Classical and Operant Conditioning

Learning Definition

  • Learning: A relatively permanent change in behavior brought about by experience or practice.

    • Relatively permanent: You can unlearn something as well as learn it.

Classical Conditioning

  • Scenario:

    • Instructor turns off the light before lecturing.
    • Students associate the light going out with an unpleasant lecture.
    • Student groans when girlfriend turns off the light outside of class.

  • Pavlov's Discovery (Early 1900s):

    • Ivan Pavlov (1849-1936) - Physiologist measuring salivation in dogs.
    • Dogs salivated before food appeared, associating footsteps with food.
    • Example: Dog/cat gets excited when you open a cupboard or a can of food.

Classical Conditioning Structure

  • Unconditioned Stimulus (UCS): Naturally occurring event.
  • Unconditioned Response (UCR): Naturally occurring response to the UCS.
  • Neutral Stimulus (NS): Does not initially elicit a response.
  • Conditioned Stimulus (CS): The neutral stimulus after pairing with the UCS.
  • Conditioned Response (CR): The response elicited by the conditioned stimulus (same as UCR).
  • Key elements:
    • Reflexes
    • Associations/Pairings (e.g., bell with food)
    • Contiguity (things next to each other)

Taste Aversion

  • Example: Eat something, get sick, then feel nauseous even smelling/thinking about that food.

Pavlov's Experiment

  • UCS: Food
  • UCR: Salivation
  • NS: Bell
  • Pairing: Bell with food
  • CS: Bell
  • CR: Salivation

Watson and Rayner - Little Albert Experiment

  • John Watson and assistant Rosalie Rayner conditioned fear into a 9-month-old baby (Little Albert).

  • Unethical experiment (causing fear).

  • Watson and Rayner had an affair and he was booted out of academia. He then did ads in New York using classical and operant conditioning.

  • Procedure:

    • Exposed Albert to furry animals (rat).
    • Banged a hammer on a pipe when presenting the rat.
    • Albert showed fear.

  • Analysis:

    • UCS: Clanging sound
    • UCR: Fear
    • NS: Furry rat
    • Pairing: Clanging with the rat
    • CS: Rat
    • CR: Fear

Generalization and Discrimination

  • Generalization: Little Albert feared all white and furry things.
  • Discrimination: Albert could tell the difference between a rat and a Santa Claus mask.
  • Experiment deemed not credible because:
    • Ethics: unethical to cause a child fear.
    • Data: Watson has no real observable data that he collected.
    • IRB (Institutional Review Board) ensures ethical experiments.

Acquisition

  • How behaviors are acquired.

  • Types of Acquisition Trials:

    • Delay: Present the neutral stimulus, then the unconditioned stimulus (best approach).
    • Simultaneous: Present food and bell at the same time.
    • Backward: Ring the bell, then give food (doesn't work well).

Extinction and Spontaneous Recovery

  • Extinction Trials: Ring the bell many times without food; the response diminishes.
  • Spontaneous Recovery: The behavior reappears after extinction (bell rings, child gets fearful).
  • Pavlov believed fear fades but doesn't disappear because there is a neural network in the brain.

Higher Order Conditioning

  • Also called secondary conditioning.
  • Adding another stimulus or a second conditioned stimulus.
  • Example: Child watches TV, garage door opens, parents come home.
    • TV show = excitement, garage door = excitement.

Limitations of Classical Conditioning

  • Doesn't always work; free will.

  • Instinctive Drift (Brelands): Instincts interfere with learned behaviors.

    • Example: Raccoon trained to put a coin in a piggy bank Commercial, but instinct to dip their food in the water took over for a bank commercial.

Classical Conditioning Recap

  • Identify UCS, UCR, NS, CS, CR in the high school classroom scenario.
  • Any stimulus an organism can perceive is capable of eliciting any reaction the organism is capable of making.
  • Classical conditioning is a reflex.

Operant Conditioning

  • You act on your environment and get feedback, which determines your behavior.

Instrumental Conditioning (Edward Thorndike)

  • Forerunner to operant conditioning.

  • Probability of a response based on its consequences.

  • Random trial and error.

  • Law of Effect:

    • Positive result = stamped in.
    • Punishment = stamped out.
    • How your brain is wiring itself.

  • Cats in puzzle boxes

    • The cat was in the box, and it wanted to get the food. So, it had to push something, anything to make that happen.
    • Trial and error, meowing a lot, pushing this or that, scratching on the side.

BF Skinner

  • Everything we do is determined by our history of rewards and punishments.
  • Operant conditioning: I operate or act on my environment and get feedback, which determines your future behavior.
  • Skinner called it the science of behavior - behaviour is predictable.
  • Walden Two (utopian community based on behavioral principles) and Beyond Freedom Dignity (theories) talk about no autonomous man, no free will.

Skinner Box (Operant Box)

  • Rats or pigeons could push on things.
  • Aversive types of stuff in it (shocks)

Skinner's Baby Tender

  • Temperature and humidity regulated.
  • Crank to change the sheets.
  • Skinner raised his own child in it.
  • Rumors that his child had psychological issues are urban legends.

Basics of Operant Conditioning

  • Discriminative Stimulus: Signals that reinforcement is available (e.g., green light).
  • Primary Reinforcers: Unlearned reinforcers needed for survival (food, water).
  • Secondary Reinforcers: Conditioned reinforcers, learned (money).
  • Accidental Reinforcement: Superstitions (lucky tie).

Types of Reinforcement

  • Enhance likelihood of repeating behavior:

    • Positive Reinforcement
    • Negative Reinforcement

  • Cause cessation of behavior:

    • Punishment

Positive Reinforcement

  • Getting a star, reward, trophy, good job.

Negative Reinforcement

  • NOT punishment.
  • Avoid/escape aversive event (unpleasant).
  • Example: Fire alarm goes off, you leave the room.
    • Example: Getting rid of someone who is annoying by actively ignoring them.

Punishment

  • Punishment by Application: Something unpleasant happens (parent yells).
  • Punishment by Removal: Taking away something liked (ball).

Examples

  • Positive Reinforcement: Works well with training animals.

  • Negative Reinforcement: Getting rid of something unpleasant.

    • Wiping hands with towel
    • Turning on windshield wipers
    • Pushing 'stop' on alarm clock
    • Spraying cat with water

  • Punishment by Removal: No more TV.

  • Punishment by Application: Soap in mouth.

Schedules of Reinforcement

  • Continuous: Reinforce every behavior (extinguishes quickly).

  • Ratio: Number of behaviors.

    • Fixed Ratio: Set number of behaviors (every quiz counts double).
    • Variable Ratio: Never know when you'll be reinforced (slot machines).

  • Interval: Time.

    • Fixed Interval: Fixed set of time (paid every two weeks).
    • Variable Interval: Never know when it's going to happen (fishing).

Shaping

  • Successive approximations.
  • Steps to get a person or animal to perform a total behavior.
  • Shaping the behavior of swimming: You're gonna reinforce you when you breathe right. I'm gonna reinforce you when you do your arms right. I'm gonna reinforce you when you do your legs right.
  • Once pigeon turns a little, it's reinforced. Turns a little more. Turns a little more
  • Call this chaining.

Premack Principle

  • Do something unpleasant to get a positive reinforcer.
  • Example: Mow the lawn, and you get to use a car.
  • Pigeons reading
    • The task was to isolate an individual piece of behavior and see how that could be changed.

Social Media Operant Conditioning

  • Advertisers spent an estimated 31,000,000,000 (31 Billion) on social media in 2016.
  • Cell phones are slot machines operating on a variable schedule reward.
  • Snapstreaks: Affecting kids' feelings of obligation to each other because they feel they have to respond and it maximises amount of time spent.