Learning – Classical, Operant & Observational Conditioning

Innate Behaviours: Reflexes & Instincts

• Reflexes
  • Rapid, involuntary, protective reactions to specific stimuli.
  • Present at birth; no prior learning required.
  • Examples (implied by lecture): pulling hand from hot object, knee-jerk.
• Instincts
  • Complex, species-typical behavioural patterns.
  • Also innate, unlearned, crucial for survival/reproduction (e.g., birds migrating, spiders spinning webs).
• Key Contrast
  • Neither reflexes nor instincts require experience, whereas learning does.

What Is Learning?

• “A relatively permanent change in behaviour or knowledge that results from experience.”
• Can be intentional (studying for an exam) or incidental (learning not to touch a hot stove after being burned).
• Occurs across species; fundamental for adaptation & survival.
• Minds naturally connect events that occur close together or in sequence → foundation of associative learning.
• Discussion prompt from slide: Recall something you learned without realising it at the time.

Three Basic Learning Processes

• Classical (Pavlovian) Conditioning – learning via stimulus–stimulus associations.
• Operant Conditioning – learning via consequences of voluntary behaviour.
• Observational Learning (Modeling) – learning by watching others.

Classical Conditioning

Pavlov’s Dog Study (Digestion → Serendipitous discovery)

• Originally measured salivation to food.
• Dogs began salivating to anticipatory cues (e.g., footsteps).
• Led Pavlov to test bell → food pairings.

Experimental Sequence

1. Before Conditioning
   • $\text{UCS (Food)} \rightarrow \text{UCR (Salivation)}$
   • $\text{NS (Bell)} \rightarrow \text{No response}$
2. During Conditioning (Acquisition)
   • $\text{NS (Bell)} + \text{UCS (Food)} \rightarrow \text{UCR (Salivation)}$
3. After Conditioning
   • $\text{CS (Bell)} \rightarrow \text{CR (Salivation)}$

• Demonstrated learning through association.

Key Terminology (Definitions + Pavlov Examples)

• Unconditioned Stimulus (UCS) – naturally elicits response (food).
• Unconditioned Response (UCR) – automatic reaction to UCS (salivating to food).
• Neutral Stimulus (NS) – initially no relevant response (bell pre-pairing).
• Conditioned Stimulus (CS) – former NS, now triggers response after pairing (bell post-pairing).
• Conditioned Response (CR) – learned response to CS (salivating to bell).

Little Albert Experiment (Watson & Rayner, 1920)

• Initial State: 9-month-old Albert shows no fear of rat, rabbit, dog, masks.
• Conditioning: Whenever Albert reaches for rat (NS), experimenters strike steel bar (UCS – loud noise) → Albert cries (UCR).
• After Several Pairings: Rat alone (now CS) → fear/crying (CR).
• Generalisation: Fear transferred to rabbit, dog, fur coat, even Santa Claus mask.
• Ethical Issues
  • No desensitisation; long-term harm possible.
  • Modern ethics would prohibit inducing lasting fear in infants.

Core Principles

• Acquisition – initial NS+UCS pairings where learning occurs.
• Extinction – CR weakens if CS is presented without UCS repeatedly.
• Spontaneous Recovery – extinguished CR can reappear after a pause.
• Generalisation – stimuli similar to CS evoke CR (e.g., child fears all dogs).
• Discrimination – learning to respond only to specific CS (e.g., fear only large barking dogs).

Higher-Order / Second-Order Conditioning

• Pair a new second-order stimulus (e.g., squeaky cabinet) with existing CS (electric can-opener) → new stimulus also elicits CR (salivation) without UCS.
• Shows conditioning can build layers of associations beyond original UCS.

Applications & Discussion Prompts

• Create own classical-conditioning example and label UCS, UCR, NS, CS, CR.
• Role in understanding phobias (conditioned fears) & marketing (pair products with positive stimuli like music, celebrities).

Operant Conditioning (B.F. Skinner)

• Learning in which voluntary behaviours are strengthened or weakened by consequences.
• Skinner Box: rat/pigeon presses lever → food (reinforcement) or avoids shock, allowing precise study.

Classical vs. Operant (Comparative Table)

• Classical: involuntary, stimulus precedes response, theorist Pavlov, explains reflexes & emotions.
• Operant: voluntary, consequence follows behaviour, theorist Skinner, used for behaviour modification.

Reinforcement & Punishment Matrix

Examples

• Student studies to receive praise → $\text{positive reinforcement}$.
• Student studies to avoid anxiety → $\text{negative reinforcement}$ (removal of unpleasant feeling).

Schedules of Reinforcement

• Continuous – reinforce every response (fast acquisition, fast extinction).
• Partial (Intermittent) – slower learning, greater resistance to extinction.
  • Fixed-Ratio (FR): after set number of responses (free coffee after 10 buys).
  • Variable-Ratio (VR): unpredictable number (slot machines, random social-media likes).
  • Fixed-Interval (FI): first response after fixed time (weekly paycheck).
  • Variable-Interval (VI): first response after varying time (checking email, package arrival).

Diagrammatic Summary

• Ratio = based on actions; Interval = based on time.
• Fixed = predictable; Variable = unpredictable.

Classroom/Group Scenarios (Analysis)

1. Coffee Shop Loyalty Card
   • Consequence: Positive reinforcement (free coffee added).
   • Schedule: Fixed-Ratio $FR\,10$.
2. Random Social-Media Likes
   • Consequence: Positive reinforcement (likes).
   • Schedule: Variable-Ratio (unpredictable likes to posts).
3. Checking for Package
   • Consequence: Positive reinforcement (package found).
   • Schedule: Variable-Interval (arrival time uncertain).
4. Teen Loses Phone
   • Consequence: Negative punishment (phone removed).
   • Schedule: One-time consequence, not a reinforcement schedule.
5. Weekly Paycheck
   • Consequence: Positive reinforcement (money added).
   • Schedule: Fixed-Interval (every 7 days).

Shaping (Successive Approximations)

• Reinforce any behaviour resembling target → gradually require closer resemblance → finally only exact target gets reinforcement.
• Enables teaching complex chains (e.g., animal tricks, speech therapy, athletic training).

Observational Learning (Modeling)

• Proposed by Albert Bandura; extends learning beyond direct experience.

Bobo Doll Experiment

• Children watched adult behave aggressively vs. non-aggressively toward inflatable doll.
• Observed aggression → children later imitated aggression; non-aggressive model → minimal aggression.
• Demonstrated learning through vicarious experience.

Four Necessary Processes

1. Attention – must notice model’s behaviour.
2. Retention – must store memory of behaviour.
3. Reproduction – must be physically capable.
4. Motivation – must have reason/desire to imitate (e.g., expectation of reward).

• Example: Younger sibling learns to cook by watching elder.

Ethical, Philosophical & Practical Implications

• Little Albert raises modern research-ethics concerns: informed consent, protection from harm, right to withdraw, debriefing.
• Marketing/Advertising leverages classical conditioning – pair products with music, attractive imagery, sponsorship.
• Behaviour Modification (operant) used in classrooms, workplaces, therapy, animal training.
• Phobias & Anxiety Disorders can develop via classical conditioning; treatments like systematic desensitisation rely on extinction & counter-conditioning.
• Social Learning highlights influence of media violence on behaviour (policy debates on TV/video-game regulation).

Quick Reference Equations & Notation

• Classical conditioning chain: $NS + UCS \xrightarrow{\text{repeated pairings}} CS \rightarrow CR$
• Reinforcement schedule symbols: $FR\,n,\ VR,\ FI,\ VI$ where (n) = fixed ratio number.
• Law of Effect (Thorndike precursor to Skinner): $\text{Responses followed by satisfying consequences are more likely to recur.}$