Fig 1 Reward pathway: one model hover or focus a region · space: pause first exposure · wanting low · control high
prefrontal cortex (control, decision) nucleus accumbens (ventral striatum) amygdala VTA (dopamine neurons) hippocampus cue context cerebellum brainstem cerebrum front back Schematic, not to scale: a simplified mid-sagittal silhouette with approximate region positions. Pulse size and brightness are qualitative, not measurements. Shows one leading model of addiction, not settled fact.
FIRST EXPOSURE
AFTER REPEATED EXPOSURE (model)
Show cue
Pause motion
D2 receptor availability (striatum)
Response to non-drug rewards
Cue-triggered 'wanting'
Prefrontal top-down control
Pulse size and brightness, glow, and the five-step bars are qualitative illustrations of direction of change, not measured values.
Hover or focus any region or pathway for a short explanation. Press 'Show cue' to fire a cue, then watch the signal reach the accumbens.
What the figure shows, and how sure we are
This is one model, not settled fact. Addiction is multi-factorial: genes, development, environment, stress, mental health and social context all contribute (Volkow & Morales, Cell 2015 ). The figure shows a leading neurobiological account, and says nothing about any individual person.
Drugs and natural rewards raise dopamine in the nucleus accumbens (animal data). In freely moving rats, drugs abused by humans increased extracellular dopamine, especially in the accumbens (Di Chiara & Imperato, PNAS 1988 ); a food reward raised accumbens dopamine in rats too (Hernandez & Hoebel, Life Sci 1988 ; see also Wise, Phil Trans R Soc B 2006 ). Midbrain dopamine neurons fire to salient, reward-related events (Schultz, Dayan & Montague, Science 1997 ).Circuit layout (textbook-level). Midbrain dopamine neurons feed forebrain circuits including ventral and dorsal striatum and prefrontal cortex (Hyman, Malenka & Nestler, Annu Rev Neurosci 2006 ). The addiction-cycle model gives the VTA and ventral striatum a focal role in binge/intoxication, and a network including orbitofrontal and prefrontal cortex, basolateral amygdala and hippocampus a role in preoccupation/craving, with disrupted prefrontal inhibitory control (Koob & Volkow, Neuropsychopharmacology 2010 ).Lower D2 receptor availability (human PET, group comparisons). Cocaine abusers showed lower D2 availability than controls, still present 3-4 months after detoxification, with lower frontal metabolism (Volkow et al., Synapse 1993 ); methamphetamine abusers likewise (Volkow et al., Am J Psychiatry 2001 ). Reviewed in Volkow et al., Arch Neurol 2007 . Cocaine-dependent subjects also released less striatal dopamine to a stimulant challenge and reported a smaller ‘high’ (Volkow et al., Nature 1997 ). These are cross-sectional comparisons: they do not by themselves show whether the difference preceded or followed use, and the D2 signal was measured mainly in dorsal striatum, not the accumbens specifically.Blunted response to non-drug rewards. Repeated drug exposure is proposed to reduce sensitivity to non-drug rewards, weaken self-regulation and enhance reactivity to drug cues (Volkow & Morales, Cell 2015 ). Supporting human data are modest: in one small study, mood did not track monetary reward in cannabis users and smokers as it did in non-smokers (Martin-Soelch et al., Neuropsychobiology 2009 ).Incentive-sensitization (a theory). Repeated use may make the dopamine system hypersensitive to drugs and drug-associated cues, sensitising ‘wanting’ (incentive salience) separately from ‘liking’ (Robinson & Berridge, Brain Res Rev 1993 ; open questions in Robinson & Berridge, Phil Trans R Soc B 2008 ). Note that this sits alongside the finding of a blunted dopamine response to a drug challenge in PET; how to reconcile a sensitised cue response with blunted reward and drug-evoked release is a model-level interpretation and remains debated.Prefrontal control. Imaging implicates prefrontal cortex in regulating limbic reward regions and in self-control, with disruption underlying compulsive taking (Goldstein & Volkow, Nat Rev Neurosci 2011 ; Koob & Volkow 2010 above).
What was simplified. Brain silhouette, region positions, sizes, route paths and all pulse speeds are illustrative. The amygdala and hippocampus lie lateral to the midline and are projected onto the cut plane; their routes to the VTA are shown as direct links but are indirect in reality. Two model states are hand-set qualitative levels (direction of change only); no number in the figure is a measurement. Many other systems (serotonin, opioid, glutamate, stress circuits, the extended amygdala and habenula) and the dorsal striatum are omitted. Cue and context inputs arrive by sensory routes not drawn.
For education only; this is not medical advice and does not describe any individual. Nothing here concerns doses, routes or how to use any substance. If you or someone you know is struggling with substance use, a local health service or helpline can help.