Working Memory Flashcards

Question

What is Atkinson-Shiffrin Encoding Claim?

Answer 1

According to the Atkinson-Shiffrin model, rehearsal (sheer stupid repetition of items) is what gets information to stick. The longer an item spends being held in short-term memory, the likelier it is to be remembered over the long term. (False) (Amount of time spent in STM determines the likelihood of getting into LTM)

Answer 2

People have remarkably little memory for information that has spent time being rehearsed in short term memory. i.e. Penny, Apple Logo or Logos (Maybe pennies and logos aren't a fair test. After all, do we **really** look at them?

Answer 3

Popular briefly in the 1970's, suggested that it’s the “**depth of coding**” that determines whether information will get into LTM memory. "**Deep**" encoding meant thinking about the meaning, while "**shallow**" encoding meant thinking about the surface properties of the stimulus such as the sound or what it looked like. For example, the level of processing theory predicts that you will remember a list of words better if you spend your study time thinking about the definitions of those words than if you spend your study time thinking about words that begin with the same letter as the study words. (False) Deep: meaning Shallow: surface properties (look or sound like)

Answer 4

Researchers started testing using a **cued recall method** (Sound vs. Meaning). This is when you give people a hint to try to recall each word. It turns out, the results depend crucially on what kind of hint you give. **When the cue directs people's attention to how a word is spelled, suddenly words that were "shallowly" encoded are remembered better than words that were "deeply" encoded**. According to LOP, this is impossible. The words that were encoded in terms of sound simply wouldn't be there in memory anymore.

Answer 5

This states that memory for something depends on how well you can enmesh that thing in a network of rich, prior associations. That is, you remember better if you take what you’re trying to remember and link it up with things you already know. i.e. It’s easier to remember a story you hear about people you know well than about people you don’t know well

Answer 6

Levels of Processing wasn't completely wrong-headed. "Deep encoding” tends to help because it offers more opportunity for elaboration, linking up with other networks of meaning, than does shallow encoding. However, “shallow encoding” could also be very elaborative—say you were a professor of phonology.

Answer 7

Ideas and concepts are represented in our minds as "nodes" in a richly interconnected network of relationships between those ideas. Activating one particular node (so in this case a word or idea) spreads activation along the network to other nodes closely linked to the first. The more connections you have to an idea, the easier it will be to find when you are searching your memory banks. Mutiple retrieval paths are better The rich get richer

Answer 8

Grouping several items together into a familiar pattern, essentially treating them as a single unit. i.e. For instance, it’s hard to remember a long string of letters: **TPJQIMCNKUACUIG**. But it’s easy to remember a few words with the same number of letters: **QUICK****CAT****JUMPING**.

Answer 9

A device, such as a formula or rhyme, used as an aid in remembering. i.e. Method of Loci, Peg word mnemonic/Acronyms, rhymes, sentences Schoolhouse Rock was shown on TV in the '70's, during Saturday morning cartoons, using catchy songs to help kids learn stuff. Many kids memorized the Preamble to the Constitution this way, and learned about various parts of grammar.

Answer 10

It's based on the assumption that you can best remember places that you are familiar with, so if you can link something you need to remember with a place that you know very well, the location will serve as a clue that will help you to remember.

Answer 11

In the Peg Word mnemonic technique, you first memorize the short rhyme. Then, using these words as pegs, you create a visual image of each of the words you wish to remember.

Answer 12

Poor memory from early age could be a result of under or not yet developed complex semantic nets, so we have nothing to attach anything to—the rich get richer, and the poor stay poor. As a side note, it’s also believed that some brain structures essential for memory (notably the prefrontal cortex and the hippocampus) lag behind in development.

Answer 13

A further twist on elaborative encoding is the effect of the context/situation on later remembering. This simply states that memory is better when context at retrieval matches that at encoding than when not. If you study underwater you’ll later remember better underwater than on land. If you study semantically and are cued semantically you’ll remember better. (This explains the experiment where Levels of Processing went wrong.) If you study expecting a multiple choice test you’ll do better on a multiple choice test than a fill-in-the-blank test, and vice versa. Also refered to as “transfer appropriate coding” or “context dependent memory.”

Answer 14

Memories for dramatic, important events. Vivid, detailed, like a moment captured on camera. They are not a special or different kind of memory. They are not always accurate. Highly emotional events are remembered better than neutral events, but only the central details. Emotion may actually draw attention away from other details of the situation Why? Hypothesis 1: Emotion heightens memory encoding Hypothesis 2: Emotion narrows your focus of attention (“Weapon focus”)

Answer 15

Infantile Amnesia Chunking Mnemonic Devices Encoding Specificity

Answer 16

We organize our memories based on what we know about the world.

Answer 17

When people are asked to remember a disorganized list of items, over time they will start to group the items into some kind of meaningful structure, based on similarity of meaning (**memory for meaning, not exact wording**). This suggests that new items are being attached to existing networks of concepts. Evidence for semantic networks supporting memory processes comes from **semantic clustering**.

Answer 18

When something is strongly implied by the stimulus (it forms part of how we understand the meaning of the stimulus), but the thing itself wasn't actually there. We remember is the gist, the overall meaning. This can somtimes get us in trouble, when we may "remember" something that wasn't even there! We don't just "replay" our episodic memories, we **reconstruct** them. i.e. Think of the *Telephone* game

Answer 19

Actual Episodic Trace **Filling in the Gaps**: Perceptual "completion" effects, Unpacking Chunks, Assumptions based on general world knowledge **Subsequent Information**: The Misinformation effect, Imagination

Answer 20

The process in which people, in essence, zoom out the field of view when remembering a picture or scene. When people see the picture shown below on the left, later they draw it something like the drawing shown on the right:

Answer 21

The process by which we normally are able to distinguish our fantasies and imaginings from real memories. Sometimes, reality monitoring fails. If you've ever had an argument with your roommate about whether you asked them to take out the garbage (when in fact, maybe you only meant to ask them).

Answer 22

In the laboratory, the more frequently you’re told to imagine doing something, the more likely you are to later truly believe you did that thing. This is especially true for mundane, or boring tasks—if you’re repeatedly asked to imagine that you tamed a wild space unicorn and rode it into the sun, your prior knowledge about reality is very likely to keep you from mistaking these thoughts for real memories.

Answer 23

Adult’s recovering memories of horrible events from their childhoods (that didn’t actually happen) is referred to as **false memory syndrome**.

Answer 24

Some kind of relatively accurate representation of at least part of an initial experience. There is debate about how this might be represented in the brain, but there is good evidence from fMRI and single-cell recording experiments that one element can simply be reactivation of the perceptual networks that were originally active during the initial experience.

Answer 25

**Recall**: Free Recall, Serial Recall, Cued Recall Recognition Savings on relearning

Answer 26

Recall (or recollection) occurs when we have to generate or pull a memory into awareness. If I were to ask you what you did this morning, you need to think back and attempt to retrieve (or recall) what happened.

Answer 27

In **free recall**, items can be recalled in any order, but in **serial recall**, they must be recalled in the same order they were learned in order to be counted as correct. Serial recall is a useful method for making sure that subjects tap into memory for an actual event, rather than relying on familiarity or a vague sense that something was heard or seen recently.

Answer 28

If I give you a list of words, say BEAR, CAT, DOG, and ELEPHANT, and then later ask you to recall the words I listed, but prompt you with: B\_\_\_, C\_\_, D\_\_, and E\_\_\_\_\_\_\_, I’ve given you a cue to aid in memory, but you still have to retrieve the actual items.

Answer 29

In **recognition memory** you’re presented with items, words, or objects, and then asked if you previously experienced them. In a recognition test I might ask: did you see the word ELEPHANT? You don’t have to recall the actual initial presentation (although that’s one way to answer the question)—you could just determine whether elephant looks familiar, or gives you a sense that you’ve previously seen it.

Answer 30

This refers to the phenomenon that, if you learn something, but then forget it, you’ll relearn it more quickly the second time you study it than you will a new, previously unlearned something. That is, the initial memory was partially saved, even if you couldn’t retrieve it via recall or recognition. "Brain Muscle Memory" This phenomenon can be particularly remarkable in the case of a language learned in early childhood, forgotten, and then relearned.

Answer 31

This is the process by which experiences and semantic knowledge become durable memories.

Answer 32

With a severe concussion, memory can be erased for a small amount of time leading up the injury. The physical trauma can interfere with the consolidation process, which means that those things will not be turned into memories.

Answer 33

When experiences prior to being consolidated interfere with the process. If you were trying to memorize a list of medical words, this could be interfered with if you had just been studying a list of scientific names of animals. Both lists will include many Ancient Greek roots, and the interference between these similar items will affect the consolidation process. This can also happen in reverse, that is, not only can the animal names you studied first proactively interfere with your ability to memorize the medical words, but the studying the medical words right after the animal words could **retroactively interfere** with the consolidation of the animal words.

Answer 34

Extremely long-term or lasting memory which forms following extensive training, learning, or experience.

Answer 35

Even the act of trying to retrieve information can cause interference! There may be some variability in priming for specific instances. If you were asked to list all the states in the US, you’d likely come up short of fifty. However, you certainly know this information—you’d have no trouble with a recognition task.

Answer 36

This refers to when you are trying to think of a word, have a sense that you do indeed know the word, but can’t actually produce it.

Answer 37

**Implicit memory** is when your brain learns how to do something, or gets better at processing a type of stimulus, without you neccessarily being aware of having experienced it before. Because it tends to involve memory for how to do things, it is also called procedural memory. Stuff you know how to do.

Answer 38

This is memory for facts, events, ideas -- stuff that you can put into words and talk about. For this reason, it's also called declarative memory. Stuff you can put into words.

Answer 39

Classical Conditioning Operant Conditioning (may have explicit components) Perceptual Fluency Motor Skills Mental Procedures (most will have some explicit components) Priming

Answer 40

An automatic reaction to a stimulus can get re-trained to a different stimulus.

Answer 41

A behavior followed by a reward is more likely to be produced again in the future.

Answer 42

The ease with which information is processed based on manipulations to perceptual quality.

Answer 43

Tasks become easier as they are completed or solved. It becomes easier to do tasks of similar nature. i.e. Tower of Hanoi Puzzle

Answer 44

Impaired ability to store new explicit memories caused by damage to the hippocampus. **Preserved ability to store new implicit memories**.

Answer 45

**Korsakoff's Syndrome**: (from lack of vitamin B1) Causes problems learning new information, inability to remember recent events and long-term memory gaps. Memory problems may be strikingly severe while other thinking and social skills are relatively unaffected. **HM**: HM’s epilepsy was clearly in both hemispheres, so they removed both his left and right hippocampus. This caused severe, permanent anterograde amnesia. While he consciously remembered nothing of recent events, he could still learn.

Answer 46

Word Fragment Completion Picture Fragment Identification Reading Backwards Mirror Tracing Tower of Hanoi

Answer 47

People do better on words seen recently. Amnesics preform normally.

Answer 48

On repeated exposures people identify pictures with less info. Amnesics preform normally.

Answer 49

Everyone improves at reading backwards in general. Normals show a stronger repetition effect.

Answer 50

The mirror-image tracing study requires that the participant use the mouse to trace the star as shown in image, first with one hand, and then with the other hand. Because mirror image tracing is primarily a visual-spatial task, and each half of the brain controls the contralateral side of the body, it is expected that right-handed participants will take longer to complete the task with their right hand (controlled by the left hemisphere) compared to their left hand ( controlled by the right hemisphere). Amnesics show normal aquisition of new skill.

Answer 51

Explictit recognition can be dissociated from priming.

Answer 52

**Semantic processing** improves recognition. **Perceptual processing** improves priming.

Answer 53

We have mental representations of concepts and categories, but how do these representations work? For instance, essentially no one will be confused if you ask them to describe a dog. But how do we know what counts as a dog and what doesn’t? One possible answer is having a list of criteria. **If you meet those criteria, you're a member of the category, and if you don't, you don't. This is sometimes called "neccessary and sufficient conditions."**

Answer 54

For instance, with a dog, we might say that it needs to be furry, bark, wag its tail, etc. If it doesn’t do all these things, we might say, it’s not a dog. But what about this? This Mexican Hairless is certainly not furry. But she’s pretty clearly a dog. **It turns out it’s extremely difficult in most cases to come up with a set of necessary and sufficient criteria that will reliably and accurately identify all, and only, those things in the world that belong in a particular category**.

Answer 55

**In prototype theory, each category has a mental representation prototype that represents the best instance of that category**. For dogs, this might be a golden retriever or a german shepherd. Then, when determining whether something is a dog, we compare it to the prototype. The closer something is, the more likely we are to assign it dog-hood. **Prototype**: Best blanket representation of a specific category

Answer 56

Category members that are less typical, that are less close to the prototype, are in some sense less "good" members of the category, even though they are clearly still members. Thus, some dogs are “doggier” than other dogs, but still get to count as dogs.

Answer 57

● “central tendency” is not always prototypical ● how do you have boundaries at all? ● typicality ratings are not always meaningful ● can't handle "disjunctive" categories ● we don't always categorize by resemblance ● categories like "dollar bill" ● disjunctive categories

Answer 58

In everyday life, we often don’t use physical resemblance to decide category membership at all. In many cases we use **functional role**. This is often true of human artifacts, which are made to serve a particular function. It doesn't much matter what they look like, so long as they get the job done. Still other things are categorized by their **causal history**. What makes a Picasso a Picasso? Resemblance is not what matters here, **causal history** is.

Answer 59

In the end, it seems that category criteria can be different for different categories. They can depend on physical appearance, function, or formal relationships with other things in the world. Ultimately, our ability to decide category membership depends on us having *mental models* of how things in the world work and are related to each other.

Answer 60

A particular type of communication that has certain peculiar features. To decide whether something counts as language, we need to compare it to that list of features.

Answer 61

* Generativity ("building block") Structure * Phonemes * Morphemes/Words * Sentences * Referentiality * Displaced Refernce * Capacity to invent new terms * Combinatorial Structure (not quite the same as generativity, though related) * Compatible with the cognitive system (easy, natural, not explicitly taught)

Answer 62

Language is generative (or productive). That is, language is flexible and can be used to produce more-or-less infinite meaningful variations from a set of constituent parts.

Answer 63

The most basic building block of language is the **phoneme** (or the basic units of sound, for example the /d/ /o/ and /g/ sounds in the spoken word “dog”). **Phonemes** do not have meaning—they are arbitrary sounds. “Dog” and “fog” are separated by only one change in phoneme, but the meanings are entirely different.

Answer 64

**Morphemes**—units of meaning. These can be whole words (for example “respect”) or pieces of language that have a set meaning but are not technically words because they cannot stand on their own (such as the “ful” in “respectful"). It is debatable how to accurately count the number of words in a language, but there are a great many—hundreds of thousands or even more. These many pieces were constructed from the few available phonemes.

Answer 65

Remarkably we do the generativity trick again, using our vocabulary of words to construct **sentences** whose variety is virtually infinite. It is not difficult to come up with a sentence that has never before been uttered by humans.

Answer 66

Another feature of natural language is **referentiality**. That is, language is designed to refer to things in the world. Words like "pickle" or "moose" refer, or point to, something in the world.

Answer 67

The extraordinary communicative power of human language derives in no small part from the ability to **refer to things that are not currently present in the here-and-now**. Displaced reference can include **talking about things that are in the past, things that we think might happen in the future, things that are far away, and things that are entirely made up**.

Answer 68

It’s not just the words we use that matter, but how we put them together. This is how we indicate "who is doing what to whom." In English, we do this primarily by word order, using a subject-verb-object system.

Answer 69

Claim: Chimpanzees have better spatial working memory than humans! Why: Chimp learns to press numbers in order faster than humans. Problem: Humans had no practice while Chimps had **two years** of practice.

Answer 70

Claim: Baby Einstein videos claimed to enhance a baby’s IQ. Result: Babies who watch them have worse vocabulary growth! –They don’t use language in natural context –Screen time for kids under 2 not recommended

Answer 71

Main Question: Can apes be taught human language? Problems: Chimp voacl track not designed for human speech. Researchers made extravagant claims, refused to share raw data.

Working Memory Flashcards

(97 cards)