to the point they are in different brain regions
• HM had anterograde amnesia where he forgets everything new after a few minutes > reduced ability to remember info acquired AFTER the onset of amnesia (after the resection for HM)
○ Memory for new personal events – impaired
○ Memory for new facts – impaired
• HOWEVER, memory for new skills remained intact > But explicit memory of learning these skills is impaired (cannot be consciously re-applied only can be used through implicit memory) Therefore – repeated items show a smaller improvement (e.g. when you repeatedly show a pattern, people w/o amnesia show improved performance each time the item is shown but HM did not show this)
Short-term memory – intact
Intact memory for skills
• HM was followed for 30 years by a memory researcher (Dr Brenda Miller) > did many tests with him + discovered new things about memory
• HM’s memory for motor learning was intact > in this task you learn how to trace a star using implicit memory > learn how to trace it in a mirror (so the trace is mirrored making it more difficult)
• HM showed he had normal memory for motor learning here because his implicit memory remained intact > amount of errors in tracing the star reduced the more he did it
- Research from Milner showed that acquisition of various skills was still normal + intact in many amnesic patients (some amnesia patients can learn certain skills like a person with intact memory would) > this is true for skills needed in the Tower of Hanoi, mirror reading and pursuit rotor task (keep a pointer/mouse on a rotating/moving dot)
Tower of Hanoi
• Point of the game is there are 3 rods and 7 discs > all discs will first be on one rod in ascending order (smallest at top) you need to get all the discs onto another rod without putting the larger discs over the smaller one (harder when you have more discs)
• These rules are complicated and doing the task itself repeatedly is something you can implicitly learn to do better
• If you ask an amnesic patient even after they have played this game, have you seen this game before? And do you know the rules to play? They will say no on both occasions because they will have no memory of doing the game > have to re-explain rules each time because the rules require explicit memory (patient has to actively think)
But there is proof of use of implicit memory because the number of errors on the moves used go down the more it is played + goes down in the same way it would for a functioning person > have a normal learning trajectory
Mirror reading
• Task where you read mirrored text > here rereading the same mirrored text will get easier so you will read it faster and faster but also get better when you read another mirrored text because it requires the same skill of reading backwards (procedural learning) > general improvement for mirror reading + specific improvement for reading the same text
• Amnesic patient NA results are compared to controls whilst Korsakoff patients were also compared
• Korsakoff patients have severe vitamin B1 deficiency > common in people with extended alcohol use > this syndrome causes impaired explicit memory and intact implicit memory (similar to amnesic patients)
• We know patient NA had impaired explicit memory because they did a word recall task compared to controls + found a significant difference where NA remembered significantly less than controls
• Non repeated words = new mirrored text, repeated words = same mirrored text shown
Patient NA + Korsakoff patients show a normal learning effect in both conditions however there is a difference between both patients and controls in the repeated condition where patients spent longer in reading the text than controls > this is because the controls use their explicit memory in remembering the words whilst the patients are impaired in this and can only rely on the implicit memory
Priming in Korsakoff patients (who have amnesia)
• Explicit memory in amnesia patients is significantly impaired compared to controls whilst implicit memory in amnesia patients show a normal learning curve however this is not exactly the same as controls > controls still perform better even when using implicit memory because controls typically use both implicit and explicit memory which improves their performance
- Results of a syntactic priming task below > prime with the syntactic structure of a sentence (e.g. passive sentence) > when you prime with a passive sentence, the person tends to respond using passive sentences > this is for both controls and patients even those amnesic patients have no explicit memory of being primed with passive sentences (there is a priming effect even though they don’t understand why or how)
Explicit memory - Neural circuits
- Circuit for explicit memory is not just the hippocampus > e.g. for HM the whole medial temporal cortex was affected which is made up of distinct subregions including the amygdala, hippocampus, perirhinal cortex and parahippocampal cortex > all very close together > entorhinal cortex is also important for explicit memory
- All of these areas are important for explicit memory, although these are not the only areas needed for explicit memory > also relies on the frontal cortex (dorsolateral + ventrolateral) and on the connections between the frontal and temporal regions
