Learning

2.1 Core concepts

NCERT Chapter 5 frames learning as "any relatively permanent change in behaviour or behavioural potential produced by experience" (NCERT §Nature of Learning, p. 78). Two parts of this definition are critical: the change must be (a) relatively permanent and (b) the result of experience. Temporary changes brought about by drugs, fatigue, illness or maturation are therefore excluded from the category of learning. Learning — an internal, inferred process — is distinct from performance, the observed behaviour from which learning is inferred (NCERT pp. 78-79). A musician may have learned a piece thoroughly but perform poorly on a given day because of nervousness; learning has occurred even though performance has dipped.

NCERT lists six paradigms of learning (NCERT §Paradigms of Learning, p. 79). The simplest are the two forms of conditioning — classical and operant (instrumental). The remaining paradigms are observational, cognitive, verbal and skill learning. Each is introduced with its proponent and characteristic experimental setting.

Classical conditioning was discovered by Ivan Pavlov while studying digestion in dogs (NCERT §Classical Conditioning, pp. 79-80). Pavlov noticed that dogs began to salivate before food actually reached the mouth — in response to footsteps or the sight of the dish — and conducted systematic experiments using a harness apparatus (Fig. 5.1, p. 80). The famous procedure pairs a bell (CS) with food (US); the food naturally elicits salivation (UR), and after repeated pairings the bell alone elicits salivation (CR). Classical conditioning is therefore an S–S learning in which one stimulus comes to signal the imminent appearance of another (NCERT Table 5.1, p. 80). Three determinants govern its strength: (1) time relations between CS and US — simultaneous, delayed, trace and backward, with delayed conditioning being the most effective and backward conditioning rarely successful; (2) type of US — appetitive USs (food, water) require many trials whereas aversive USs (mild shock, loud noise) can establish a CR in just 1-3 trials; and (3) intensity of CS — more intense conditioned stimuli require fewer trials to produce conditioning (NCERT §Determinants of Classical Conditioning, p. 81).

Operant or instrumental conditioning was developed by B.F. Skinner, who studied voluntary responses ("operants") emitted by the organism and strengthened by their consequences (NCERT §Operant Conditioning, pp. 81-82). His apparatus, the Skinner Box, contained a lever and food dispenser; hungry rats placed inside learned to press the lever to obtain food pellets. The chief determinant is reinforcement — any stimulus or event that increases the probability that the response will occur again. Reinforcement varies in (a) type — positive (presenting a desired stimulus) and negative (removing an aversive stimulus); (b) number/frequency; (c) quality (superior such as preferred food vs inferior); and (d) schedule — continuous (every response reinforced) versus partial/intermittent. Crucially, negative reinforcement is not punishment: negative reinforcement increases the probability of avoidance/escape responses while punishment suppresses a response (NCERT §Determinants of Operant Conditioning, pp. 82-83). The partial reinforcement effect holds that intermittent reinforcement produces greater resistance to extinction than continuous reinforcement because the organism cannot readily detect when reinforcement has been discontinued (NCERT pp. 83, 91). Delayed reinforcement also weakens learning — smaller immediate rewards typically outweigh larger delayed ones (NCERT p. 83). Box 5.1 (p. 83) systematically contrasts the two forms: classical responses are reflexive and elicited (respondent), the CS and US are well-defined, and the experimenter controls reinforcement; operant responses are voluntary, the CS is not directly known, and the organism itself controls when the reinforcer occurs.

Several key learning processes apply to both paradigms (NCERT pp. 84-85; Fig. 5.3, p. 85): reinforcement (strengthens responses), extinction (removal of reinforcement weakens the response), generalisation (responding to stimuli similar to the CS), discrimination (responding differently to similar stimuli), and spontaneous recovery (reappearance of an extinguished CR after a rest period). NCERT also distinguishes primary reinforcers (biologically important — food for a hungry organism) from secondary reinforcers (money, praise, grades — acquiring reinforcing properties through experience) (NCERT §Reinforcement, p. 84). Box 5.2 (p. 84) describes learned helplessness, demonstrated by Martin Seligman and Steven Maier in dogs: animals given inescapable shock in a first (classical) phase later failed to escape avoidable shock in a second (operant) phase. The phenomenon is widely regarded as an animal analogue of human depression.

Observational learning, also called imitation, modeling or social learning, was demonstrated by Albert Bandura in his classic Bobo doll study (NCERT §Observational Learning, p. 86). Children who watched an adult model behave aggressively toward an inflatable doll later imitated those specific aggressive acts; whether the children spontaneously performed the behaviour depended on whether the model had been rewarded or punished. Cognitive learning focuses on what the learner knows rather than does. Wolfgang Köhler demonstrated insight learning in chimpanzees who suddenly grasped the solution to a banana-reaching problem in a "flash"; Edward Tolman demonstrated latent learning in rats who explored a maze without reinforcement and developed an internal cognitive map that was only displayed once food reward was introduced (NCERT §Cognitive Learning, p. 87).

Verbal learning is unique to humans and uses nonsense syllables, familiar and unfamiliar words as material. The standard methods are paired-associates learning (similar to S-R conditioning), serial learning with the serial-anticipation procedure, and free recall, in which Bousfield showed category clustering — participants spontaneously recall words from the same semantic category together (NCERT pp. 87-89; Tables 5.2, 5.3). Determinants of verbal learning include list length, meaningfulness and the total-time principle — a fixed amount of total practice time is required to master a list regardless of how trials are distributed (NCERT p. 89).

Skill learning treats a skill as a chain of perceptual-motor S-R associations. Paul Fitts described three phases — cognitive (understanding the task), associative (linking responses into sequences) and autonomous (smooth, automatic execution); transitions between phases produce a performance plateau, and the final phase achieves automaticity (NCERT §Skill Learning, p. 90). Two facilitating factors are motivation (intrinsic preferred over extrinsic) and preparedness — a species-specific biological constraint on what associations can easily be learnt. Finally, learning disabilities are heterogeneous CNS-based disorders involving difficulty in acquisition, reading, writing, speaking, reasoning or mathematical activity; symptoms include attentional problems, poor space/time orientation, poor motor coordination, perceptual disorders and dyslexia (confusion between letters such as b/d and p/q). Learning disabilities are seen even in children of average to superior intelligence (NCERT §Learning Disabilities, pp. 92-93).

2.2 Definitions to memorise

2.3 Diagrams / processes to remember

• Fig. 5.1 — Pavlovian harness (p. 80): Dog in harness with tube from salivary glands and cup measuring saliva; one-way glass wall separates dog from experimenter so that only the bell (CS) and food (US) function as stimuli — the canonical apparatus for classical conditioning.
• Table 5.1 — Stages of conditioning (p. 80): Before — Food (US)→Salivation (UR), Bell→Alertness; During — Bell (CS)+Food (US)→Salivation (UR); After — Bell (CS)→Salivation (CR). Three rows must be memorised in this order.
• Fig. 5.2 — Skinner Box (p. 82): Sound-proofed chamber with lever, food container and recording mechanism for studying operant conditioning of rats; the rat's spontaneous lever press is followed by a food pellet, raising the probability of the response.
• Box 5.1 — Classical vs Operant differences (p. 83): Reflexive vs voluntary; CS/US well-defined vs not; experimenter vs organism controls reinforcement; passive vs active subject — a high-yield comparison table.
• Fig. 5.3 — Spontaneous recovery curve (p. 85): Three phases — Acquisition (CS+US presented together, CR strength rises), Extinction (CS alone, CR strength falls), Spontaneous Recovery (CS alone after rest, CR strength partially returns).
• Tables 5.2 & 5.3 (p. 88): Sample nonsense syllables, unfamiliar and familiar words used in verbal-learning experiments and the paired-associate stimulus-response format.

2.4 Common confusions / NTA trap points

• Negative reinforcement ≠ punishment. Negative reinforcement increases an avoidance/escape response; punishment suppresses a response (p. 82).
• Most effective time-relation = delayed conditioning, not simultaneous; backward conditioning is rarely successful (p. 81).
• Partial reinforcement produces greater resistance to extinction, not continuous (pp. 83, 91).
• Insight vs latent learning: Kohler used chimpanzees (insight); Tolman used rats in a maze (latent learning + cognitive map) — students often swap.
• Pavlov vs Skinner vs Bandura vs Kohler vs Tolman vs Seligman: Match the psychologist with the paradigm — Pavlov (classical), Skinner (operant/Skinner Box), Bandura (observational/Bobo doll), Kohler (insight), Tolman (latent learning), Seligman & Maier (learned helplessness).
• Learning is inferred from changes in performance — performance itself is the observable behaviour, not learning (p. 79).
• Appetitive US takes more trials; aversive US establishes a CR in 1-3 trials. NTA often flips this.
• Primary vs Secondary reinforcers: primary = biological (food, water); secondary = acquired (money, praise, grades).
• Generalisation vs Discrimination: generalisation = responding to similar stimuli alike; discrimination = responding differently to similar stimuli. The same child who fears a bearded man (generalisation) but not a clean-shaven one (discrimination) illustrates both.
• Fitts' phases: Cognitive → Associative → Autonomous (in that order). Confusing the order is a frequent trap.

2.5 Thinkers and theories at a glance