13 - Classical Conditioning

Question 1

Explanation of fixed action patterns and habituation

Answer 1

• Preexisting mental links (associations) between certain stimuli and responses
• Unlearned association – experiencing stimulus automatically triggers response

Question 2

Unconditioned stimulus and its motivational value

Answer 2

• Another example of stimuli that can automatically elicit unlearned unconditioned responses because of a pre-existing, unlearned association
• USs often elicit unconditioned responses because they have motivational value (they are nice or nasty)

Question 3

Unconditioned response

Answer 3

• URs can be like a fixed action pattern – a type of involuntary reflex
• They can overt muscular responses (approach/avoid) or internal (like/ dislike responses)
• Different species have different USs and Urd evolved to be useful – anticipate US

Question 4

Preexisting and acquiring associations

Answer 4

1. Pre-existing mental links
   - Fixed action patterns
   - Unconditioned stimulus (US) – unconditioned response (UR) links
2. Acquired/learned mental links
   - Classical conditioning: SS
   - Instrumental conditioning: RS

Question 5

What happens during classical conditioning?

Answer 5

• Cough followed by tickle
• Tickle is a US because already linked to a UR
• Cough is a Conditioned Stimulus CS – it isn’t already linked to a UR
• When cough or tickle perceived, patterns of mental/neural activity occur
• This activates mental representations of those events - correspond to perceiving/thinking about the event
• Their two mental representations become simultaneously activated and mentally associated
• Now hearing cough excites its mental representation
• This activates mental representation of tickle via the association between them
• So CS presentation now indirectly elicits UR
• This is the conditioned response – CR
• Now when child hears cough, she remembers the tickle vividly, which makes her squirm and giggle

Question 6

Classical conditioning is NOT S-R learning

Answer 6

• Sometimes textbooks and online resources, and maybe even your lessons in school, give the impression that classical conditioning is SR learning.
• This is not true. CR not directly associated to the CS
• How do we know? Thought experiment
• You now tickle the child constantly until they are exhausted, their stomach hurts, and they don’t want you to do it any more
• The child won’t giggle any more.
• Why? because tickling US has changed – has become uncomfortable now – doesn’t make them giggle anymore!
• Classical conditioning is S-S learning. Cough activates tickle, but this is now unpleasant, so response is different
• If classical conditioning were a direct SR link
• They would still giggle, regardless of the tickle no longer being rewarding
• Dog would still salivate despite not being hungry

Question 7

Form of conditioned response (CR)

Answer 7

• Conditioned response is indirect elicitation of UR
• what CR looks like depends on nature of UR and also on what CS is
• CR can resemble UR closely (which can make animal look a bit stupid) – but can also be more sophisticated
• Reflects knowing that the US will happen
• Compare key peck CR after AutoShaping; US is GRAIN or WATER
• Pigeons eat grain with open beak and closed eyes, but drink water with closed beak and open eyes
• AutoShaping = classical conditioning procedure e.g. pair keylight with food; CR is to peck. Pigeon ‘shapes itself’ in sense that pecking happens spontaneously

Question 7

Stimulus substitution

Answer 7

• CS makes you think of US – you expect it CS acts as substitute for the US
• Explains form of CR – it’s like UR
• Pigeons peck grain; when keylight paired with grain, they peck keylight
• But what if CS is a tone? if you pair a tone with food pigeon cannot peck it; but it knows food is coming – becomes more active, flaps about
• Physical (sensory) properties of food affect CR

Question 8

Stimulus substitution also applies to US value

Answer 8

• Pigeons will also approach a signal for food – remember the bird in the long box? and they will also work to obtain it
• Motivational properties of US transfer to CS
• E.g. an unconditioned response for nice food is liking these transfers to the keylight – likes the keylight, will approach it, work to obtain it
• This is called evaluative conditioning

Question 9

Advertisements: classical conditioning

Answer 9

• Pair branding of a product with motivationally positive images
• Companies hire associative learning experts to exploit this

Question 10

Chains of associations: second order condition

Answer 10

a) Purple wrapper = chocolate & sweet taste (CS1 = US)
b) Box = purple wrapper (CS2 = CS1)
- = I want that box!
- Our world is full of things that were once neutral, but we learned to value positively or negatively through associations
- Chocolate are USs that are nice or nasty
- Purple wrappers predict those USs – CS1, these CSs have now acquired motivational value
- Chocolate box predict those CSs –– CS2, now house and chocolate box have motivational value too

Question 11

Why do people think classical conditioning is S-R learning?

Answer 11

• The fault of psychologists like Hull & Thorndike
• Hull thought it important only to study observable things. Learning had to be between visible S and visible R. He argued all learning was SR. Thorndike said to form an S-R association you need a reinforcer to stamp it in, but not to be learned about itself. So they thought all learning requires a reinforcer and a response, but is between stimulus and response
• Now we know learning can happen without a response or a reinforcer – but how do we know if there is no response?
• We can see it if we test in the right way:
• Need a motivationally valuable US to see learning

Question 12

Classical conditioning without reinforcement or response

Answer 12

• Don’t need motivationally significant US to learn!
• Hull was wrong – don’t need reinforcer to stamp in learning
• Shown by a variation of second order conditioning – sensory preconditioning
• You have never tried chocolate. But you know that the purple sweets are in the Quality Street box. Then one day you try one. It’s delicious! Now you like the purple wrapping and the Quality Street box
• You associated sweets and box, when neither had value for you, and you didn’t behave differently. But once they acquire value the hidden learning is revealed

Question 13

Sensory preconditioning: classical conditioning without reinforcement or response

Answer 13

• Tone predicts light, box with purple sweet; you see no response.
• Then pair light with food/purple sweet with chocolate.
• Now rat likes the tone, looks for food during tone – tone predicts food
• It learned in the first stage despite making no response
• Now you will buy the box, even though you weren’t bothered before
• In sensory preconditioning pair two neutral things, in second order conditioning you don’t

Question 14

Methods of studying classical conditioning

Answer 14

• Animal experiments used Skinner boxes and rats or pigeons
• Pair tones with lights with shock or food
• Measure conditioned response: food experiments (appetitive reinforcer); CR is searching for food in food hopper

Question 15

Measuring appetitive conditioning

Answer 15

• With conditioned approach, the learning graph goes UP
• measure is number of head entry responses per minute (rpm) during CS (often compared to number before CS)
  a) Tone  food
  b) Click  no food
• Click line increases even though click never followed by food (generalization)

Question 16

Measuring aversive conditioning

Answer 16

• Shock experiments (aversive reinforcer). CR is freezing – hard to measure directly, so often measure suppression of lever pressing (can’t lever press while freezing)
• Conditioned emotional response procedure (CER)
• Suppression ratio: rate CS / (rate before + rate CS)
• Before training:
• 20 responses per minute (rpm) before CS
• 20 rpm during CS – the same! rat couldn’t care less
• Ratio = 20/(20+20) = 0.5
• After training:
• 20 responses per minute (rpm) before CS
• 5 rpm during CS – less! means rat is slightly afraid
• Ratio = 5/(5+20) = 0.20
• Minimum ratio = 0: no responses during CS – very afraid
• With conditioned suppression, the learning graph goes DOWN
• Suppression ratio: rate during / (rate before + rate during)
  a) Tone  shock
  b) Click  no shock
• Again we can see generalisation between tone, that’s frightening, and click that isn’t

Question 17

Extinction: removes US

Answer 17

• What happens when you take away the US? CR goes away
  a) Tone = nothing, Click = nothing
• CR slowly dissipates
• Spontaneous recovery: if you wait and test again, the CR returns (but never to its original level)
• Extinction involves removing the US – so the CS that used to predict the US doesn’t anymore. CS now predicts the omission of the expected US
• This does not eliminate the original learning – it produces additional inhibitory learning that counteracts the excitatory learning
• inhibitory association more disrupted e.g. by passage of time, so excitatory association can show and CR returns

Question 18

Conditioned inhibition

Answer 18

• You really like those purple Quality box sweets, and eat them whenever you have the chance
• But your friend has a very old box of the same sweets, and when you eat the purple ones, they taste weird and not very nice
• Purple sweet  lovely chocolate
• Purple sweet from the old box  not nice at all
• This particular box predicts the omission of the nice chocolate you expected
• A conditioned inhibitor signals the omission of the US
• Tone  food tone+light  no food, light signals absence of expected food
• Tone EXCITES mental representation of food… light INHIBITS it
• Tone makes you expect food… light counteracts that expectation

Question 19

Properties of inhibitors

Answer 19

• Inhibitors can also have motivational value
• If you are hungry and expected food and didn’t get it, that can make you sad…
• But if you expected something bad that didn’t happen, that will make you happy…
• Motivational states can be positive or negative