Comprehensive Guide to Operant Conditioning and Reinforcement Schedules

Refresh: Classical vs. Operant Conditioning

  • Classical Conditioning Recap:

    • Elicited Response: The response is drawn out automatically, functioning like a reflex.

    • Autonomic Nature: These are involuntary responses (e.g., salivation, heart rate, sweaty palms).

    • Pavlov’s Dogs: Dogs did not choose to salivate; it was an autonomic response. Over time, they associated the sound of a metronome with the arrival of food.

    • Reward/Stimulus Timing: In classical conditioning, the reinforcement (food) simply arrives after the stimulus (sound), regardless of the subject's actions.

  • Operant Conditioning Introduction:

    • Emitted Response: This refers to voluntary movements and behaviors that an organism chooses to perform.

    • Action-Dependent Consequences: The reward or consequence depends entirely on the action. If a specific behavior leads to a reward, it is more likely to be repeated; if it leads to nothing or a negative outcome, it is less likely to be repeated.

    • Comparison of Nervous Systems:

      • Classical Conditioning: Involves automatic bodily responses (reflexes and associations).

      • Operant Conditioning: Involves the voluntary nervous system (e.g., raising a hand, pressing a button).

    • Fundamental Principle: Classical conditioning happens to you; operant conditioning happens because of you.

Foundational Principles of Operant Conditioning

  • Definition: Operant conditioning is a form of learning controlled by the consequences of a living being’s behavior.

  • Instrumental Conditioning: Psychologists also use this term because the behavior plays a "functional" or "instrumental" role. The organism acts to gain something (food, safety, social interaction) or avoid something (pain, discomfort).

  • Edward Thorndike and the Puzzle Box:

    • Experimental Setup: Thorndike placed a hungry cat inside a cage that could only be opened from the inside by pressing a lever. Food was placed outside as motivation.

    • Process of Learning: The cat initially struggled (scratching at bars) and accidentally stepped on the lever. Performance improved gradually through trial and error, rather than an instant epiphany.

    • Ethical Considerations: Contemporary ethics committees would likely deem this experiment unethical due to the distress, confinement, and food deprivation imposed on the animals. Modern guidelines prioritize minimizing harm and using humane alternatives.

    • The Law of Effect: This principle states that behaviors followed by satisfying consequences are more likely to be repeated, while those followed by annoying consequences fade away.

B.F. Skinner and the Skinner Box

  • Expansion of Thorndike’s Work: Skinner coined the term "operant behavior" and developed a highly controlled environment for study.

  • The Skinner Box (Operant Chamber):

    • Consists of a box containing a lever (for rats) or a button/key (for pigeons) and a food dispenser.

    • Designed to study behavior precisely without outside distractions.

  • Skinner's Philosophy: Behavior is controlled by the environment regardless of the species. Principles learned from a rat pressing a lever apply to complex human behaviors, such as checking a phone for notifications.

The ABC Model of Behavior

Skinner explained behavior through a three-part system called a contingency:

  • A - Antecedent: What happens before the behavior (e.g., a command like "sit" or the presence of a Skinner box).

  • B - Behavior: The voluntary action that takes place (e.g., the dog sitting or the rat pressing the lever).

  • C - Consequence: What happens after the behavior (e.g., receiving a treat or a shock).

Types of Consequences

There are four distinct ways to shape behavior based on whether a stimulus is added (positivepositive) or removed (negativenegative), and whether the goal is to increase (reinforcementreinforcement) or decrease (punishmentpunishment) the behavior.

1. Reinforcement (Strengthens Behavior)

  • Positive Reinforcement: Adding something pleasant to increase a behavior (e.g., praise for good work, a bonus for performance).

    • Primary Reinforcers: Satisfy basic biological needs naturally (e.g., food, water).

    • Secondary (Conditioned) Reinforcers: Items that become rewarding through association with primary reinforcers (e.g., money, tokens, praise).

  • Negative Reinforcement: Removing something unpleasant to increase a behavior.

    • Important Distinction: Negative reinforcement is not punishment. It strengthens behavior by providing relief.

    • Escape Behavior: Acting to stop an ongoing unpleasant stimulus (e.g., taking paracetamol to stop a headache).

    • Avoidance Behavior: Acting before the unpleasant stimulus occurs to prevent it (e.g., wearing sunscreen to avoid a burn).

2. Punishment (Reduces Behavior)

  • Positive Punishment: Adding an aversive stimulus to decrease behavior (e.g., a speeding fine, a scolding, or the pain from touching a hot stove).

  • Negative Punishment: Removing a pleasant stimulus to decrease behavior (e.g., losing a driver’s license, a child losing tablet time).

Operant Extinction and Stimulus Control

  • Operant Extinction: Occurs when a learned behavior no longer receives reinforcement, leading the behavior to eventually disappear (e.g., a parent stoping the reward for a child's tantrum).

  • Extinction Burst: A temporary increase in the frequency, duration, or intensity of the behavior immediately after reinforcement is withdrawn (e.g., the child screaming louder before giving up).

  • Discriminative Stimuli (SDS^D): Signals that tell an organism when a behavior will lead to a consequence.

    • Example: A light in a Skinner box. When the light is on, the lever works; when off, it doesn't. The rat learns to only press when the light signals reward availability.

  • Operant Generalization: When a learned response extends to new but similar situations (e.g., avoiding all hot stoves after being burned by one).

  • Operant Discrimination: Learning to respond only in specific conditions (e.g., a child only taking a cookie when a parent is not watching).

Delay Discounting

  • Definition: The longer one has to wait for a reward or consequence, the less impact it has on behavior.

  • Immediate vs. Delayed: People naturally prioritize short-term, immediate gratification over long-term benefits.

  • Examples:

    • New Year’s Resolutions: The immediate pleasure of junk food outweighs the distant health benefits.

    • Smoking: The immediate nicotine high outweighs the delayed risk of future health issues.

Shaping and Chaining

  • Shaping: Reinforcing successive approximations of a final desired behavior. We reward small improvements until the goal is reached.

    • Example: Treating selective mutism by rewarding whispering, then single words, then full sentences.

  • Chaining: Teaching a complex, multi-step behavior by reinforcing each step in a specific order.

    • Example: Teaching a child to brush their teeth by reinforcing each link in the chain: picking up the brush, applying toothpaste, brushing top teeth, etc.

Schedules of Reinforcement

Reinforcement schedules determine how often and when a behavior is rewarded, influencing learning speed and resistance to extinction.

1. Continuous Reinforcement

  • Reward is given every time the behavior occurs.

  • Result: Fast learning, but very low resistance to extinction (behavior stops quickly once rewards stop).

2. Partial (Intermittent) Reinforcement

  • Reward is given only sometimes.

  • Result: Slower learning, but very high resistance to extinction.

Four Intermittent Schedules:
  • Fixed Ratio (FRFR): Reinforcement occurs after a set number of responses.

    • Example: A coffee card (buy 7, get 1 free) or commission-based sales.

    • Pattern: High response rates with a brief pause after reinforcement.

  • Variable Ratio (VRVR): Reinforcement occurs after an unpredictable, changing number of responses (averaging out over time).

    • Example: Poker machines, gambling, video game rewards.

    • Pattern: Steady, extremely high response rates with no pauses. This is the most resistant to extinction.

  • Fixed Interval (FIFI): Reinforcement occurs after a set period of time has passed, regardless of the number of responses.

    • Example: Getting paid every two weeks; checking an oven as timer approaches zero; cramming for tests.

    • Pattern: A scalloped response curve—slow activity after reward, followed by a burst of activity as the next interval deadline approaches.

  • Variable Interval (VIVI): Reinforcement occurs after an unpredictable, changing amount of time.

    • Example: Pop quizzes, checking emails/notifications, fishing.

    • Pattern: Steady, constant response rates because the reward is unpredictable but time-based.

Applied Behavior Analysis (ABA) and Token Economies

  • Token Economies: A system where desirable behaviors are reinforced with tokens (stickers, points) that have no inherent value but can be exchanged for primary or meaningful rewards (prizes, privileges).

  • Applied Behavioral Analysis (ABAABA): A scientific approach to changing behavior through evidence-based techniques.

    1. Data Collection: Observing and measuring behavior at baseline.

    2. Implementation: Using reinforcement/shaping strategies.

    3. Measurement: Tracking data to determine effectiveness.

Controversy regarding ABA and Autism

  • Criticism from the Autistic Community: Many advocates argue ABA focuses on "normalizing" behavior to neurotypical standards rather than supporting needs.

  • Aversive History: Earlier forms included ethically unacceptable techniques like electric shocks.

  • Masking: Concerns that ABA forces individuals to suppress natural behaviors (stimming, eye contact avoidance), which may lead to long-term anxiety and harm.

  • Neurodiversity Affirming Care: Modern focus shifts toward improving quality of life and supporting traits rather than enforcing behavioral conformity.