cogcsci test 2

Now imagine a different patient, A.D., with selective damage to the amygdala, with her hippocampus spared. Imagine that she got the same static shock from the doorknob and then returns to it later that day. How would her reaction differ from H.D.'s? (Hint: there are two major ways in which it would differ. Please give us both).

(1) A.D. will not hesitate or feel nervous about touching the doorknob (because even though the static shock felt painful to her, she is unable to develop a CR to the doorknob) (2) A.D. will remember that she got a static shock from the doorknob earlier that day.

how to make speech sounds

Sound is created by a stream of air, for which the lungs are the source, rushing and resonating through a tunnel. The tunnel for the sax is the tube and the keys, and the tunnel for speech is our vocal tract: the throat, tongue, teeth, lips, and for some sounds, the nose. As the tunnel changes its shape—by the pressing of the keys on the sax—the traveling of the airstream alters speed, changes course, and consequently produces various sounds.

response different classical conditioning vs operant

c: the conditioned response is usually an automatic behavior or reflex ( salivating, sweating, withdrawing) o: the conditioned response is usually a voluntary behavior (could become automatic over time) Behaviors to maximize a reward

9/11 brains

closer to world trade center--> more vivid memories and more reactions in amygdala because of the emotions remembered place the best more confident in memories for emotional events--doesnt make them more right

Evidence for phonological loop word length effect

people were able to remember more of a list of short words than long words even tho there was the same amount of words if miller was correct with the 7 plus or minus two, they should remember the same amount of words and the length of the words shouldn't matter but since saying the words in head and the longer the words are the longer it takes to say in your head so the less words you will remember

/ba/ vs. /pa/

b: voiced p: voiceless ba: vocal cord vibrates almost right away (20ms) pa has some delay (80ms): this is called Voicing Onset Time (VOT)

dopamine spikes signal prediction errors

before conditioning, dopamine neurons spike more when monkeys get a reward after conditioning, dopamine neurons spike in response to the CS, but no longer to the reward after conditioning, dopamine neurons actually suppress firing when the expected reward is omitted

why does emotion help us remember

emotions help us remember events associated with those emotions the brain structure important for this is the amygdala

classical conditioning pavlovs dogs process? US? CS? UR? CR?

food--> salivation bell + food--> salivation Bell alone --> salivation US- food CS- bell UR-salivation CR- saivation

temporal difference learning model

a model of learning that takes into consideration not only the size, but also the timing of the prediction error signal, so that it can respond to rewards that are earlier or later than expected, as well as rewards that are larger or smaller than expected (betting on football, expected values would not change in rescorla-wagner model (distinct time stamp), but temporal difference would take into account injuries during the game or the score and it would change)

Brandford & Johnson 1973

read two sentences group 1: POUNDING the nail group 2: LOOKING FOR the nail then asked "John was using a hammer to fix the birdhouse...." to see if people reported if they saw the sentence before or not 57% from group one said they saw the sentence w the hammer (THIS GROUP THOUGHT THE SAW HAMMAR BECAUSE THEY PICTURED POUNDING THE NAIL W A HAMMER EVEN THO IT DOESN'T SAY HAMMER 20% from group two said they saw the sentence w the hammer

You ask your friend Samantha to pick up a long list of items (e.g., "tea," "bread," "eggs," "tofu," "hummus", etc.) at the grocery store. (Assume she doesn't write anything down.) Later on, you find that what she remembered to buy depended on the order in which you said the items to her, like this: The fact that the probability of remembering an item from the end of the list was also somewhat high is known as the

recency effect

recency effect

remember the last things in a list because the least amount of time has passed since we heard it

how do stories help us remember

schemas- a preconceived notions about how things are that we have built up over experience - influence how we interpret things and how we recall things. we tend to recall things that are consistent with our schemas ex. stereotypes, social schemas, event schemas

primacy effect

tendency to remember words at the beginning of a list especially well because we rehearse it more/pay more attention to it

positive punishment

the administration of a bad stimulus to decrease the probability of a behavior's recurring ex. you will have to do the dishes if you break curfew

phonological loop

the part of working memory that holds and processes verbal and auditory information phonological store and the articulatory rehearsal process

retrieval

the process of getting information out of long term memory storage

articulatory rehearsal process

the process of internally speaking what you are trying to remember rehearsal that can keep items in the phonological store from decaying

Encoding

the processing of information into the long term memory (often involves exerting attention and/or rehearsal, but is usually unintentional

negative punishment

the removal of a stimulus to decrease the probability of a behavior's recurring ex. you will have your phone taken away if you break curfew

visuospatial sketchpad

the store for mental imagery= the creation of visual images in the mind in the absence of a visual stimulus

cognitive maps are analog but systematically distorted bc perspective

things that are closer to us, we view those distances as being longer than things that are further away from us

Brook's Dual-task inference experiment

trial 1: scan this block letter F in head and say yes when you hit an outside corner and no when you hit an inside corner Trial 2: scan block letter F in your head and point to the word yes when you hit an outside corner and no when u hit an inside corner ------------ memorize this sentence "a bird in the hand is worth two in the bush" say yes when hit a noun and no when u hit a verb... then trial two point --------------- results: shows a double dissociation between the phonological loop and visuosketehpad one task is a lot harder than other

contiguity

two events happening at or around the same time

iconic memory

visual sensory memory

Evidence for phonological loop phonological similarity mistakes

when people make mistakes, they are more likely to remember words or letters that sound similar - ex. F is often more confused w X or S even tho it looks more like E because it sounds more like X/S and you are saying the words in your head

phonological store

where sound-based info is temporarily stores in working memory (limited capacity)

are the phonological loop and visuospatial sketchpad different modules

yes

When things are very consistent and you get a reward each time.... when things are more spastic and you only get rewards sometimes...

you learn things quicker and easier it takes awhile to learn things

Consonants VS Vowels sound source

• Vowels: no air flow constriction • Consonants are produced by some kind of air flow constriction during articulation

Compare and contrast the three memory systems within long-term memory (classical conditioning, operant conditioning, and relational memory). What are the neural regions underlying each system?

● Classical conditioning ○ Memory for an association between a stimulus (CS) and a physiological response (CR); emotional learning. (Implicit) ○ Neural region: amygdala ● Operant conditioning ○ Memory for an association between a stimulus and an action; based on feedback. (Implicit) ○ Neural region: striatum (caudate + putamen) ● Relational memory ○ Memory for an association between a stimulus and another stimulus, or between a stimulus and a context (conscious/explicit: "declarative memory") Neural region: hippocampus

In lecture, we talked about an experiment trying to prove the human tendency to group words into phrases. The participants heard sentences in one ear and clicks in the other. They were then asked where in the sentence they heard the click. What were the findings?

● Click at break, they were accurate ● Click before break, they reported hearing it after ● Click after break, they reported hearing it before

What helps us remember?

-emotion -place -a story (make words that u have to remember into a story)

echoic memory

auditory sensory memory

What is the human simulation paradigm? What does it demonstrate?

- Demonstrates the degree of referential ambiguity (word learning from an infants perspective - some words are easier to learn than others) - Adults (Penn undergraduates) watch videos of a child-parent interaction, except the audio was muted, Instead they hear beeps every now and then and are instructed to guess the word uttered by the parent at that moment - To simplify, they are told when the word is a noun or a verb, and only needs to select from a small list of 25 possibilities - Word learning from the infant's perspective (cues) - To see the visual field of the child

A grammar G is a computational system that accepts positive examples, but rejects negative examples. Give one example of each category that helps the system learn how to phrase [boy] + [throw] + [ball] in past tense.

- The boy threw the ball (positive) - The ball was thrown (positive) - The boy throws the ball (negative) - The boy throwed the ball (negative)

Evidence for phonological loop articulatory suppression

- can disrupt people's internal speech - can make them say words (the the the the) out loud while trying to remember the words and repeat them in their heads - this will make it a lot more difficult

What was the result of Sperling varying the time of the partial report cue

- did terrible without the partial report cue -if played tone right at the beginning, they remembered almost all of the letters of the row - but as he played it later and later, they started not being able to report it - after 1 second they were not able to report it

Packard and McGaugh (1996) fill in the blanks ______________ intact saline in hippocampus Test Day 8 Strategy: Rats were more likely to rely on place information. Test Day 16 strategy: ______ Lidocaine in the striatum Test Day 8 Strategy:_______ Test Day 16 Strategy:________

- hippocampus intact - Test Day 16 Strategy: Rats were more likely to rely on the trained response. - Test Day 8 Strategy:Rats were more likely to rely on place information. - Test Day 16 Strategy: Rats were more likely to rely on place information.

cognitive maps are analog but systematically distorted: heirarchial organization

-one source of the distortion: hierarchical organization - when you represent space on a larger scale (ex. relationships between countries or states) the small-scale relationships (ex relationships between cities or neighborhoods) or lost -ex which is more west, Reno or San Diego? --people think about California vs Nevada, then say San Diego is more west than reno --reno is actually more west

central executive

-part of Alan Baddeley's model of working memory that oversees the visuospatial sketchpad, phonological loop, and episodic buffer. Responsible for shifting and dividing attention -The part of working memory that is responsible for monitoring and directing attention and other mental resources. -central executive can go in-between the two and focus attention

working memory 3 components

-phonological loop -central executive -visuospatial sketch pad

evidence of visuospatial sketchpad

-rotate objects in mind to see if two are the same

Sterling's Sensory Memory Experiment

-showed participants an array of letters for 50 ms only - they reported seeing about 4.5 letters -when used a "partial report" cue AFTER the array went off the screen (high, medium, low tone to signal the top middle or bottom row), they remembered 3.3 of the 4 letters on average - It is as if they still "saw' the image right after and then it started to fade. (sensory memory)

kosslyn mental scanning experiment

-subjects first memorized a map - then they imagined the map and "scanned" from location to another on it - their time to scan was measured - then asked them to "walk from ____ to _____" -distance in the map was proportional for time for scanning (good evidence that people use mental images) -just using logical relations, shouldn't take longer for long distances

a) In the partial report condition, the grid is again flashed for 100 msec, but immediately after the grid goes off, one row of the grid is cued and the subject is asked to just report the items in that row. Suppose we consider trials in which 15 items are flashed and there are 5 items in the row that was cued. We discover that on such trials, subjects are on average able to correctly recall 4 items. From these data, at least how many items on average do subjects register from each presentation, even if they are not subsequently able to encode all these items into memory? (By the term register we mean that the information about the item was available somewhere in the brain after the grid was turned off.)

12

what is the capacity of working memory

7 items plus or minus 2 about 10 seconds

Write down their past tense forms. Some share the same past tense form Apply the Tolerance Principle to see if any of these sub patterns is productive. Bring Cling Ding Fling Ring Sing Sling Spring String Swing Wring Wing Zing

Bring brought Cling clung Ding dinged Fling flung Ring rang Sing sang Sling slung Spring sprang String strung Swing swung Wring wrung Wing winged Zing zinged bring-brought i—>a: sing, ring, spring i—>u: swing, string, fling, cling, sling, wring -ed: wing, zing, ding N = 13, N/lnN = 5. Thus a productive pattern must work for 13-5 = 8 verbs in this class. Even the most dominant pattern (i—>u) is inadequate.

Categorical perception examples

Color perception (example given in the lecture), smell perception (although odors generally vary in the molecular composition to varying degrees, they are perceived as odor categories), recognition of emotions in faces (see figure below, I think this may help illustrate the concept of categorical perception in general)

Do we perceive speech stimuli as continuous, or as categorical?

Even though speech stimuli are distributed in a continuous space, we appear to perceive them as categories.

Continuous perception examples

Height (example given in the lecture), loudness of a tone (example given in the lecture), temperature (we don't just have two categories of warm vs. cold

a) You no longer want either rat to expect cheese when they hear the bell, so you want toextinguishthe conditioned response. Based on the expected values at the beginning of Trial 5, in which rat do you think extinction will take a longer time - Timothy or Jasper? Explain why.

I expect extinction to take longer in Timothy because his expected value is higher. Jasper has not learned as well that the bell predicts the cheese so it will be easier for him to extinguish.

Describe what you expect to see in the dopamine neuron firing rate in response to the outcome for Jasper's trial 3.

I expect to see a decrease (dip, suppression) in dopamine firing because the prediction error is negative.

a) You bring Timothy and Jasper to the lab, where they put electrodes into their ventral tegmental areas in order to record from their dopamine neurons. Describe what you expect to see in the dopamine neuron firing rate for each of the following (you can assume for this exercise that Rescorla-Wagner prediction errors are the same as temporal difference prediction errors). Describe what you expect to see in the dopamine neuron firing rate in response to the bell and in response to the outcome for Timothy's trial 1.

In response to the bell, I expect to see a "flat" firing rate (no firing), but then when the outcome (the cheese) arrives, there will be a large spike in the dopamine neuron firing because the prediction error is positive and large.

Describe the Sperling Memory Experiment. Discuss the main findings of this study.

In this study, participants were shown an array of letters for 50 ms. They reported seeing about 4.5 of the 12 letters. When participants were shown a "partial report" cue AFTER the array went off the screen (i.e. a high, medium, or low tone to signal the top, middle, or bottom row), they remembered 3.3 of the 4 letters on average. It's as if they still "saw" the image right after and then it started to fade. The longer Sperling waited until playing the partial report cue, the worse they did. The study suggests that sensory memory lasts up to only about 1 sec. Sensory memory for vision is called iconic memory. Sensory memory for hearing is called echoic memory. Problems with Miller's conclusion of his experiment: Main problem is chunking. Some numbers or letters can be combined into larger, more meaningful units allowing the participant to hold more information in their working memory. Some participants might be able to chunk some items together. For example, C I A F BI N B C C B S is easier to remember because a person can group the letters like this (C I A) (F B I) (N B C) ( C B S). also... Some languages have longer or shorter words than others, which can cause variations in this number in terms of remembering words. Items also might have different significances for different people, creating another pneumonic to help them remember the items.

positive reinforcement

Increasing behaviors by presenting positive stimuli, such as food. A positive reinforcer is any stimulus that, when presented after a response, strengthens the response. ex. you can have ice cream if you do your work

negative reinforcement

Increasing behaviors by stopping or reducing negative stimuli, such as shock. A negative reinforcer is any stimulus that, when removed after a response, strengthens the response. (Note: negative reinforcement is not punishment.) ex. you do not have to wash the dishes if you finish your hw

infant sound recognition

Infants at birth can distinguish virtually all consonant pairs such as ba~pa even though they lose the ability to do when later

You ask your friend Samantha to pick up a long list of items (e.g., "tea," "bread," "eggs," "tofu," "hummus", etc.) at the grocery store. (Assume she doesn't write anything down.) Later on, you find that what she remembered to buy depended on the order in which you said the items to her, like this: If you asked Samantha to color in a picture in a coloring book during that 30 seconds, would it interfere with her ability to rehearse the grocery list? Why or why not?

It would not. Since coloring does not use the phonological loop (It is a visuospatial task), then it won't interfere with her ability to rehearse the list.

What methods allow us to know which brain regions are necessary for a certain process? Explain each method and their limitations.

Lesion method: when patients have damage to a particular brain region and are unable to perform a task, this means that region is necessary for that task. A limitation is that we do not have control of the lesion's location. "Temporary lesion" method: when a patient's brain region is temporarily impaired (drug or TMS) and are unable to perform a task, this means that the region is necessary for that task. A limitation is that it is not very precise. Another is that TMS can only affect external brain regions.

Packard and McGaugh (1996) fill in the blanks Striatum Intact Lidocaine in Hippocampus Test Day 8 Strategy:________ Test Day 16 Strategy: ______ Saline in the striatum (caudate) Test Day 8 Strategy:_________ Test Day 16 Strategy: _________

Lidocaine in Hippocampus Test Day 8 Strategy: Rats were equally likely to rely on place information and the trained response. Test Day 16 Strategy: Rats were more likely to rely on the trained response. Saline in the striatum (caudate) Test Day 8 Strategy: Rats were more likely to rely on place information. Test Day 16 Strategy: Rats were more likely to rely on the trained response.

Can finite state machines describe the structure of human language? Why or why not?

No, finite state machines cannot describe the structure of human language. There are long-distance relations in English, where corresponding pairs of words (either-or, if-then, etc.) can straddle an arbitrarily long sentence (e.g., the boy eats ...), and these long- distance relations cannot be described by a finite state machine.

Patients with damage to the hippocampus have been shown to have no problem navigating around their own houses, if they have been living in the same place for many years. They do, however, have trouble navigating in new locations. Explain this discrepancy. (Hint: what memory system is intact that allows them to navigate around their own house?)

Patients with hippocampal damage still have their striatal / operant conditioning / habit learning system intact, so because they have been living in their houses a long time, they form habits that allow them to associate stimuli in the house with actions (e.g., "turn right" at the end of the hallway to go to your bathroom). So even if they can't tell you what room is connected to which, they can still manage to get where they need to go. They can't navigate in new places because they don't have a habit for new places and would need their hippocampal / relational memory system to learn the relationships between places.

Describe the four types of operant conditioning. Provide examples of each.

Positive reinforcement: provide something good to increase behavior treating a child to an ice cream cone when he stays quiet and obedient during a shopping trip is positive reinforcement. The child's behavior (being quiet and obedient while out shopping) is reinforced by adding something pleasant (an ice cream cone). Negative reinforcement: take something bad away to increase behavior removing restrictions from a child when she follows the rules is an example of negative reinforcement. Something unpleasant (a set of restrictions) is removed to encourage the child's good behavior (following the rules). Positive punishment: provide something bad to decrease behavior spanking a child when he throws a tantrum is an example of positive punishment. Something is added to the mix (spanking) to discourage a bad behavior (throwing a tantrum). Negative punishment: take something good away to decrease behavior After two children get into a fight over who gets to play with a new toy, the mother simply takes the toy away from both children. positive = add stimulus, negative = remove stimulus || Reinforcement = increase behavior, Punishment = decrease behavior

What are some of the problems with Ekman's idea of the 6 basic emotions? What is another model that can be used to understand and measure emotions?

Problems with Ekman's idea ● Emotions are not binary (can be feeling more than one at a time) ● Different levels/intensity of an emotion but same facial expression ● What makes an emotion "basic" Dimensional model of emotion, allows for different levels of intensity, arousal, valence

Explain the difference between productive and unproductive rules of language. Give examples for each of the two types of rules.

Productive rules ● They can apply to new instances, thereby giving the power of infinity to language. ● Examples: -s for plurals, -ed for simple past, -ing for progressive form -Unproductive rules ● They only apply to a fixed list of words, and do not generalize. ● Examples: warm - warmth, grow - growth, true - truth

You incorrectly recall a conversation that occurred in your dream as reality. What kind of error is this? Explain the error in your own words.

Source monitoring errorThis term generally refers to memory errors that occur when we remember that specific events happened, but not in what context.

Explain the blocking effect. How does the Rescorla Wagner model account for this phenomenon?

The blocking effect describes the phenomenon that the conditioning of an association between two stimuli, a conditioned stimulus (CS) and an unconditioned stimulus (US), is impaired if, during the conditioning process, the CS is presented together with a second CS that has already been associated with the unconditioned stimulus. the Rescorla Wagner model would account for this effect by stating that during the second phase of the conditioning training (when both CS are paired with the US) there is no prediction error that would allow for learning about the newly introduced CS.

Do you expect the dopamine neuron firing rate to be higher in response to the cheese in Jasper's trial 2 or Jasper's trial 4? Explain your answer.

The firing rate will be higher for trial 4 where the prediction error is 0.625 compared to trial 2 where the prediction error is 0.5, since dopamine firing rates correspond to prediction errors. (Without referencing the numbers, students could also say that Jasper will be more "surprised" by the outcome in trial 4 compared to trial 2, hence the dopamine firing rate will be higher for trial 4).

Rescorla-Wagner model

a cognitive model of classical conditioning; it holds that the strength of the CS-US association is determined by the extent to which the unconditioned stimulus is unexpected first time going to restaurant expected value (t=1)=0 you like the chicken reward(t=1)=1 prediction error =reward(t=1) -EV(t=1)=1 to updat: learning rate=.2 new EV(t=2)= EV(t=1) +LR*(prediction error)

Explain the difference between Cannon-Bard theory of emotion and James-Lange theory of emotion. Provide an example for a Cannon-Bard response to a sensory stimulus and a James-Lange response to a sensory stimulus.

The two theories differ in which comes first: feeling versus the bodily response comes first After you perceive the stimulus, James-Lange theory argues you experience the physiological response then you label the experience. For example, you first have a physiological response of shock like a gasp and then label the experience as shocking. (Any example works as long as the subjective feeling is a result of the bodily response, e.g. you are sad because you cry.) After you perceive the stimulus. Cannon-Bard theory argues you first have the subjective feeling then you have the expression and physiological response, or the subjective feeling and physiological response occurs at the same time. For example, you feel tired then you yawn. (Any example works as long as the subjective feeling comes first then the physiological response.)

Draw the sensory-working-long term memory diagram. What is the capacity of the working memory? Do you see any problem with the capacity number? How does the working memory bump up information capacity within this capacity?

The working memory can hold up to 7 units of information. Working memory bumps up information withholding capacity by chunking, which means to group similar or related information into one chunk which forms one unit. As a result, the working memory can remember up to 7 chunks of information. (Side note: chunking helps to store information into the long-term memory as well)

Kamin's blocking experiment

Then he gave the animal numerous trials with the elementary CS, such as a noise followed by cheese; at the end of this phase, the animal will show a conditioned response to the tone. Next he gave the animal additional trials with a compound CS, in which the noise and light appear simultaneously, followed by cheese When we test the animal's reaction to light alone we see that testing with the light yields no conditioned response. It's as if they never received any pairings of the light and noise at all. - Kamin's blocking effect demonstrated that if a CS provides no new information about a US, there will be no learning of the CS-US pairing - CONTIGUITY BETWEEN A CS AND US IS NEITHER NECESSARY NOR SUFFICIENT FOR LEARNING

a) If H.D.'s partner were to ask her why she is hesitating to touch the doorknob, what do you think she would say? Explain your answer

There are many correct answers here, but the key thing is that she would not explain that she got a static shock earlier in the day. She might say "I'm not sure why," or "Sometimes doorknobs give you static shocks," (in very general terms) or she might make up a completely different story, like "The doorknob is really hot to touch." The reason she is unable to explain the actual reason is because she is missing her hippocampus (relational memory system) so she cannot form new memories, and therefore she will not have any memory of what happened earlier in the day. She has no factual/declarative memory/learning.

a) Imagine a patient with severe hippocampal damage named H.D. She is going to open the door to the bedroom in her apartment when she feels a static shock from the doorknob. It is somewhat painful and causes her to automatically withdraw her hand and raises her heart rate a bit. The next time she goes to her bedroom door later that day, she feels a bit nervous and hesitates before touching the doorknob. Since H.D. is feeling nervous before touching the doorknob, that means that which memory system, and corresponding brain structure, are intact?

This means that the classical conditioning system (or emotional learning system), and amygdala are intact.

a) How would you show that the rule for "add -ed" you have just described in (b) is productive?

Three made-up words, one for each rule You don't need to actually make up words, but you do need to say that the experiment would involve made up words, one with each type of ending. Partial credit if you use words that already exist but the problem with that is that people may have memorized those endings so it's not really proof. To see if a rule is productive you need to see what people would apply in new situations.

How can we tell if infants have a categorical perception?

Using High-amplitude Sucking Technique. Let the infant listen to a sound for a long time until she is bored (habituation), then change to a new sound. If she notices that there is a difference, she will increase her sucking on the pacifier.

You ask your friend Samantha to pick up a long list of items (e.g., "tea," "bread," "eggs," "tofu," "hummus", etc.) at the grocery store. (Assume she doesn't write anything down.) Later on, you find that what she remembered to buy depended on the order in which you said the items to her, like this: a) You decide to test this "working memory hypothesis" on Samantha by having her repeat back the list to you (in no particular order) after about 30 seconds. If you wanted to interfere with her ability to use her working memory during that 30 seconds, describe a task that you could ask her to do while she is rehearsing the list during that time.

You can ask her to say something, for example, "the the the." (There will be multiple correct answers here but they only get full credit if the thing that you ask her to do involves an articulatory process (speaking something).

a) You talk to your brother and he corrects you about where you were (you trust he has a better memory than you do!). What kind of memory error have you made by concluding what you did in part (b)? Explain this error in your own words.

You have made a source monitoring or source misattribution error. You remembered something about the memory but you were not able to link it to the right source or context.

Dopamine prediction errors: Draw the expected response from a dopamine neuron in the ventral tegmental area in the following conditions: a) Before conditioning: b) After conditioning I c) After conditioning II

a) Before conditioning: No prediction + reward occurs Spike in response when reward comes, flat otherwise b) After conditioning I: Reward predicted (cued) + reward occurs spike in response Spike in response to CS, but not to reward (flat afterwards) c) After conditioning II: Reward predicted (cued) + no reward occurs Spike in response to CS, reduced firing when the expected reward does not occur

You have decided to train your pet rat Timothy to expect a piece of cheese whenever you ring a bell. Rats naturally salivate in response to being presented with cheese. a) In this classical conditioning situation, what corresponds to each of the following?Conditioned stimulus (CS): __________________________ Unconditioned stimulus (US): ______________________ Conditioned response (CR): _________________________ Unconditioned response (UR): ______________________

a) Conditioned stimulus (CS): bell or bell ringing Unconditioned stimulus (US): cheese Conditioned response (CR): salivation Unconditioned response (UR): salivation

What is cross-situational learning? How is this different than propose but verify (hypothesis testing)?

a) Cross situational learning: Keeping track of co-occurrences between labels and objects and the one with the most co-occurrences emerges as the winner. PbV does not keep "memories" of previously unchosen meanings at trial n+1, instead the success/failure at trial n+1 is dependent on the hypothesis made at trial n. Different way of phrasing it: This is a big data global approach. This is when people over different situations keep track of all co-occurrences between labels and objects. From the information across different situations a winner emerges. b) Hypothesis testing: PbV does not keep "memories" of previously unchosen meanings at trial n+1, instead the success/failure at trial n+1 is dependent on the hypothesis made at trial n. Different way of phrasing it: This is concluding a hypothesis object for a label. Hypothesis testing means that people verify their hypothesis against the data presented and revise it if necessary. Cross situational learning is deterministic which means that over trials the highest frequency of an object in each scenario wins. Hypothesis testing is not deterministic, it is stochastic meaning that the learners can make different choices along during the different situations.

How should we characterize the pattern in the pronunciation of "ed" in these three sets of words? That is, can you formulate a rule that would account for these different pronunciations in different cases? You should use the example in the Halle reading as a template. Note that we are interested in how the sounds are pronounced, not spelled!

a) Depends on the final "sound"/phoneme of the word: t/d: id Voiced: d (note that this includes both voiced consonants and vowels, which are always voiced) Voiceless: t A pat on the back if the answer includes some order of the rules: because t/d are voiceless and voiced respective, the rule must apply before the rest.

Compare and contrast the Rescorla-Wagner model and the "temporal difference" model.

a) Rescorla-Wagner model ● R-W assumes discrete time steps ("trials"), so any new information during the trial can't be incorporated ● R-W assumes that we update only from outcomes (rewards or punishments), not just information, so it can't account for phenomena like second-order conditioning ● Update value of start based on end b) "Temporal difference" model ● You can have a prediction error at any point (or "state") along the way, even without an actual reward or punishment ● Update based on how much closer to the reward each state gets you ● Instead of just updating your expectations after you get a reward/outcome, you can also update your expectation based on any new information that you get about rewards. c) Both models ● Learning from prediction error

analog vs logical relations

a: imagining a map or metaling walking between the buildings l: recalling "huntsman hall is 2 blocks west of vp"

Brain region: Amygdala

association: Stimulus -> Physiological response memory system: classical conditioning, emotional learning

Brain region: Hippocampus

association: stimulus --> or stimulus --> context memory system: relational memory

brain region: striatum (caudate+putamen)

association: stimulus --> response stimulus--> action memory system: operant conditioning

You ask your friend Samantha to pick up a long list of items (e.g., "tea," "bread," "eggs," "tofu," "hummus", etc.) at the grocery store. (Assume she doesn't write anything down.) Later on, you find that what she remembered to buy depended on the order in which you said the items to her, like this: Samantha is disappointed in how few items she remembered to buy at the store, so the next time you give her a list, she uses some memory strategies in order to encode the list into her long-term memory. First, she groups the items into categories, like "dairy," "drinks," and "frozen foods." What is this strategy called?

chunking

Seeing speech example

closing your eyes: here ba opening your eyes: ga today mind hears: da

Reinforcement marrs 3 levels

computational level: the goal of classical conditioning is making predictions algorithmic level: Rescorla-wagner and temporal difference models are proposed algorithms. Temporal difference learning is more general and explains more implementational level: dopamine neurons spike when receiving a reward

Explain the blocking principle, and give examples of how it applies both to language and general communication.

i. Blocking principle: 1. When there are two forms to realize the meaning of a word, the more specific form wins out 2. Irregular verbs are not predictable and their past tense forms must be rote memorized; when retrieval of the irregular/specific form (e.g., felt) is not successful, the productive rule "add -ed" kicks in and children say "feeled" 3. Innate (because children often don't get negative feedback) ii.Example (language): went (specific) blocks goed (general) iii.Example (general): square (specific) blocks rectangle (general)

You ask your friend Samantha to pick up a long list of items (e.g., "tea," "bread," "eggs," "tofu," "hummus", etc.) at the grocery store. (Assume she doesn't write anything down.) Later on, you find that what she remembered to buy depended on the order in which you said the items to her, like this: She also imagined where the items were in the grocery store as you read the list to her, so that she only had to walk around the store in her mind to recall them all. What is this strategy called?

method of loci (or memory palace)

how does place help us remember

method of loci- associating items with areas of space in order to remember them better later - hippocampus has place cells

Is contiguity necessary for learning

no not necessary because you learn things that are separated in time consequences happen much last and we still associated with something earlier (conditioned taste aversion: eat something new, next day feel sick, associate sickness w new food even tho it happened a day before)

You ask your friend Samantha to pick up a long list of items (e.g., "tea," "bread," "eggs," "tofu," "hummus", etc.) at the grocery store. (Assume she doesn't write anything down.) Later on, you find that what she remembered to buy depended on the order in which you said the items to her, like this: Some scientists have claimed that the effect in (b) exists because the last item on the list is still being rehearsed in working memory when the list is recalled. Given the nature of the stimuli being remembered, what is the name of the working memory store that the list of items would be in?

phonological store (or phonological loop)

You ask your friend Samantha to pick up a long list of items (e.g., "tea," "bread," "eggs," "tofu," "hummus", etc.) at the grocery store. (Assume she doesn't write anything down.) Later on, you find that what she remembered to buy depended on the order in which you said the items to her, like this: The fact that the probability of remembering an item from the beginning of the list is higher is known as the

primacy effect