Chapter 7 - Memory Flashcards
(68 cards)
What are the three stages and pathways of the Atkinson Shiffrin model of memory storage?
- Information comes in as sensory input, and is stored in sensory memory, which is extremely detailed, short lasting memory. This allows us to see after images of input for a while after they have occurred.
- Short-term memory: This is where memory is stored for a bit longer, and if rehearsed it can stay in short term memory for a while.
- Long-term memory is where memories are stored to once repeated enough. Once stored in long term memory they can never be lost, but they might just not be able to be found. Long term memories can also be retrieved by short term memory.
What is sensory memory? How long does it last? What phenomenon does it explain?
Sensory memory is the storage of visual information for a short time after it is initially collected, so it stays long enough that you can actually process and see it. This memory lasts a bit more then half a second, and explains why we see lightning bolts as all one. In reality, they are actually extremely fast and occur in quick succession, but we see it as all one. This is because they stay in sensory memory as an after image, which lasts about half a second — when the actual bolt lasts about 0.2 seconds.
What is another example an after image, and when applied for 0.05 seconds how long does it last in sensory memory?
When a display of letter is given for 0.05 seconds, and then disappears, one can see those letters for around 0.2 to 0.3 seconds after. So it lasts about 4 times longer then the input.
What is the purpose of sensory memory? What is its capacity and what is its duration?
Why does the after image appear?
Sensory memory is used to hold information long enough that we can process its basic characteristics and actually understand what we saw. This is why sometimes when you hear something you ask someone to repeat themselves, but a short time later you realize what they said before they can repeat themselves.
Sensory memory has a very large capacity so that it can hold a lot of details. This is the tradeoff for duration, as it can only last for about 0.3 seconds for visual information and 2 seconds for auditory information.
Nerve impulses take some time to completely stop after last sensroy input is entered due to propagation, which is why an after image appears.
What is short term memory? How does duration compare to sensory memory? What is the capacity of short term?
Short term memory is based on the details of the image that one stores based on what they pay attention to. The duration is much longer then sensory memory (around 20-30 seconds) but it comes with a tradeoff- the capacity is much smaller. The capacity for short term memory is around 7 +- 2 items (or chunks) of information.
What is an experiment to test the capacity for short term memory?
The test to do this is called a serial recall test. You have to remember them in order and this forced them to not be grouped and stored in long term memory, but instead just recited in short term and that is where it is drawn from.
What are the patterns for free recall tests in the way that people remember things? Why is this and what are these effects called?
People will generally remember things near the end (in the last 7 seconds due to short term memory) and at the beginning of the list (due to long term memory) Because they will try to rehearse what they heard at the beginning and hence it will stay in their mind, and then they will neaturally remember the things recently heard in the last 20 seconds due to short term memory. The effect of remembering what was just heard is called the recency effect, and remembering what was heard at the beginning is called the primacy effect. This test is clearly used for both long-term and short-term memory testing as it is looking at how the two interplay.
What is maintenance rehearsal and what can it do?
Maintenance rehearsal is repetitive exposure of information to the brain to allow it to be stored in short term memory for longer then 30 seconds. On its own it will not be enough to cement it in long term memory, unless repeated long enough. More detailed and deeper thinking is required to do so.
What is long term memory? What is its capacity and duration? Is anything truly lost in memory — what are the two viewpoints on this?
Long term memory is when information is recited and memorized in a stronger way then just maintenance rehearsal, so that it is stored in a place that it can be retrieved at any time. This type of memory has an infinite capacity, meaning that it can store as much information as one needs, and it will never disappear.
There are two viewpoints on whether information is ever truly lost from memory:
1. The retrieval failure viewpoint: Information may be lost in memory, but never lost FROM memory, as there is no way to prove that it is no longer present. Just because you fail to find something doesn’t mean that it has vanished, because you can’t prove that something is gone.
2. Decay viewpoint: Long term memory can decay if not regularly used or if stored a long time ago. This says that information is lost FROM memory over time, and although this is more likely it can never truly be proven.
What evidence do we have that information is never lost from memory?
We can use a free recall test and a recognition test.
How can free recall and recognition tests be used to support the belief that nothing is ever lost from memory? Why does this work?
Free recall tests are used for reproducing information without any cues, meaning that people have to search in their memory to try to find previously learned information. This often fails, which appears to support the decay viewpoint. However, one cannot know if that information was lost from memory or lost in it — if lost in memory then maybe it can be recalled in another way even if it can’t be searched for and retrieved.
And this is proven by the recognition test, where previously learned information is embedded in new information and people have to indicate if information is old or new. They may not remember how or when they learned something, but they can recognize that they had learned that information before, highlighting that it was lost IN memory not from it.
This works because:
Information is often easier to recognize than recall due to the nature of the task. Recognition relies on a feeling of familiarity, requiring less active retrieval and relying more on the environment or context in which the information was encountered. Recall, on the other hand, necessitates actively retrieving specific information from memory without any external cues, making it a more demanding process.
Then, if you were to test for implicit memory, it ewoukd be even more accurate then recognition test, further highlighting that information is not lost in memory.
What is the difference between free recall and serial recall tests? What is the difference between their results? Which types of memory does each primarily test for?
Free recall tests allow subjects to recall a list of items in any order, and hence they test more for long term memory (as well as short term). But because it is in any order, it is simply based on how the participants encoded the information in the first place, and hence this can be based on rehearsed information that was stored in long term memory, or rehearsed information that simply just stayed in short term for a longer period, or it could be information that was just heard and hence it is based on short term memory.
Serial recalls tests are where subjects have to recall the list of items in their original order of presentation, making it much more focused on short term memory because you can’t take the time to commit information to memory, you have to keep listening to the new words coming in. This causes interference, because our short term memory can only hold so much and so new information is going to interrupt old information. In addition, this tests for rehearsal abilities, because we will try to repeat the information we were taking in so that it stays in short term memory for longer. Some of this information will then be stored in long term memory, but new information coming in will be ignored because rehearsal is very hard to do while trying to take in new info.
For free recall, the serial position effect occurs, where individuals have a strong memory for the beginning and end of the list, just not he middle. This is because information in the beginning will be rehearsed into long term memory, and hence if remembered it must be due to long term memory retrieval. This is called the primacy effect., then for information at the end of the list, it will be remembered because short term memory captured it, whilst you were trying to commit other information to long term memory.
What is some more evidence that suggests that information is never lost FROM memory?
The difference between first and second time learning — relearning of forgotten information is faster then first time learning. If you are learning a topic that you could not explicitly state any information about, then it may appear that it is lost from memory. But then when relearning you might feel familiar or remember things you have learned before, and that clearly proves that you just were unable to retrieve the information.
When the tasks is relearned, retention is measured (based on what you knew before). This retention is measured as a saving score, meaning what percentage of time did you save, or what percent of the information did you retain (based on the amount of time saved).
Example: The first time you learn a list of words it takes 20 minutes, and the next time only 5 minutes. You saved 15 minutes, which is 15/20 =0.75 or 75% of time saved. This also means that you have retained 75% of the information.
What are two of the main ways to effectively encode and store information in long term memory?
- Elaboration — how welll do you understand the information? Explain it deeply and connect it with other topics.
- Chunking — how do you group information together so it can be more easily remembered and also connected.
Why do some people have better memory? Why are some people better problem solvers?
The answers to these questions are the same:
Ones ability to do deep processing, their use of visual imagery to understand, and organization (chunking) are all the things that can effect storage.
This is why simple rehearsal will not be enough to store information to long term memory, you have to do deeper processing to have it stored more strongly.
What are the three types of encoding, and rank them based on their depth/strength?
- Structural encoding: This emphasizes the structure or physical appearance of the stimulus, this is the shallowest level of processing and will be the least likely to be stored and remembered in long term memory.
Ex: is the word written in capital letters?
PROCESSING WORDS - Intermediate processing: Phonemic encoding, meaning what does the word sound like? You have to do a bit more deeper processing and relating it to other words you’ve heard before, and hence this will be a bit stronger level of processing.
Ex: does the word rhyme with weight?
PROCESSING SOUNDS - Semantic encoding: This is where you actually understand the meaning and hence can relate it to information you’ve heard before, and so this is the deepest level of processing and is the most likely to be stored and remembered in long term memory.
Ex: Would the word fit in the sentence: “He met a ________ on the street?”
PROCESSING MEANING
Depth of processing theory says that _______ ________ of processing result in __________ _______ memory codes.
Depth of processing theory says that deeper levels of processing result in longer lasting memory codes.
This is why processing meaning and connecting it to previously learned information will make it much more strongly stored in long term memory.
What does the graph for results look like for type of encoding vs the proportion of words recalled? What study is this from?
When asked to process the case (upper or lower) of the word, this has the smallest proportion of words recalled. Then for processing rhyming, this had an intermediate proportion of words recalled, and lastly for semantic or understand the meaning in a sentence, this will have the highest proportion of words recalled. This once again proves that structural will be the weakest type of encoding, phonemic in the middle, and semantic the strongest type.
This is the Craik and Tulving study from 1975.
What is chunking? What is an experiment that represents it and what is a confounding variable for that experiment?
Chunking is trying to break down large amounts of information into smaller more manageable chunks, grouping that information together based on similarities. For example, remembering a phone number as three groups, rather then one long stream.
Another example, for experienced chess players they use chunking to group pieces on the board based on their positioning, and this makes it easier to remember that information because it is stored as one chunk of information. Masters can do this more effectively because they understand the board and possible positions, and so when studying a chess board they can group things based on their long term memory storage for how they have chunked information int he last. But novices only encodes the information in short term memory, and hence reproduces a lot less of the game positions.
When not allowed to chunk, the master and beginner had the same amount of reproduction of the game board because it was simply based on short term memory, and as long as they have the same short term memory capacity nothing should differ between them.
A confounding variable for this experiment would be that one person has a larger short term memory.
So when the to be remembered chess pieces were from actaul games, the master chess player was much more accurate and needed less trials to completely reproduce the board, based on previous experience for where pieces should go in relation to each other based on their chunking abilities.
This is because the chess master had stored many meaningful patterns from real games in long term memory, and so drawing from that chunked information he could be much more effiecient and accurate in reproducing those positions.
So to remember random patterns, you want to make up ________ for them, because _________ helps to strengthen memory and storage in long term memory.
So to remember random patterns, you want to make up meanings and relationships for them, because organization and chunking helps to strengthen memory and storage in long term memory.
Long term memory requires….
CONSOLIDATION
What is memory consolidation? When a new memory is formed, how is the brain changed?
This is the neural process by which encoded information is stored in memory. In other words, synaptic connection is strengthened and new connections are formed, but never new neurons! Just new synapses! This is the strengthening of brain plasticity.
When a new memory is formed, the brain is changed by the strengthening of synapses. This means that the postsynaptic neurons are more easily activated by presynaptic neurons, or in other words, it has a lower activation threshold.
What was the traditional view for memory consolidation — in terms of how the information was changed or not changed, and how accurate it was?
Traditionally, it was believed that information when encoded will be accurate and unchanged from the input, and this will be stored in long term memory just as is.
What is the new view for memory consolidation?
Now it is believed that when stored, information can be changed. Then when retrieved it will be reprocessed and hence this changes its value based on what you are retrieving it for and the context of the situation. You record what YOU remember, and what your brain filled in based on what was lost upon encoding. This changes each time information is stored.
Each memory is a memory of the previous retrieval, not of the original experience.