HookA fruit machine pays out on a schedule Skinner designed for pigeons
Sit a pigeon in a Skinner box and reward it every single time it pecks, and the moment you stop, it stops. But reward it unpredictably — sometimes on the third peck, sometimes the twentieth, never on a fixed count — and the pigeon pecks like its life depends on it, and keeps pecking long after the food dries up. Skinner called that a variable-ratio schedule, and it produces the most persistent, extinction-resistant behaviour he could engineer. A modern fruit machine, a loot box and a betting app all run on exactly that schedule. Addiction, in other words, is not a mystery of weak willpower; it is behaviour built by mechanisms psychologists can name.
That is the through-line of this option: addiction is explained at several levels at once, and the examiner wants you to move between them. At the biological level, dopamine floods the brain's reward pathway. At the behavioural level, reinforcement schedules and conditioned cues keep the habit alive. At the cognitive level, faulty beliefs — the gambler's fallacy, the illusion of control — do the rest. And treatment, and the models of behaviour change behind it, only make sense once you can say which mechanism you are trying to interrupt.
ModelWhat addiction actually is — dependence, tolerance and withdrawal
Precision here earns easy AO1 marks, and losing it loses them. Physical dependence is a bodily state in which the drug has become part of the system's normal functioning, so that stopping produces a withdrawal syndrome — the cluster of physiological symptoms (nausea, tremor, sweating, seizures in severe cases) that appear on cessation. Psychological dependence is the compulsion to keep using because of the reward or relief it delivers: the craving, the sense that you cannot cope without it. The two usually travel together but are distinct — a gambler can be psychologically dependent with no physical withdrawal at all.
Tolerance is the reason yesterday's dose stops working: repeated exposure means the body adapts, so more of the drug is needed to achieve the original effect. It comes in flavours worth naming — metabolic (the liver clears the drug faster), cellular (receptors down-regulate) and behavioural (the user learns to compensate). Tolerance and withdrawal are two faces of the same adaptation: the system re-tunes itself around the drug, which is why removing the drug feels like the illness rather than the cure.
DataWhy some people get hooked and others don't — the risk factors
No single cause; a stack of probabilities. Genetic vulnerability is real — twin and adoption studies put the heritability of alcohol dependence at roughly a half, and the A1 variant of the DRD2 dopamine-receptor gene is associated with fewer D2 receptors and a blunted reward response, the 'reward deficiency' that may push some people to seek stronger stimulation. Stress feeds addiction through self-medication: adverse childhood experiences and chronic strain raise the odds, and substances that dampen the stress response become negatively reinforcing. Personality matters too — high impulsivity and sensation-seeking, and traits associated with antisocial personality disorder, predict earlier and heavier use, though the causal arrow is hard to pin down. Family influences work partly through genes and partly through social learning: children model parental use and read parental attitudes as permission. Peers supply the availability, the norms and the direct modelling — adolescents massively over-estimate how much their friends use, and conform to the estimate. The examiner's caution: these are correlational risk factors, not switches, and they interact — which is why a diathesis-stress framing beats any single-factor story.
MechanismNicotine — dopamine plus a lifetime of learned cues
Nicotine reaches the brain in about ten seconds. There it binds nicotinic acetylcholine receptors in the ventral tegmental area, which triggers a surge of dopamine in the nucleus accumbens — the mesolimbic 'reward pathway' the brain uses to tag things worth repeating. With repeated exposure the receptors down-regulate, so the smoker needs nicotine simply to feel normal: that adaptation is tolerance, and its removal is withdrawal. This is the neurochemical half of the explanation.
The other half is learning theory, and it explains why quitting is so cue-dependent. Through operant conditioning, each cigarette delivers positive reinforcement (the dopamine hit) and, once dependence sets in, powerful negative reinforcement (relief from withdrawal). Through classical conditioning, the reward becomes associated with contexts — the coffee, the pub, the drive to work — so those cues alone trigger craving, a phenomenon called cue reactivity. Secondary reinforcers such as the lighter and the ritual acquire reinforcing power of their own. The strength of the account is that it explains relapse: a smoker can clear the nicotine in days but meet the conditioned cues for years.
MechanismGambling — the variable schedule meets the faulty calculator
Gambling has no ingested drug, so it is the cleanest test of the psychological explanations. Learning theory centres on the reinforcement schedule from the intro: gambling pays out on a variable-ratio basis — unpredictable, intermittent wins — which Skinner showed produces behaviour that is exceptionally resistant to extinction. A run of losses does not extinguish it because the gambler has learned that the next bet might be the one. Add the 'near miss' (two jackpot symbols and a blank), which the brain processes almost like a win, and early 'big wins' that anchor the habit, and you have operant conditioning by design.
Cognitive theory supplies the beliefs that keep it going: the gambler's fallacy (a loss makes a win 'due'), the illusion of control (blowing on the dice, choosing 'lucky' numbers), and a recall bias that remembers wins vividly and edits out losses. Griffiths recorded regular and non-regular gamblers thinking aloud on a fruit machine and found the regulars made far more irrational verbalisations ('the machine likes me') and believed they were more skilful, despite no real difference in outcome. The two explanations dovetail: the schedule builds the habit, the distortions justify it.
A worked mini-answer shows how to fuse AO1 and AO3 economically. AO1 first, tightly: 'Learning theory explains gambling through operant conditioning on a variable-ratio schedule, in which unpredictable wins act as intermittent positive reinforcement and produce behaviour highly resistant to extinction.' That single sentence earns description marks without padding. Then evaluate with elaboration: 'A strength is that this predicts a specific, testable pattern — gambling should persist through long losing runs, which is exactly what is observed and which a simple reward account cannot explain.' Notice the 'which cannot explain' — you are showing what the theory does that a rival does not. Then the limitation: 'However, learning theory struggles to explain why most people exposed to the same schedules never become addicted, implying that reinforcement interacts with cognitive biases and genetic vulnerability rather than acting alone.' That sentence does three jobs at once — states a weakness, explains it, and points toward the diathesis-stress line of argument the whole answer should build. Compression like this is how you fit 6 AO1 and 10 AO3 into the time available.
CaseReducing addiction — chemistry, conditioning and cognition
Three families of treatment, each targeting a different mechanism. Drug therapy works on the biology: nicotine replacement supplies an agonist (the drug at a controlled, tapering dose), while bupropion and varenicline act on the reward pathway to blunt craving; for gambling, opioid antagonists such as naltrexone dampen the reward. Drugs treat symptoms fast but not the learned habit, so relapse is common once they stop. Behavioural interventions attack the conditioning directly. Aversion therapy uses classical conditioning to pair the addictive behaviour with something unpleasant — a nausea-inducing drug with alcohol, rapid over-smoking, a mild shock with gambling imagery — so the cue that once predicted reward comes to predict discomfort. Covert sensitisation does the same in imagination, pairing a vividly imagined aversive scene with the behaviour, which is gentler and more ethical. Both suffer poor long-term generalisation. Cognitive behaviour therapy targets the thinking: functional analysis of the triggers, direct challenge to the gambler's fallacy and illusion of control, and skills training for high-risk situations. CBT tends to hold up best over time precisely because it re-tools the beliefs and cues the other treatments leave intact — but it demands motivation and effort the drug route does not.
ModelChanging behaviour — planned intentions and the stages of quitting
Two models explain how change happens rather than what causes addiction. The theory of planned behaviour (Ajzen) says behaviour follows behavioural intention, and intention is predicted by three things: attitudes to the behaviour ('quitting will make me healthier'), subjective norms (whether important others approve of quitting) and perceived behavioural control (how able you feel to do it) — and control also predicts behaviour directly. Its strength is that it identifies precisely which lever an intervention should pull; its weakness is the notorious intention–behaviour gap, since many people who fully intend to quit simply do not.
Prochaska's six-stage model (the transtheoretical model) treats change as a process, not an event: precontemplation (not yet considering change), contemplation (weighing it up), preparation (getting ready), action (actively changing), maintenance (holding the change) and termination (the behaviour no longer tempts). Its great insight is that it treats relapse as a normal part of a cycle rather than a failure, and that it matches the intervention to the stage — you do not offer nicotine patches to someone in precontemplation who does not yet want to stop. Its critics note the stages can be arbitrary and hard to distinguish. Between them, the two models turn 'just decide to quit' into a set of things a psychologist can actually target.
VocabularyKey terms the mark scheme pays for
TrapsMisconceptions that cost marks
ExamWhat examiners want
Addiction is a Paper 3 option, so it lives or dies on the 16-mark essay, marked 6 AO1 + 10 AO3 (or 6 AO1 + 4 AO2 + 6 AO3 with a scenario). The commonest failure is spending too long describing a mechanism and leaving no room to evaluate it — examiner reports single this out year after year. Fix the balance at the planning stage: budget a tight paragraph of accurate AO1, then three or four developed AO3 points.
Definitions are worth guarding. 'Describe what is meant by tolerance / withdrawal / dependence' questions are pure AO1, and the marks turn on precision — distinguish physical from psychological dependence, and name the type of tolerance. In evaluation, use point → evidence → explain → link and make the 'explain' step spell out why a study strengthens or weakens the explanation; a named study you cannot connect to a conclusion is half a mark, not a full one. The reliable line of argument across this topic is diathesis-stress — no single factor (dopamine, reinforcement, a gene, a bias) explains why most exposed people never become addicted — so signpost it early and cash it out at the end.
Where a question carries a stem (a described smoker, gambler or dieter), you must apply to it explicitly for the AO2 marks: quote the detail, then map it onto the mechanism, as in the gambling worked example. And where a treatment or model is asked for, judge it on outcomes — relapse rates, the intention–behaviour gap — not just on how it is meant to work.