PS261 Lecture Notes - Lecture 7: Operant Conditioning, Reinforcement, Contiguity
Document Summary
Learning which behaviors to emit (which one"s lead to appetitive outcomes) occurs by trial and error. Behaviours which are rewarded become more likely to be repeated. Puzzle boxes: cats initially tries many random behaviours, accidentally opens box, behaviour that led to box opening now more frequent, box is opened faster and faster. In presence of s1 (light): r1 (lever) leads to o1 (water, r2 (button) leads to o2 (food) In presence of s2 (sound): r1 (lever) leads to o2 (food, r2 (button) leads to o1 (water) The response depends on the s and the o: not simply s-r learning, operant learning involves the outcome (o) as well. Reinforce can be: appetitive: rewarding, aversive: punishing. Operant (response) can be: positive: increases probability of reinforce, negative: decreases probability of reinforce. Belief in a (probably) incorrect correlation: breaking the mirror (r) 7 years back luck (o) Matching to samples (mts: sample presented (green, subject rewarded for pecking same color.