Learning Exam #2

• Classical vs. Instrumental Conditioning

• Classical conditioning:

  • Learning associations between stimuli

    • Example: A bell meant food was coming

• Instrumental conditioning:

  • Learning associations between stimuli, response,and outcome, leads to goal-directed behavior 

Example: Brush teeth, no cavities

E.L. Thorndike (1874–1949)

• Used puzzle box:

  • Food-restricted cats

  • Goal: escape and food

  • Measured latency to escape 

• Cats learned association between stimulus and response:

  • Association affected by outcome

  • Rope in box (S) + pulling rope ® —> escape (O)

Law of effect:

• Response to a stimulus followed by a satisfying event strengthened S-R relationship

• Pull rope—> food and escape —> less time to escape

• Response to a stimulus followed by an annoying event —> weakened S-R relationship

  • Ex: Pull rope in the box—> shock —< longer time to escape 

1. Discrete-Trial Approach

   1. Response is performed once 

   2. behavior of subject terminates the trial

   3. timing of trial determined g the experimenter 

Two Maze types:

• Straight alley

• T-maze

Behavior Measured:

• Running speed

• Latency to move from start box (S)

• Choice behavior: T-Maze only

2. Free-Operant Approach 

   1. Subject is “free” to respond at any time

   2. May be repeated many times

   3. Timing responses determined by subject 

Free-Operant Approaches: Skinner Box

• Skinner box measures operant response rate:

  • Example: # of lever presses for food

• Operant response: Behavior that “operates” on the environment

Advantages of free-operant? 

Magazine Training

How to produce a target response 

• Step 1: Magazine Training: Classical Conditioning

  • Sound of the “magazine: (aka, food dispenser) is a CS+, followed by a food US that orients organism

• Examples?

• Step 2: Shaping

  • Instrumental Conditioning

  • Rewarding successive approximations to target behavior by:

    • Reinforcing closer actions to correct response

    • Not reinforcing early non-target responses

  • Correct step[s are preserved

    • Ex: the Rat still tears up to push the bar 

SHAPING

• Shaping familiar responses:

  • Must be a variable response

  • Slowly step-up criteria

  • Can bring about SUPER or miniature responses

    • Deich, Allan, & Ziegler (1988): Training beak openings in pigeons 

Terminology:

• A response may produce 1 of 2 outcomes:

  • Appetitive stimulus:

    • A pleasant outcome (getting paid, food, sunshine)

  • Aversive stimulus:

    • Negative outcome (yelling, shock, cold)

• Contingency: something likely to occur because of something else 

  • Positive contingency: response turns on/ causes an outcome 

    • A rat can press the lever to gate food

  • Negative contingency: response turns off/ inhibits an outcome

    • A rat can turn off a loud by pressing the lever

Appetitive Stimulus              Aversive Stimulus 

Positive Contingency —Response causes outcome

Negative Contingency—Response inhibits outcome 

Positive Reinforcement: 

• Positive contingency between response and appetite stimulus (response turns on a good thing)

• Outcome: Increased responding

• Examples:

  • Dog training

  • Good grades

  • Paid by doing a job

• WHAT ELSE?

Punishment: 

• Positive contingency between response and aversive stimulus (response turns on a bad thing)

• Outcome: DECREASED RESPONDING

• Examples:

  • Tickets for speeding

Omission Training:

• Negative contingency between a response and an appetitive stimulus (response removed an appetite stimulus)

• Outcome: Decreased responding

• Examples:

  • Swearing leads to loss of TV

  • Autoshaping reversal 

Negative reinforcement:

• Negative contingency between response and aversive stimulus (response turns off or avoids a bad thing)

• Outcome: Increased responding 

• Types: 

  • Escape: aversive stimulus present at the time of behavior, stopped by response

  • Avoidance: aversive stimulus scheduled to happen but it is prevented from happening by response 

  • Examples:

    • Putting on sunscreen

Barton, Bull, & Rep (1986):

Differential reinforcement of other behaviors (DRO)

• Subjects: students with Autism Spectrum Disorder

• Behaviors:

  • Hand flapping-decrease

  • Other behaviors-increase

• Reinforcers: Yummy snacks

• What types of instrumental conditioning are used 

• What’s the response?

  • Drinking a Four Loko

• What is the contingency?

  • Positive contingency 

• Is the outcome reinforcing or inhibiting?

  • Reinforcing

• What procedure is used?

  • POSITIVE REINFORCEMENT

• What is the response?

  • Running And hiding

• What is the contingency?

  • Negative contingency

  Is the outcome reinforcing or inhibiting?

  • Reinforcing

• What procedure is used?

  • Negative reinforcement (ESCAPE)