5. Classical Conditioning Flashcards
problem
classical conditioning is likely to depend on changes in synaptic plasticity - but there are trillions of synapses in the brain so where do we look?
- need to make problem simple
how do we make understanding learning simple?
pick a simple form of learning - classical conditioning - hoping that there are fewer synapses involved
- gives us a chance to identify them
- understanding simple behaviour could help us understand more complex
unconditional stimulus (US)
(food/airpuff) produces the unconditioned response (UR)
- not learnt
- before training
- e.g. salivation/shut NM
conditioned stimulus (CS)
paired with the US (bell/tone)
- initially produces no response or an irrelevant response
- eventually leading to conditioned response (CR)
conditioned reflex/conditioned response (CR)
CS produces conditioned response (after training)
e.g. tone produces salivation without the presence of food
eye blink conditioning
US is usually a puff of air into the eye (periorbital shock)
- CS is usually a tone
- UR and CR are movements of eye lid to protect eye
nictitating membrane response (NMR)
some animals have a third eyelid
- this can also be conditioned but the animal has no voluntary control over its movement
- very low level of spontaneous activity
- SIMPLE
what happens in NMR
day 1 - NM stationary until US (air puff)
day 2 - NM moving before US = CR
day 5 - NM moves well before US, shutting to protect eye preemptively
eye blink conditioning features
- US overlaps with CS = delay conditioning - gap between the end of the CS and the start of the US (trace conditioning) - more neural structures = delay
- CR does not effect US = closing the eye has no affect on the US, otherwise its avoidance learning
forebrain?
Mauk and Thompson (1987)
- delay in NMR conditioning still possible in rabbits lacking either hippocampus or cortex
- also possible with rabbits who have forebrain separate from brainstem and cerebellum (decerebrate)
brainstem and cerebellum?
Thompson (1983)
- technique 1 = electrophysiological mapping during conditioning.
- recorded from multiple units in cerebellar cortex and deep cerebellar nuclei
- unpaired control condition in which CS’s and US’s are presented as frequently as in training condition but unrelated (no neural response)
- when paired, units increase responding as CR develops
- but recording does not establish a causal role (need lesions)
- technique 2 = first large lesion (through cerebellum and underlying brainstem) of entire cerebellum plus output (deep cerebellar nuclei)
- CR abolished and cannot be relearnt
- similar results when study focused on deep cerebellar nuclei
where are lesions in the deep cerebellar nuclei successful at abolishing CR
Yeo et al (1985)
- lesions confined to the anterior portion of deep cerebellar nucleus (anterior interpositus nuclei) are effective
hippocampus?
Weiskrantz and Warrington (1979)
- patients with anterograde amnesia (damage to hippocampus) showed clear evidence of learning even though patients couldn’t remember apparatus
Gabrieli et al., (1995)
- verified damage to medial temporal lobe (contains hippocampus) compared to control
- rates of conditioning the same for both groups
Squire et al (1993)
- LTM is divided between conscious memory for facts and non-conscious memory (skill, habit learning, classical conditioning and learning)
- does not depend on hippocampus and surrounding tissue
lesions: behavioural effects
- after lesioning there is selective loss of the CR, but UR not affected
- lesion does NOT produce a simple motor deficit
- effects of unilateral lesions are unilateral, can do it to the other eye and get the same response (Yeo et al., 1987)
its loss of memory trace
lesions dont cause: (Thompson, 1983)
- paralysis = UR still evident (e.g. NM shutting)
- deafness = unilateral lesions affect conditioning of ipsilateral eye
- loss of conditioned response appears to be genuine loss of the memory trace (the engram)
