Learning (4-5) Flashcards
How does a stimulus control operant behaviour
A response is reinforced, it really becomes more likely to occur again in the environment, or stimulus context, in which it was reinforced
What happens if you change the stimulus context
Changing the probability of the response
- The behavior changes in a systematic way - this Is what stimulus control means:
Respondent (classical, Pavlovian) conditioning
Respondent conditioning, or classical conditioning, is learning through association between two stimuli.
Key Components:
Unconditioned Stimulus (US): Naturally triggers a response (e.g., food causes salivation).
Unconditioned Response (UR): Automatic response to US (e.g., salivation to food).
Neutral Stimulus (NS): Initially does not trigger UR (e.g., bell).
Conditioned Stimulus (CS): NS becomes CS after being paired with US (e.g., bell after pairing with food).
Conditioned Response (CR): Learned response to CS (e.g., salivation to bell).
Process:
Before Conditioning: US (food) → UR (salivation); NS (bell) → No response.
During Conditioning: NS (bell) + US (food) paired repeatedly.
After Conditioning: CS (bell) → CR (salivation).
Describe how decremental gradient is shown on a Generalization gradient
- Identify the Gradient of responses, focusing on the peak at the training stimulus (S+).
- Observe Generalization to similar stimuli to S+, the amount of generalisation depends on how close the stimuli is to S+
- Responding decreases progressively as stimuli become more different from the training stimulus.
- The less generalisation, the more stimulus control over the behaviour
- More stimulus control = more discrimination (behaving differently in the presence of different stimuli)
- Usually obtained when the training stimulus was associated with reinforcement (i.e., an S+). Also called a positive or excitatory gradient.
What does stimulus control mean?
The extent to which stimuli that comes before operant behaviour, affects the rate or probability of that behaviour
What is Generalization gradient
it displays how much does the original trained behavior generalization new stimuli
What is Decremental gradient
Responses decreases as you move away from the training stimulus
The ABC of behavior is
Antecedent: Something that happens in the environment before a behaviour, which signals that a behaviour is more or less likely to contact a consequence (reinforcement or punishment)
Behavior: response to antecedent
Consequence: reinforcer
Concept of formation ad case study example
- Grouping related stimuli together in classes
Example: Can animals group related stimuli together into classes
-** Herrnstein and Loveland - can pigeons learn the concept of “human”**
○ Reinforce response to pictures contain humans
○ Don’t reinforce response to picture without humans
○ Then test with new pictures, not seen before. Unseen pictures will be introduced to see if pigeons can identify human presence without prior exposure. **The null hypothesis suggests pigeons don't understand humans but apply learned rules to familiar stimuli. **Success in identifying humans in new pictures would challenge this hypothesis, indicating pigeons comprehend the concept of humans beyond memorized rules. ○ Evidence for concept of "human" if more. Reponses to slides contain humans. ○ Pigeons were very good at this, better than the experimenters ○ Pigeons can form the concept of humans
Concurrent schedules are + example (generic)
Have two simple scheduled available simultaneously, or both at the same time.
Vary the schedules, and therefore the rate of reinforcement, on each alternative and measure choice
Example
VI= variable interval
VI 30s (left) VI 60s (right)
The average time for reinforcers is 60 secs for right
1 reinforcer
The average time for reinforcers is 30 seconds for left
2 reinforcer
The individual would pick the left over right
Would respond twice as much on the left than on the right
Higher rate of reinforcement
What is the matching law
In a concurrent schedule, the proportion of responses that an organism will emit on one alternative is equal to (matches) the proportion of reinforcers it obtains from that alternative. As an equation, where B means responses (behaviour) and R means reinforcers:
Not perfect at describing choice