Chapter 6 - Learning

Question 1

What are the three types of learning? What are the two types for each?

Answer 1

Nonassociative: learning about a stimulus such as a sight or sound in the external world.
2 types:
Habituation = when our behavioural response to a stimulus decreases as more stimulation occurs, because we no longer need to focus on it as much as it is not posing a danger and is taking up our limited attention. We still process it but we ignore the input.
Sensitization = when the behavioural response to a stimulation increases as we are repeatedly exposed to it — meaning we can more easily recognize it and so we more quickly respond. So this would be when it is dangerous, and so our response pathway is amplified to low us to have that larger response to live.

Associative: learning the relationship between two pieces of information — how are they associated with each other?
2 types:
Classical conditioning = when we learn that a stimulus predicts another stimulus, then our response to the second stimulus will also become our response to the other one.
Operant conditioning = when we learn that a behaviour leads to a certain outcome, then we will learn to do that behaviour or not based on the outcome.

Observational: learning by watching how others behave
2 types:
Modeling = imitating a behaviour seen by others
Vicarious learning = learning to engage in a behaviour or not after seeing others being rewarded or punished for performing that action.

Question 2

What is classical conditioning? What is an example? What is the operational definition for it?

What is the unconditioned response? What about the unconditioned stimulus?

What is the conditioned response? The conditioned stimulus? What about the neutral stimulus? What is the response to a neutral stimulus?

Answer 2

A type of learning where a neutral stimulus acquires the capacity to evoke a response that was originally evoked by another stimulus.
An example of this would be that food produces drooling, so if you pair a bell with food, then eventually the bell will produce drooling.

The unconditioned response is the response which naturally occurs due to normal psychological pathways. For this example this would be drooling in response to food.
The unconditioned stimulus is the stimulus that requires a response without training the thing to do so (like the food).

The conditioned response is the response that was originally due to the unconditioned stimulus, but which is now paired with some other stimulus. So it is a response to an object that would not normally occurring.
The conditioned stimulus is the stimulus which did not originally produce a response but now does because it was paired with an unconditioned stimulus.

The neutral stimulus is the stimulus which originally produced no response but once paired with the unconditioned stimulus will produce the same response in the person.
When a neutral stimulus is provided, no response will occur until it is paired with an unconditioned stimulus and produces an unconditioned response.

The operational definition is a measurement of the unconditioned response. So drooling for the dog one.

Question 3

Why does classical conditioning happen?

Answer 3

Classical conditioning happens because when you pair an unconditioned stimulus with a neutral stimulus, and they always show up together, they will be stored together in memory. So when you just produce the neutral stimulus, it will now be conditioned because it will be recognized in the same memory and trigger that neurological pathway to produce the conditioned response.

Question 4

What is the process called when trying to remove this conditioning? How is this done?

Answer 4

This process is called extinction, and to do this each time the unconditioned stimulus appears, the conditioned stimulus is not there. Eventually, these two sources will stop being stored together in memory and it will be extinguished.

Also after an extended time, if the conditioned stimulus is presented without the unconditioned stimulus, the conditioned response will extinguish because they realize they are no longer connected.

Question 5

When trying to condition someone, what will happen after a while if the CS is presented without the US?

Answer 5

If it hasn’t been very long, the CR will quickly decrease, but if they have been ingrained in memory for a long time, then it will take longer for the CR to extinguish.

Question 6

What happens if after a 24 hour rest from conditioning and then extinction, the CS is presented alone once again? What about another 24 hours later?

Answer 6

The brain will revert back to its stored neurological pathways and have a SPONTANEOUS RECOVERY, responding weakly to the CS with a CR. It will then extinguish again.
The another 24 hours later there will be another weaker spontaneous recovery, because the CS has been presented along for so long. The longer it has been ingrained in your memory, the more breaks that will be required to make it go extinct.

Question 7

What is generalization and what is discrimination?

Answer 7

Generalization: this is a learning relationship where one conditioned response is generalized to other similar stimuli, meaning that response will be produced even when the conditioned stimuli is not the one producing anything. So they have to sound similar to do this.

Discrimination:
This is when the conditioned stimulus is highly specified, and the learner is not going to react with their conditioned response to any similar stimulus.

Question 9

How to do discrimination training?

Answer 9

To do discrimination training, pair a certain frequency of stimulus with the unconditioned stimulus repetitively, and then introduce a different frequency but similar stimulus WITHOUT the unconditioned stimulus repeatedly, and so they will learn to only respond to the specific frequency stimulus.

Question 10

For a graph of frequency in tone of the conditioned stimulus vs the strength of the conditioned response, what will the generalization trained graph look like, and what will the discrimination trained graph look like?

Answer 10

For this graph, the closer you get to the conditioned stimulus frequency, the larger the conditioned response will be for both.

HOWEVER, for the generalization plot, it will be a wide shorter curve, because there is a larger deviation from the exact conditioned stimulus. Frequencies close to that conditioned stimulus will still induce a response.
For the discrimination plot, it will have a very low response rate until very close to the conditioned stimulus, because the person has been trained to not react unless it is the specific frequency. And any small deviations might just be because they cannot notice a difference between the specific frequency and the ones close to it. So discrimination training ensures there is almost no response to the different frequencies.

Question 11

What is second order conditioning? What is an example of this?

Answer 11

Second order conditioning still follows the classical conditioning pattern. However, now the conditioned stimulus has been trained to produce another conditioned response. Meaning the US produces a UR, and is US and CS1 are paired, this produces a CR1, which will occur when CS1 is shown on its own. Then, if you pair another item with the now conditioned CS1, you will get a newly conditioned response to that second item (which is the same response as UR and CR1), and can be called CR2. Eventually after enough time CS2 will produce a CR2 on its own.

An example of this would be a nurse giving a shot. At first, since a needle produces pain (US) and pain produces crying (UR), the needle is the conditioned response because is always paired with pain. So just seeing the needle creates crying. But the more often that the person gets the shot, the more they will realize that the nurse means there’s a needle coming which means there is pain. Therefore the nurse will become the the secondary conditioned response, which will produce a reaction before the shot is even shown.

So here:

US: Pain
UR: Crying

CS1: needle
CR1: crying

CS2: nurse
CR2: crying

Question 12

If there is a rat in a chamber, and you place a harmless blue bar in the chamber, the rat hides. Then you put a red light with the blue bar, and eventually the rat will hide due to the red light on its own. What type of conditioning is this, and what are the various parts? (CS, CR, US, etc.)

Answer 12

In this case, the blue bar is harmless, and so the rat’s fear of it is irrational. Therefore, this is not an unconditioned response, unconditioned is natural and based on your autonomic nervous system.

Therefore, this bar is already the first conditioned stimulus, which was originally conditioned with something else we do not know.
Then the red light is the second conditioned stimulus, as it was paired with CS1 to produce the same CR.

So:
US = ?
UR = hide

CS1 = blue bar
CR1 = hide

CS2 = red light
CR2 = hide

So the responses will always all be the same.

Question 13

What is the law of effect proposed by Edward Thorndike?

Answer 13

The law of effect says that:
Any behaviour that leads to a “satisfying state of affairs — very ambiguous — is likely to occur again, and any behaviour that leads to an “annoying state of affairs” is less likely to occur again.

Or in other words:
The likelihood of the occurrence of a behaviour is influenced by its consequences.

Question 14

What is Thorndike’s puzzle box? What did it indicate?

Answer 14

This puzzle box had a window to a bowl of food that the cat could see. There was then a lever attached to a string which opens the door of the box.

When first placed in there, it will take a lot of time for the cate to get out, because they don’t know how. Only once they accidentally step on the lever will the door open.

So the second time they are put in the box they will realize that this lever opens the door, and therefore this is a pattern that their brain recognizes. It’s conditioned learning, the cat wants the food, and the lever gets them to the food. So the UR is getting to the food, and the lever is the CR1 so they realize this will open the door which will give them food.

So over time, the escape time drastically lowers. However, if they take a break and come back they will forget, but as soon as they escape and try again the response time will lower faster then the first time, since they will remember quicker.

Question 15

What did B.F. skinner do as an experiment to elaborate on thorndike’s law? What theory did he develop?

Answer 15

B.F. Skinner invented the operant chamber, which was just a more complicated version of the puzzle box from before. It had a lever, signal lights, a speaker, a shock generator and an electric grid that the animal sits on, and a pellet dispenser that goes into the box.
So over time, the rat would learn that the lever disperses the pellets, but maybe something else creates a loud noise that they don’t want, and another button produces a shock which they don’t like, and based on the consequences they will either continue with that response or not.

He developed the theory of operant conditioning which was based off of Thorndike’s ideas.

Question 16

What is the difference between classical and operant conditioning?

Answer 16

Classical conditioning: A type of learning where a neutral stimulus acquires the ability to evoke an originally UR (involuntary or reflexive), which was originally NATURALLY evoked by another stimulus.

Operant conditioning: A form of learning where voluntary responses come to be controlled by their consequences. So this was based on thorndike’s belief that the likelihood of the occurrence of a behaviour is influenced by its consequences, but this theory was created by skinner.

Question 17

What are the two main categories of response to an action and consequence?

Answer 17

You either will do that action again due to a good consequence, or will not do it again due to a bad consequence.

Question 18

What are the 4 subcategories (2 in each) within these responses? (Think about what the action is and the corresponding response)

Answer 18

Positive reinforcement:
GET SOMETHING GOOD, WILL DO THIS AGAIN.

Negative reinforcement:
SOMETHING BAD IS REMOVED, WILL DO THIS AGAIN.

Positive punishment:
GET SOMETHING BAD, WILL NOT DO THIS AGAIN.

Negative punishment:
SOMETHING GOOD IS REMOVED, WILL NOT DO THIS AGAIN.

Question 19

What does the word POSITIVE indicate for learning? What about punishment? NEGATIVE? Reinforcement? How to remember these 4 subcategories of consequence and response?

Answer 19

POSITIVE = you GET something due to your action.

NEGATIVE = something gets TAKEN AWAY due to your action.

REINFORCEMENT = You WILL do that action again, because the consequences reinforced your decision and the subsequent action that occurred.

PUNISHMENT = you WILL NOT do that action again, because you were punished and so your decision was NOT reinforced.

To remember these 4 categories, just know if YOU GET SOMETHING = POSITIVE. If SOMETHING IS TAKEN AWAY = NEGATIVE.

If the consequence is GOOD = REINFORCEMENT (the consequence reinforces your action and will cause you to do it again).

If the consequence is BAD = PUNISHMENT (the consequence harmed you and hence will cause you to not do that again).

Question 20

Punishment will always _________ ones behaviour, because the consequence was _______.

Reinforcement will always ________ ones behaviour, because the consequence was _________.

Answer 20

Punishment will always DECREASE one’s behaviour because the consequence was BAD.

Reinforcement will always INCREASE one’s behaviour because the consequence was GOOD.

Question 21

Using Skinner’s box, what would be an example of a response in each of these 4 categories?

Answer 21

Positive reinforcement: the animal presses the lever, food comes into the chamber. They got something good, so this reinforces their actions and causes them to act in the same way again.

Negative reinforcement: Push the lever and it turns off bright lights and loud speaker that harm the animal. Something bad was taken away, so their actions were reinforced by this good consequence and they will do it again.

Positive punishment: Pull the lever and it electrifies the grid — causes a shock. This will harm the animal and hence something bad was added, punishing them and causing them to not do it again.

Negative punishment: food is continuously coming in, until you pull the lever then it stops. Something good was taken away so you don’t want to do that again and your behaviour in that manner decreases.

Question 22

In a disciplinary procedure — disciplining kids, do you use positive or negative punishment?

Answer 22

PP = add something bad
NP = take away something good

Positive punishment would be like corporal punishment, which is harming the child with physical abuse. This is illegal.

But a mild form of physical punishment is not effective because its consequence will not overtake the joy of what the person is violating.

So in this case negative punishment (take away valuable privileges) will be the best.

Question 24

What are the three things to consider in effectively using punishment (acronym)?

Answer 24

Timing — apply immediately after behaviour so brain associates action with that punishment (similar to classical conditioning).

Intensity — intensity of punishment should match or surpass the intensity of the misbehaviour.

Consistency — have to do it every time there is a misbehaviour, or there will be a weak association between the punishment and the misbehaviour.

Acronym:
TIC

Question 25

Where is positive reinforcement most commonly used? What are the two main types of positive reinforcement?

Answer 25

Positive reinforcement is commonly used in drawing customers, by saying if they buy a certain amount they will get rewards. The schedules of reinforcement would be continuous (every time they buy something they get a reward) or partial (sometimes they get a reward, sometimes they do not).

Question 26

What are the two and then 4 subcategories of partial positive reinforcement?

Answer 26

The two subcategories of partial reinforcement are interval and ratio, meaning the reinforcement is given at each interval, or it is given based on the ratio of things bought or ratio of people present. Within interval, there is fixed intervals, and there is variable intervals. Within ratio, there is fixed ratios (buy 7 get one free) and there is variable ratios (or a lottery), where a certain ratio of people will get reinforcement but the people getting this changes each time. For lottery, a random amount of responses must be made before reinforcement occurs.

Question 27

What does a graph of time vs cumulative number of responses look like for each of these 4 schedules of partial reinforcement? What do the lines coming off of the line graphs indicate? Which has the steepest slope, which has the shallowest?

Answer 27

Overall, ratio reinforcement is the least predictable and hence motivates people to respond the most because they never know when he reinforcement will be. For ratio it is also strongly linked to the number of responses, linking behaviour with reward, rather then time. The steepest slope (fastest growing cumulative amount for responses) is the variable ratio reinforcement. This is also known as lottery, and this is because each time your chance to win varies but is unknown, and this unknown factor creates a strong anticipation of reward, motivating consistent engagement in the behaviour . Then it would be the fixed ratio, as ratios are always more unpredicatable then intervals since they don’t rely on time. Then it is variable interval. Lastly fixed interval. Then the two variable ones will have straight lines, (the short horizontal lines are just the times at which you apply the reinforcer). For the variable ones people don’t know when the next reinforcement will be so the response is constant, anticipating when that will be. The fixed ones will have larger responses at points close to and leading up to the reinforcement, as they know when it will be so they response accordingly. For fixed ratio it is slightly more constant thought because that is required to produce the reward sometimes. Fixed interval will have the most drastic exponential rises close to the reinforcer.

Question 28

What is extinction? What response is the easiest and quickest to make extinct (out of the 5 reinforcement methods?)

Answer 28

Extinction: in operant conditioning this is the gradual weakening and disappearance of a response tendency because the response is no longer followed by reinforcement. This is also the same for classical conditioning, when the conditioned response no longer produces a certain reward. The response with the quickest rate of extinction is continuous reinforcement, because people will notice instantly that they are nor receiving their reward. Then fixed will be quicker extinction then variable! Because fixed is known so you are less willing to wait around! Then fixed interval will be the next fastest since you will realize as soon as that interval passes that you are no longer being rewarded. The longer the interval the more resistance to extinction, once again because it will take longer to realize no reward is coming. Next will be fixed ratio because again you are expecting when you will receive your reward and if you don’t receive it then you will know it’s over. But the higher the ratio the more resistance there will be to resistance because it takes longer to go through that many trials until you realize there will be no reward. Then variable interval will be next because you don’t know how long that interval is, and only once it has been a very long time will people stop acting that way because they have less and less faith in receiving a reward. Lastly variable ratio will have the most resistance because this is not based on time and its just chances, so it could take a very long time for people to realize the reward is gone.

Question 29

What is the difference between primary and secondary reinforcers? What is the special generalized reinforcer?

Answer 29

Primary reinforcers: Things that are rewarding on their own, and without learning produce the unconditioned response that they are satisfying, and these are primary reinforcers. So getting food when hungry is a primary reinforcer, it immediately satisfies you. Conditioned reinforcer (learned like CS) or secondary reinforcer: Is a stimulus that gains reinforcing power by being associated with a primary reinforcer. So this is like the conditioned response. Good grades and money are both things that are associated with primary reinforcers — they can get us the innately satisfying things that we crave. Money is the special generalized reinforcer because it can be traded for about any primary reinforcer.

Question 31

What is shaping? What is it mainly used for?

Answer 31

Shaping: The reinforcement of closer and closer approximation of a desired response, so rewarding them with small increments of change until they finally get to the desired response. This used for animal trick training, because to get them to do a big trick you teach them in small increments.

Question 32

What is observational learning?

Answer 32

Observational learning: People learn how to act by watching others, especially children will learn from the way that the adults act. For example, children in one group watched adults playing with the Bobo doll, one group had adults ignore that doll and play with toys, and one group had not adult. The children did whatever the adults did, and were less aggressive to the bobo doll if the adults wasn’t playing with it.

Question 33

What is the Dunning-Kruger effect?

Answer 33

The dunning-Kruger effect is a sense of false confidence that happens when you are learning something for the first time, because you learn the surface and think you know it all. But as you learn more and more you become less confident until you become an expert, and then you are (still less then the beginning) but more confident.

Question 34

What is illusory superiority?

Answer 34

This is a form of false confidence similar to what you believe when you are at the beginning of learning something. These people form a halo around themselves and think they are above average in everything, as they are blind to their shortcomings.

Question 35

What does a graph of the actual performance percentile vs percentile, with lines for perceived mastery of material, perceived test performance and actual test performance look like?

Answer 35

This graph will be increasing from left to right, but the difference between actual test performance and perceived mastery of material will be very larger in the bottom percentile of performance, because again when you are a novice you don’t know enough to know how much you don’t know. Then in the top percentile, your test performance will actually be higher then what you believe, because as an expert you realize that there is so much that you don’t know, and this makes you worried you don’t know anything on the test.