Neurotransmitter systems: dopamine, serotonin, norepinephrine, GABA, glutamate

Intuition Beginner

Neurons communicate using chemical messengers called neurotransmitters. One neuron releases them into the tiny gap — the synapse — at the next neuron, where they bind to receptors and pass the signal along. More than 100 different neurotransmitters shape thought, mood, and movement.

Glutamate is the brain's main "go" signal — it excites neurons, pushing them to fire. GABA is the main "stop" signal — it calms neurons down. Together these two keep the brain's activity in balance. Most anti-anxiety drugs, including Valium, work by enhancing GABA, which quiets overactive circuits.

Dopamine is famous as the "reward" chemical — it surges when you eat good food, win a game, or fall in love. But dopamine is really about prediction and motivation: it fires when something turns out better than expected. That signal drives learning and keeps you pursuing goals.

Serotonin regulates mood, sleep, and appetite. Antidepressants like Prozac work by keeping serotonin active longer in the brain. Norepinephrine drives arousal and the body's fight-or-flight response to stress. Acetylcholine activates muscles and is crucial for attention and memory. When acetylcholine neurons die, memory fails — as in Alzheimer's disease.

Visual Beginner

The table below maps the major neurotransmitters to their primary roles and the drugs that target them. The same chemical can have very different effects depending on which receptors it binds and which brain circuits it activates.

Neurotransmitter	Main role	Drug that targets it
Glutamate	Excitation (the main "go")	Ketamine (blocks NMDA)
GABA	Inhibition (the main "stop")	Valium, alcohol
Dopamine	Reward, motivation, movement	L-DOPA (Parkinson's)
Serotonin	Mood, sleep, appetite	Prozac (SSRI)
Norepinephrine	Arousal, fight-or-flight	SNRI antidepressants
Acetylcholine	Muscle activation, memory	Alzheimer's drugs

Each neurotransmitter plays many roles depending on its receptor and circuit. A drug that enhances one system can treat a disorder while producing side effects elsewhere, because the same chemical is reused across the brain.

Worked example Beginner

In Parkinson's disease, dopamine-making neurons in the substantia nigra slowly die. Without enough dopamine, patients develop a tremor at rest, stiff muscles, and slow movement. The drug L-DOPA gives the brain the raw material to build more dopamine, easing these symptoms. This works because dopamine drives smooth, voluntary movement, not just reward.

Depression is linked, in part, to low serotonin. Prozac and similar drugs are called SSRIs — selective serotonin reuptake inhibitors. After release, serotonin is normally recycled back into the sending cell. An SSRI blocks this recycling, leaving more serotonin in the synaptic gap for longer. Mood steadies, though the full effect takes weeks to appear.

Check your understanding Beginner

Formal definition Intermediate+

The life cycle of a neurotransmitter

Neurotransmission unfolds in five stages. Synthesis: the transmitter is built from precursor molecules by specific enzymes (tyrosine hydroxylase for the catecholamines, tryptophan hydroxylase for serotonin, glutamic acid decarboxylase for GABA). Packaging: the transmitter is loaded into synaptic vesicles by transporters (VMAT for monoamines, VGlut for glutamate, VIAAT for GABA). Release: an arriving action potential opens voltage-gated calcium channels, calcium enters the terminal, and SNARE proteins fuse vesicles with the membrane, spilling transmitter into the cleft. Receptor binding: transmitter molecules bind receptors on the postsynaptic (and sometimes presynaptic) membrane. Termination: the signal ends by reuptake through plasma-membrane transporters (DAT, SERT, NET), enzymatic degradation (acetylcholinesterase, monoamine oxidase), or diffusion. Drugs act at every stage — synthesis (L-DOPA), release (amphetamines), receptors (agonists and antagonists), reuptake (SSRIs, cocaine), and degradation (MAO inhibitors, cholinesterase inhibitors).

Two receptor families

Neurotransmitter receptors come in two structural families with distinct time courses. Ionotropic receptors are ligand-gated ion channels: the transmitter binds, the channel opens within milliseconds, and ions flow, producing a fast postsynaptic potential (AMPA, NMDA, kainate for glutamate; GABA-A; nicotinic acetylcholine; 5-HT3). Metabotropic receptors are G-protein-coupled receptors (GPCRs): binding activates an intracellular G-protein, which triggers second-messenger cascades (cAMP, IP3) that open or close channels indirectly. Metabotropic effects are slower (hundreds of milliseconds to seconds) but longer-lasting and amplifying. Most amine transmitters act chiefly through GPCRs: all dopamine receptors (D1-D5), most serotonin receptors (all but 5-HT3), adrenergic receptors (alpha and beta), and muscarinic acetylcholine receptors (M1-M5).

Glutamate: the workhorse of excitation

Glutamate is the predominant excitatory neurotransmitter in the brain; roughly half of all synapses release it. Its ionotropic receptors divide by pharmacology and kinetics. AMPA receptors mediate the fast component of the excitatory postsynaptic current (sodium influx, mild depolarization). NMDA receptors are voltage-dependent: a magnesium ion blocks the channel at rest, and only when the membrane is already depolarized (by AMPA activity) does the block lift, allowing calcium to enter. That calcium influx is the trigger for long-term potentiation (LTP) and long-term depression (LTD) — the synaptic changes thought to underlie learning. Kainate receptors modulate release and contribute to excitatory signalling in specific circuits.

Excess glutamate is toxic. Excitotoxicity — the overactivation of glutamate receptors, especially NMDA, leading to calcium overload — kills neurons during stroke, traumatic brain injury, and in neurodegenerative diseases such as ALS. The ischemic cascade starves neurons of ATP, depolarizes them, releases glutamate, and triggers self-destruction. Memantine, a low-affinity NMDA antagonist, slows cognitive decline in Alzheimer's disease by blocking pathological NMDA activation while sparing normal signalling. Ketamine, a higher-affinity NMDA antagonist, produces dissociative anaesthesia at high doses and a rapid antidepressant effect at subanaesthetic doses — a pharmacology explored further in the master tier.

GABA: the workhorse of inhibition

GABA is the brain's chief inhibitory neurotransmitter. GABA-A receptors are ionotropic chloride channels: opening them lets chloride enter, hyperpolarizing the neuron and suppressing firing. GABA-A is the target of a remarkable cluster of sedative drugs. Benzodiazepines (Valium, Xanax) bind an allosteric site that increases the frequency of channel opening. Barbiturates increase the duration of opening. Alcohol and the Z-drugs (zolpidem) bind overlapping sites. All potentiate GABA, producing sedation, anxiolysis, and — at high doses — fatal respiratory depression (especially in combination). GABA-B receptors are metabotropic (Gi/o); the muscle relaxant baclofen is a GABA-B agonist. Too little GABA-mediated inhibition causes seizures; many anticonvulsants (vigabatrin, tiagabine) enhance GABAergic tone.

Dopamine: pathways and receptors

Dopamine is synthesized from tyrosine via L-DOPA (tyrosine hydroxylase is the rate-limiting enzyme). Four pathways define its anatomy. The mesolimbic pathway runs from the ventral tegmental area (VTA) to the nucleus accumbens and underlies reward, motivation, and addiction; its overactivity is linked to the positive symptoms of schizophrenia. The mesocortical pathway runs from the VTA to the prefrontal cortex and supports cognition, working memory, and motivation; its underactivity is implicated in the negative and cognitive symptoms of schizophrenia. The nigrostriatal pathway runs from the substantia nigra to the dorsal striatum and controls voluntary movement; its degeneration produces Parkinson's disease. The tuberoinfundibular pathway runs from the hypothalamus to the pituitary and inhibits prolactin release; D2-blocking antipsychotics interfere here, causing elevated prolactin (hyperprolactinemia).

Dopamine receptors divide into two families. D1-like receptors (D1, D5) couple to Gs and stimulate adenylate cyclase, increasing cAMP; they are generally postsynaptic and excitatory in their downstream effects. D2-like receptors (D2, D3, D4) couple to Gi and inhibit adenylate cyclase; they include presynaptic autoreceptors that throttle dopamine release. All dopamine receptors are GPCRs. Typical antipsychotics (haloperidol) are D2 antagonists; atypical antipsychotics (clozapine) combine D2 antagonism with 5-HT2A antagonism. The therapeutic window for D2 blockade is narrow: too little fails to control psychosis; too much produces motor side effects (tardive dyskinesia, parkinsonism) by disrupting the nigrostriatal pathway.

Serotonin: many receptors, many roles

Serotonin (5-hydroxytryptamine, 5-HT) is synthesized from the amino acid tryptophan via 5-HTP. The cell bodies of serotonergic neurons cluster in the raphe nuclei of the brainstem and project throughout the brain. At least 14 serotonin receptor subtypes have been identified, nearly all metabotropic except 5-HT3 (ionotropic). This receptor diversity lets one transmitter regulate mood, sleep, appetite, pain, and gastrointestinal function through distinct pathways. SSRIs (Prozac/fluoxetine, Zoloft/sertraline) block the serotonin transporter (SERT), raising synaptic serotonin; they treat depression, anxiety, and obsessive-compulsive disorder, with the therapeutic effect emerging over weeks — a delay that complicates the simple "low serotonin causes depression" story. The 5-HT2A receptor is the principal target of classical psychedelics (psilocybin, LSD, mescaline). Tryptophan depletion — acutely lowering brain serotonin — can transiently relapse symptoms in recovered depressed patients, a key tool for probing the causal role of serotonin.

Norepinephrine: arousal and stress

Norepinephrine is synthesized from dopamine by dopamine beta-hydroxylase. Its principal source is the locus coeruleus, a small nucleus in the pons whose neurons branch widely across the cortex, thalamus, and spinal cord. Norepinephrine acts on alpha receptors (alpha-1, Gq, excitatory; alpha-2, Gi, often presynaptic autoreceptors) and beta receptors (beta-1, beta-2, beta-3, Gs). The system governs arousal, vigilance, attention, and the stress response: locus coeruleus firing rises with wakefulness and stress and falls during sleep. SNRIs (venlafaxine, duloxetine) block reuptake of both serotonin and norepinephrine and treat depression and chronic pain. Stimulants used for ADHD (methylphenidate, amphetamines) raise synaptic dopamine and norepinephrine in the prefrontal cortex, improving attention. Beta-blockers blunt the peripheral adrenergic symptoms of stage fright.

Acetylcholine: muscle and memory

Acetylcholine is synthesized from choline and acetyl-CoA by choline acetyltransferase and is broken down in the cleft by acetylcholinesterase within milliseconds — a degradation so rapid that cholinergic signalling is terminated chiefly by enzymes, not reuptake. Two receptor families: nicotinic (ionotropic cation channels; at the neuromuscular junction they trigger contraction, and in the brain they mediate arousal and nicotine's addictive effects) and muscarinic (metabotropic, M1-M5). Cholinergic neurons of the basal forebrain (nucleus basalis) project to the cortex and are central to attention and memory. These neurons degenerate in Alzheimer's disease, which is why cholinesterase inhibitors (donepezil/Aricept, rivastigmine) — which prolong acetylcholine's action — modestly slow cognitive decline. At the neuromuscular junction, curare and botox act by blocking (curare) or preventing (botox) acetylcholine release, causing paralysis.

Endocannabinoids and endogenous opioids

Beyond the classical transmitters, two modulatory lipid and peptide systems shape behaviour. Endocannabinoids (anandamide, 2-AG) are synthesized on demand and signal retrogradely — they are released by the postsynaptic cell and bind CB1 receptors on the presynaptic terminal to suppress further transmitter release. This retrograde signalling lets neurons regulate their own input. THC (the active compound in cannabis) mimics anandamide at CB1 receptors, producing euphoria, appetite stimulation, and memory impairment. The endogenous opioid system comprises endorphins, enkephalins, and dynorphin acting at mu, delta, and kappa receptors. It mediates pain relief, reward, and the calming effects of social bonding. Morphine and heroin are mu-agonists; naloxone (Narcan) is a mu-antagonist that reverses opioid overdose.

Key mechanism: the dopamine reward prediction error Intermediate+

Of all the mechanisms described above, the most influential is the reward prediction error — the algorithmic account of what dopamine neurons actually compute. It reframes dopamine from a "pleasure chemical" into a teaching signal, and it is the bridge between neurochemistry and reinforcement learning.

Schultz's discovery

Wolfram Schultz recorded from midbrain dopamine neurons in monkeys while they received juice rewards, sometimes predicted by a visual cue. The neurons' firing pattern was striking and counterintuitive. An unexpected reward evoked a burst of firing (positive response). A fully predicted reward — one that always followed the cue — evoked no response at all. And the unexpected omission of a predicted reward produced a pause or dip below baseline firing (negative response). Over the course of learning, the burst shifted backward in time, from the reward itself to the earliest reliable predictor (the cue). Dopamine neurons, in other words, did not report reward. They reported the difference between the reward obtained and the reward expected.

The prediction-error equation

The signal can be written compactly. Let R be the actual reward delivered on a trial, and let V be the predicted value of the current state. The prediction error is

δ = R − V.

A positive δ (reward exceeds expectation) drives a burst and strengthens the actions and cues that predicted it. A δ near zero (reward matches expectation) produces no change — learning has converged. A negative δ (reward is worse than expected) drives a dip and weakens the preceding actions. Formally, this is the update rule of temporal-difference learning: the predicted value V is nudged toward the outcome, step by step, with δ as the teaching signal. The same algorithm underlies much of modern reinforcement learning in artificial intelligence, where it appears as the TD error.

Why this matters

The prediction-error account explains several otherwise puzzling facts. It explains why dopamine fires for cues that merely predict reward (the cue carries the prediction error once the reward is itself expected). It explains addiction: addictive drugs release dopamine pharmacologically, bypassing the prediction and delivering an unnaturally large δ that never fully habituates, locking learning onto drug-associated cues. It explains motivation: a cue that predicts reward becomes attractive and triggers approach behaviour (incentive salience), even if the reward itself is not yet present. And it explains why dopamine depletion does not merely reduce pleasure — it abolishes the drive to pursue rewards. The mechanism is the foundation for the master-tier topics that follow: optogenetic dissection, wanting versus liking, and the computational rivalry between dopamine and serotonin.

Exercises Intermediate+

Advanced topics Master

Phasic versus tonic dopamine and optogenetics

Dopamine neurons fire in two modes. Tonic firing is the slow, steady, few-hertz background that sets the overall level of receptor stimulation and governs baseline motivation and motor function. Phasic firing consists of rapid bursts (up to 20 Hz) lasting tens to hundreds of milliseconds, which carry the prediction-error signal described above. The tonic-phasic distinction matters because the same neurons, the same transmitter, and the same receptors support two computationally different functions depending on the timescale. Tonic dopamine is thought to set the "gain" on phasic signals and to regulate the threshold for action.

The causal dissection of these modes was made possible by optogenetics. Tsai, Deisseroth, and colleagues (2009) expressed channelrhodopsin selectively in VTA dopamine neurons of mice and drove phasic bursts with light at specific frequencies. Phasic, but not tonic, stimulation was sufficient to drive behavioural conditioning — mice learned to prefer the chamber where they received phasic dopamine stimulation. This established that phasic dopamine is causally sufficient for reinforcement, a claim that recording studies alone could not make. Subsequent work used inhibitory opsin halorhodopsin and the slower DREADD chemogenetic actuators to dissect the contributions of specific projections (VTA-to-nucleus-accumbens versus VTA-to-prefrontal-cortex) to different aspects of reward and aversion.

The D1 inverted-U: dopamine and prefrontal working memory

Dopamine's effect on the prefrontal cortex is nonmonotonic. Working memory and executive function depend on D1 receptor stimulation in a characteristic inverted-U dose-response: too little D1 activity destabilizes the persistent firing that holds information online; too much D1 activity over-narrows the network into a rigid state, impairing flexibility. The optimum sits in the middle. Williams and Goldman-Rakic (1995) demonstrated this in primate prefrontal neurons, and Arnsten extended it to stress: stress-induced catecholamine release pushes the system past the peak of the inverted-U, explaining why acute stress impairs prefrontal function while simultaneously sharpening more posterior, habit-based systems. The inverted-U also explains why the same stimulant (methylphenidate) improves attention in people with ADHD (who may sit on the rising slope) while disrupting focus in people with already-optimal dopamine tone.

Wanting versus liking: incentive salience

Kent Berridge's dissociation of "wanting" (incentive salience, the motivation to pursue a reward) from "liking" (hedonic pleasure, the enjoyment of consuming it) is one of the most consequential revisions to dopamine theory. Berridge showed that dopamine depletion or blockade powerfully suppresses wanting — rats cease to approach and work for food — while leaving liking intact, as measured by the orofacial "hedonic" responses to sweet tastes. Conversely, the brain contains small hedonic hotspots (in the nucleus accumbens shell and ventral pallidum) where opioid and endocannabinoid stimulation amplifies liking without affecting wanting. The two systems are anatomically and neurochemically distinct. This dissociation reframes addiction: addicts may "want" a drug intensely (high incentive salience, driven by dopamine-triggered cue learning) while "liking" it little (tolerance having eroded the hedonic response). It also predicts that dopamine is not, despite popular framing, the chemical of pleasure.

Serotonin, behavioural inhibition, and computational models

If dopamine is the transmitter of approach and reward learning, serotonin is increasingly cast as the transmitter of behavioural inhibition, patience, and the suppression of maladaptive impulses. Peter Dayan, Nathaniel Daw, and colleagues (2002) proposed that serotonin and dopamine act as opponent systems: dopamine drives action toward reward, serotonin holds behaviour in check when the environment is punitive or when waiting is adaptive. On this view, serotonin signals the average punishment rate, just as dopamine signals reward prediction error. Depleting serotonin (acute tryptophan depletion) makes animals and humans more impulsive and less able to wait for large delayed rewards. The account is computationally elegant but contested: serotonin's 14 receptor subtypes make a single-function summary implausible, and recent evidence implicates serotonin in the valuation of time, social behaviour, and even moral decision-making (serotonin depletion reduces harm aversion in moral dilemmas). The honest conclusion is that serotonin supports a family of inhibitory and value-related functions whose unifying computational principle remains under active debate.

The locus coeruleus–norepinephrine system and adaptive gain

The locus coeruleus is a few thousand neurons in the pons that supply nearly all of the brain's norepinephrine. Aston-Jones and Cohen (2005) proposed the adaptive gain theory: the locus coeruleus optimizes the trade-off between exploiting a current task (phasic, task-evoked bursts that sharpen cortical processing) and disengaging to explore alternatives (tonic elevation that lowers the threshold for shifting behaviour). Phasic locus coeruleus firing tracks task performance and precedes correct decisions; tonic elevation marks drowsiness, distraction, or the decision to disengage. Disorders of this system are implicated in ADHD (under-arousal, poorly sustained attention), post-traumatic stress disorder (hyperarousal and exaggerated noradrenergic stress responses), and the cognitive decline of aging. The system also gates synaptic plasticity: norepinephrine release during emotionally arousing events strengthens memory consolidation, which is why beta-blockers given shortly after trauma can blunt the formation of intrusive memories.

Acetylcholine, the basal forebrain, and attention

Acetylcholine from the basal forebrain (nucleus basalis of Meynert) and the brainstem pedunculopontine nucleus sets the cortex's attentional state. Cholinergic tone rises with wakefulness and task engagement and sharpens the signal-to-noise of cortical responses, allowing relevant inputs to drive neurons while suppressing irrelevant ones. Sarter and colleagues argue that acetylcholine specifically enables the detection of signals — the ability to sustain attention under uncertainty. The degeneration of basal forebrain cholinergic neurons in Alzheimer's disease underlies the attentional and memory deficits and justifies cholinesterase-inhibitor treatment. Nicotine's stimulation of nicotinic receptors in the prefrontal cortex transiently improves attention, which is part of its addictive grip and the basis for off-label cognitive effects.

Measuring and manipulating neurotransmitters in vivo

For decades, extracellular neurotransmitter levels were inferred indirectly (microdialysis, slow and coarse) or from postmortem tissue. Fast-scan cyclic voltammetry (FSCV), developed by the Wightman group, measures neurotransmitter concentration at subsecond resolution by applying a rapid voltage ramp to a carbon-fibre electrode and reading the resulting current — each oxidizable species (dopamine, serotonin, norepinephrine) produces a characteristic voltammogram. FSCV confirmed the phasic dopamine transients that the prediction-error model predicts, in real time, in behaving animals. On the manipulation side, DREADDs (designer receptors exclusively activated by designer drugs) let researchers switch specific cell types on or off with a systemically injected inert ligand — slower than optogenetics but far less invasive, and usable in freely behaving animals over days. Together, FSCV and DREADDs let neuroscience move from correlation ("dopamine is correlated with reward") to circuit-specific causation ("this projection's phasic dopamine causes this animal to approach").

Trace amines and TAAR1

Beyond the classical monoamines, the brain contains trace amounts of trace amines (tyramine, beta-phenylethylamine, tryptamine) — endogenous relatives of the catecholamines and serotonin. Long dismissed as metabolic byproducts, they were rescued by the 2001 discovery of the trace amine-associated receptor 1 (TAAR1), a GPCR activated by trace amines, amphetamines, and the psychedelic MDMA-related compounds. TAAR1 modulates monoamine transporters and firing, and its antagonism of dopamine and serotonin systems gives it a striking pharmacology: TAAR1 agonists show antipsychotic and pro-cognitive effects in animal models without the motor side effects of D2 blockade. A TAAR1 agonist (ulotaront) has reached late-stage clinical trials for schizophrenia, offering one of the first mechanistically novel antipsychotic strategies in a generation.

Psychedelic pharmacology and the 5-HT2A receptor

Classical psychedelics — psilocybin, LSD, DMT, mescaline — are, at the molecular level, 5-HT2A receptor agonists. The 5-HT2A receptor is densely expressed in layer V pyramidal neurons of the prefrontal cortex, where its activation increases cortical excitability and, at the network level, increases the entropy and diversity of brain activity (Carhart-Harris's finding that psychedelics raise brain entropy, inverting the reduced entropy seen in depression and obsession). The REBUS model (relaxed beliefs under psychedelics) proposes that 5-HT2A activation transiently relaxes the weight of high-level prior beliefs, allowing maladaptive priors (rigid depressive or addictive schemas) to be revised. MDMA acts differently — it is a potent releaser of serotonin and dopamine via the transporter, producing empathy and euphoria, and is being tested for PTSD. Ketamine, structurally distinct, is an NMDA antagonist whose downstream effects on glutamate and the growth factor BDNF produce a within-hours antidepressant response that no monoaminergic drug can match.

Psychedelic-assisted therapy: psilocybin, MDMA, ketamine

After decades of prohibition-driven dormancy, psychedelic research has revived. Johns Hopkins and Imperial College London trials show that psilocybin, delivered in a structured therapeutic setting, produces large and sustained reductions in treatment-resistant depression, end-of-life anxiety, and tobacco and alcohol addiction. The MAPS-sponsored trials of MDMA-assisted therapy for severe PTSD have reported response rates far exceeding those of SSRIs, with effects persisting long after a few sessions — a pattern consistent with the REBUS account of belief revision rather than chronic symptom suppression. Ketamine and its enantiomer esketamine are now licensed for treatment-resistant depression, validating glutamatergic, rapid-acting pharmacology. These trials raise deep questions: is the therapeutic effect pharmacological (a receptor event) or experiential (the subjective trip), or both? What are the risks of loosened cognition in vulnerable individuals? And how should a therapy that works in a handful of guided sessions be integrated into a medical system built around daily dosing?

The neurocircuitry of addiction: the Koob–Volkow model

Nora Volkow and George Koob frame addiction as a three-stage cycle that progressively shifts across brain systems. Binge/intoxication (driven by the basal ganglia and dopamine, the acute rewarding and habit-forming effects) gives way to withdrawal/negative affect (driven by the extended amygdala, where stress neurotransmitters like CRF and dynorphin produce the dysphoria of abstinence), which in turn drives preoccupation/anticipation (driven by the prefrontal cortex, where cue-driven craving and impaired executive control conspire to drive relapse). Across the cycle, the balance shifts from impulsive (ventral striatum, dopamine-driven, "I want it now") to compulsive (dorsal striatum, habit-driven, stimulus-response) behaviour, and from liking to wanting. The model explains why willpower alone rarely overcomes established addiction (the prefrontal control system is itself compromised) and why treatment must address dysphoria and cue-driven craving, not just acute intoxication. It also predicts that no single pharmacology will suffice: the disorder spans dopamine, stress, and glutamate systems across months and years.

Connections Master

• Neuroscience: brain and behaviour 29.02.01. This unit extends the survey of neurotransmitters introduced in 29.02.01 into full systems. The prerequisite covered the synapse and the major transmitters at a survey level; this unit formalizes receptor families, the dopamine pathways, and the reward prediction error.

• Brain regions and function 29.02.02 pending. The chemical systems treated here wire the regions treated in the preceding unit. Parkinson's disease is the conjunction of the substantia nigra (region) and the nigrostriatal dopamine pathway (system); fear conditioning joins the amygdala with GABAergic and noradrenergic signalling; reward learning joins the nucleus accumbens with mesolimbic dopamine. The two units are most naturally read as a pair.

• Neuroplasticity 29.02.04 pending. The successor unit depends directly on this one. Long-term potentiation and depression are glutamatergic phenomena — the NMDA receptor's calcium influx is the molecular trigger for synaptic change. Dopamine gates which synapses get strengthened. Recovery from injury is, at bottom, the re-weighting of neurotransmitter-driven circuits.

• Learning and memory 29.04.01. The dopamine prediction error is the neural implementation of reinforcement learning. The memory unit builds the psychological theory of conditioning and skill acquisition on this neurochemical foundation.

• Psychological disorders 29.09.01. Depression (serotonin, norepinephrine), schizophrenia (dopamine), anxiety (GABA, serotonin), addiction (dopamine, opioid, stress), and Alzheimer's (acetylcholine) are the clinical face of the systems described here. The disorders unit reads these systems through the lens of psychopathology.

• Therapy and treatment 29.10.01. Pharmacotherapy is applied neurotransmitter science: SSRIs, antipsychotics, benzodiazepines, cholinesterase inhibitors, and the new psychedelic and ketamine protocols all act on the receptors and transporters catalogued in this unit.

• Biology: cellular electrophysiology 17.09.01. The action potentials, vesicle fusion (SNARE machinery), and ion-channel biophysics that underlie neurotransmission are treated in the biology strand. Optogenetics and FSCV, the causal and measurement tools of this unit, are built directly on that biophysics.

Historical and philosophical context Master

Olds and Milner: the discovery of "reward circuits"

In 1954, James Olds and Peter Milner, working at McGill, implanted electrodes in rat brains and discovered that the animals would press a lever thousands of times an hour to deliver electrical stimulation to certain sites — the septal area, and later the medial forebrain bundle and nucleus accumbens. The rats would forgo food and even cross a shock grid to self-stimulate. This was the first demonstration that the brain contains circuits whose activation is itself reinforcing, independent of any external reward. Olds and Milner's "pleasure centres" framed reward as a localized brain function and launched four decades of research that located the effect in the mesolimbic dopamine pathway. The framing has since been refined — Berridge's wanting-versus-liking work shows the rats were stimulating "wanting" circuits, not necessarily experiencing pleasure — but the empirical discovery stands as the founding experiment of behavioural neuroscience of reward.

Arvid Carlsson and the identification of dopamine

When Arvid Carlsson began his work in the 1950s, dopamine was thought to be merely a precursor to norepinephrine, with no signalling role of its own. In 1957-1958, Carlsson showed that dopamine is concentrated in the striatum (not evenly distributed as a precursor would be), that reserpine depletes it and causes parkinsonian immobility in animals, and that L-DOPA — which restores dopamine — reverses that immobility. This established dopamine as a neurotransmitter in its own right and pinpointed its role in movement. The work earned Carlsson a share of the 2000 Nobel Prize in Physiology or Medicine and laid the foundation for L-DOPA therapy in Parkinson's disease, still the gold-standard treatment. It also seeded the dopamine hypothesis of the disorders that would dominate psychopharmacology for the next half-century.

The dopamine hypothesis of schizophrenia and its revisions

The finding that dopamine-blocking drugs (chlorpromazine, haloperidol) calm psychosis, while dopamine-releasing drugs (amphetamines) can induce it, gave rise to the dopamine hypothesis: schizophrenia as dopamine overactivity. The hypothesis explained the drugs but was too coarse. It could not account for the negative and cognitive symptoms, for the delayed onset of antipsychotic action, or for why some patients respond poorly. The refined dopamine hypothesis (Howes and Kapur, 2009) localizes the overactivity to the mesolimbic pathway (positive symptoms) while attributing the negative and cognitive symptoms to mesocortical underactivity, and traces the upstream cause to cortical dysfunction (involving glutamate and the NMDA receptor) that dysregulates midbrain dopamine. This layered account — symptoms as pathway-specific dopamine imbalances, driven by glutamatergic cortical dysfunction — illustrates how a simple "chemical imbalance" framing gives way, under evidence, to a circuit-level theory.

The serotonin hypothesis of depression under scrutiny

For decades, depression was popularly explained as a "chemical imbalance" of serotonin, and SSRIs were marketed on that basis. The hypothesis has come under serious challenge. A 2022 umbrella review by Moncrieff and colleagues found limited direct evidence that low serotonin causes depression, and acute tryptophan depletion does not reliably lower mood in untreated depressed people (though it can in recovered ones). Defenders argue that the serotonin system is too complex (14 receptors, multiple pathways) for a single-depletion study to falsify, that SSRIs work through downstream neuroplastic and anti-inflammatory mechanisms rather than acute serotonin elevation, and that clinical trial evidence supports modest but real efficacy. The debate does not show that SSRIs are useless; it shows that the mechanism story sold to the public was an oversimplification. The honest position is that we know less about the cause of depression than the "chemical imbalance" slogan implied, even as the drugs remain among the most prescribed in psychiatry.

The psychedelic renaissance

Psychedelics were researched seriously in the 1950s and 1960s — over a thousand papers were published on LSD — before being criminalized and effectively banished from science for a generation. Their return, since the early 2000s, is one of the most striking reversals in modern medicine. The renaissance rests on three pillars: mechanistic (the 5-HT2A receptor and the REBUS model give a principled account of how a single receptor event could revise entrenched beliefs), clinical (rigorous randomized trials of psilocybin and MDMA show effects that conventional treatments do not), and cultural (a growing willingness to treat subjective experience as therapeutically central rather than epiphenomenal). The philosophical stakes are real: if the therapeutic effect of psilocybin depends on the subjective trip, then consciousness is not a side effect of pharmacology but its vehicle — a claim that cuts against a purely mechanistic, brain-state-reduces-to-receptor-event view of psychiatry. Whether the early promise survives large-scale replication, and whether the model can scale beyond the intensive guided-setting format, are the open questions of the next decade.

Bibliography Master

1. Olds, J. and Milner, P., "Positive Reinforcement Produced by Electrical Stimulation of Septal Area and Other Regions of Rat Brain," Journal of Comparative and Physiological Psychology 47 (1954), 419-427. The founding discovery of brain reward circuits and intracranial self-stimulation.

2. Carlsson, A., Lindqvist, M., and Magnusson, T., "On the Biochemistry and Possible Functions of Dopamine and Noradrenaline in Brain," Naunyn-Schmiedeberg's Archiv für experimentelle Pathologie und Pharmakologie 231 (1957), 17-25. The discovery that dopamine is a neurotransmitter and its localization to the striatum.

3. Carlsson, A., "A Half-Century of Neurotransmitter Research: Impact on Neurology and Psychiatry," Nobel Lecture, Bioscience Reports 21 (2001), 359-373. The 2000 Nobel Prize synthesis of the dopamine story and the L-DOPA therapy it enabled.

4. Schultz, W., Dayan, P., and Montague, P. R., "A Neural Substrate of Prediction and Reward," Science 275 (1997), 1593-1599. The prediction-error account of dopamine neurons, with the temporal-difference framing.

5. Schultz, W., "Predictive Reward Signal of Dopamine Neurons," Journal of Neurophysiology 80 (1998), 1-27. The full monograph establishing the dopamine reward prediction error across learning.

6. Williams, G. V. and Goldman-Rakic, P. S., "Modulation of Memory Fields by Dopamine D1 Receptors in Prefrontal Cortex," Nature 376 (1995), 572-575. The D1 inverted-U and the modulation of prefrontal working memory.

7. Berridge, K. C. and Robinson, T. E., "What Is the Role of Dopamine in Reward: Hedonic Impact, Reward Learning, or Incentive Salience?," Brain Research Reviews 28 (1998), 309-369. The wanting-versus-liking dissociation and incentive salience theory.

8. Tsai, H.-C., Zhang, F., Adamantidis, A., Stuber, G. D., Bonci, A., de Lecea, L., and Deisseroth, K., "Phasic Firing in Dopaminergic Neurons Is Sufficient for Behavioral Conditioning," Science 324 (2009), 1080-1084. The optogenetic demonstration that phasic dopamine is causally sufficient for reinforcement.

9. Daw, N. D., Kakade, S., and Dayan, P., "Opponent Interactions Between Serotonin and Dopamine," Neural Networks 15 (2002), 603-616. The computational opponent-systems model of serotonin and dopamine.

10. Aston-Jones, G. and Cohen, J. D., "An Integrative Theory of Locus Coeruleus-Norepinephrine Function: Adaptive Gain and Optimal Performance," Annual Review of Neuroscience 28 (2005), 403-450. The adaptive-gain theory of the locus coeruleus.

11. Robinson, D. L., Hermans, A., Seipel, A. T., and Wightman, R. M., "Monitoring Rapid Chemical Communication in the Brain," Chemical Reviews 108 (2008), 2554-2584. Fast-scan cyclic voltammetry for real-time neurotransmitter measurement.

12. Roth, B. L., "DREADDs for Neuroscientists," Neuron 89 (2016), 683-694. Chemogenetic actuators for circuit-specific manipulation of neurotransmitter systems.

13. Grandy, D. K., "Trace Amine-Associated Receptor 1 — Family Archetype or Iconoclast?," Pharmacology and Therapeutics 116 (2007), 355-390. The pharmacology of TAAR1 and the trace amines.

14. Carhart-Harris, R. L. and Friston, K. J., "REBUS and the Anarchic Brain: Toward a Unified Model of the Brain Action of Psychedelics," Pharmacological Reviews 71 (2019), 316-344. The relaxed-beliefs-under-psychedelics model of 5-HT2A action.

15. Nichols, D. E., "Psychedelics," Pharmacological Reviews 68 (2016), 264-355. The authoritative pharmacology of classical psychedelics and the 5-HT2A receptor.

16. Koob, G. F. and Volkow, N. D., "Neurocircuitry of Addiction," Neuropsychopharmacology 35 (2010), 217-238. The three-stage addiction cycle and the shift from impulsive to compulsive behaviour.

17. Howes, O. D. and Kapur, S., "The Dopamine Hypothesis of Schizophrenia: Version III — The Final Common Pathway," Schizophrenia Bulletin 35 (2009), 549-562. The revised circuit-level dopamine hypothesis.

18. Moncrieff, J., Cooper, R. E., Stockmann, T., Amendola, S., Hengartner, M. P., and Horowitz, M. A., "The Serotonin Theory of Depression: A Systematic Umbrella Review of the Evidence," Molecular Psychiatry 28 (2023), 3243-3256. The critical reassessment of the serotonin hypothesis of depression.

19. Kandel, E. R., Schwartz, J. H., Jessell, T. M., Siegelbaum, S. A., and Hudspeth, A. J., Principles of Neural Science, 6th ed. (McGraw-Hill, 2021). Ch. 6-15 cover neurotransmitter systems, receptor physiology, and synaptic transmission in depth.

20. Myers, D. G. and DeWall, C. N., Psychology, 13th ed. (Worth, 2021). Ch. 2 provides the introductory treatment of neurotransmitters and their behavioural roles.