Neuroscience: brain and behaviour

Intuition Beginner

Your brain weighs about 1.4 kilograms. It has the consistency of firm gelatin and looks, at first glance, like a wrinkled walnut. From this organ everything you experience emerges: every thought, memory, emotion, decision, and dream. Neuroscience is the attempt to understand how.

The project is ambitious and, depending on who you ask, either making extraordinary progress or barely scratching the surface. Brain imaging studies appear in headlines weekly, claiming to have found the "centre" of love, morality, or political orientation. These claims are almost always overstated. Understanding what neuroscience genuinely knows — and where its limits lie — requires stepping back from the hype and looking at the biological machinery itself.

The cellular building blocks

The nervous system is built from two cell types: neurons and glial cells. Neurons are the signal-processing units. A typical neuron has three parts: a cell body (soma), dendrites that receive incoming signals, and a single axon that carries outgoing signals to other cells. The human brain contains roughly 86 billion neurons, each connected to thousands of others, producing something on the order of 100 trillion synaptic connections.

Glial cells outnumber neurons and perform support functions: oligodendrocytes wrap axons in a fatty insulating sheath called myelin, which dramatically speeds signal transmission; astrocytes regulate the chemical environment around synapses and contribute to metabolic support; microglia act as the brain's immune cells, clearing debris and fighting infection. Glia were once dismissed as mere structural scaffolding, but research since the 1990s has shown they participate actively in synaptic signalling and plasticity.

How neurons communicate

Neurons communicate through a combination of electrical and chemical signals. Within a single neuron, information travels electrically. The neuron maintains a voltage difference across its membrane — the resting membrane potential — of approximately $-70$ millivolts (negative inside relative to outside). This potential arises from an unequal distribution of ions: sodium (${\text{Na}}^{+}$) is concentrated outside the cell, potassium (${\text{K}}^{+}$) inside, and the membrane is selectively permeable.

When a neuron receives sufficient excitatory input at its dendrites, the membrane potential depolarises (becomes less negative). If it reaches a threshold of approximately $-55$ mV, voltage-gated sodium channels open. Sodium rushes in, the membrane potential shoots up to roughly $+40$ mV, and this rapid depolarisation is the action potential — the basic unit of neural signalling. The action potential is an all-or-nothing event: it either fires completely or does not fire at all. There is no such thing as a "half-strength" spike.

After the peak, voltage-gated potassium channels open, potassium flows out, and the membrane repolarises back toward rest. There follows a brief refractory period during which the neuron cannot fire again, ensuring that action potentials propagate in one direction along the axon — from the soma toward the axon terminals.

At the axon terminals, the electrical signal is converted to a chemical one. The terminal contains synaptic vesicles packed with neurotransmitters. When an action potential arrives, calcium channels open, triggering vesicle fusion with the cell membrane and releasing neurotransmitter molecules into the synaptic cleft — the roughly 20-nanometre gap between the sending (presynaptic) neuron and the receiving (postsynaptic) neuron.

Neurotransmitters bind to receptors on the postsynaptic membrane, which can be ionotropic (directly opening ion channels) or metabotropic (triggering intracellular signalling cascades via G-proteins). The effect can be excitatory (depolarising the postsynaptic cell, making it more likely to fire) or inhibitory (hyperpolarising it, making firing less likely).

Major neurotransmitter systems include:

• Glutamate — the brain's primary excitatory neurotransmitter. Essential for learning and memory. Excessive glutamate release causes excitotoxicity, a mechanism implicated in stroke and neurodegenerative diseases.
• GABA (gamma-aminobutyric acid) — the primary inhibitory neurotransmitter. Many anti-anxiety medications (benzodiazepines) enhance GABA signalling.
• Dopamine — involved in reward, motivation, and motor control. Imbalances are central to Parkinson's disease (dopamine deficiency in the substantia nigra) and schizophrenia (excessive dopaminergic signalling in mesolimbic pathways).
• Serotonin — modulates mood, sleep, appetite, and cognition. Selective serotonin reuptake inhibitors (SSRIs) are the most widely prescribed antidepressants, though their mechanism is more complex than the "chemical imbalance" narrative suggests.
• Acetylcholine — critical for attention, learning, and memory. Degeneration of cholinergic neurons in the basal forebrain is a hallmark of Alzheimer's disease.
• Norepinephrine — regulates arousal, vigilance, and the stress response.

The picture of neurotransmission presented above is a simplification. Most neurons release multiple neurotransmitters. Receptors exist in multiple subtypes with different effects. Neuromodulators like dopamine and serotonin do not simply transmit a signal but alter the gain of neural circuits, changing how responsive entire networks are to other inputs.

Visual Beginner

The diagram illustrates the fundamental signal path: input arrives at dendrites, is integrated at the axon hillock, propagates as an action potential along the myelinated axon, and triggers neurotransmitter release at the synapse. Myelin allows the action potential to "jump" between nodes of Ranvier (saltatory conduction), increasing transmission speed up to 120 metres per second in large motor neurons.

Brain anatomy

The brain can be described at multiple levels. A useful division separates it into three major regions visible in any vertebrate: the forebrain, midbrain, and hindbrain.

The hindbrain sits at the top of the spinal cord and includes the medulla (controlling vital autonomic functions: breathing, heart rate, blood pressure), the pons (a relay station between the cerebrum and cerebellum), and the cerebellum (critical for motor coordination, balance, and motor learning — and increasingly recognised as involved in cognitive and emotional processing).

The midbrain is small but contains structures essential for sensory processing (the superior and inferior colliculi for visual and auditory reflexes), motor control (the substantia nigra, whose dopamine-producing neurons degenerate in Parkinson's disease), and reward processing (the ventral tegmental area, a key source of dopaminergic signals).

The forebrain dominates the human brain both in size and in the functions most associated with "higher cognition." Its major external structure is the cerebral cortex, the deeply folded outer layer. The folding (gyri and sulci) allows a large cortical surface area to fit inside the skull. The cortex is divided into two hemispheres, each further divided into four lobes:

• Frontal lobe — executive functions (planning, decision-making, working memory), voluntary motor control (primary motor cortex), and speech production (Broca's area in the left hemisphere). Damage to the frontal lobe can produce profound personality changes. The most famous historical case is Phineas Gage, a railroad worker who survived an iron rod being driven through his frontal lobe in 1848; his personality was reportedly transformed from responsible and well-mannered to impulsive and profane.

• Parietal lobe — somatosensory processing (touch, temperature, pain, proprioception), spatial reasoning, and attention. The primary somatosensory cortex contains a mapped representation of the body (the sensory homunculus), with disproportionate areas devoted to sensitive regions like the fingers and lips.

• Temporal lobe — auditory processing, language comprehension (Wernicke's area), memory formation (the hippocampus is tucked inside the medial temporal lobe), and emotional processing. The amygdala, critical for fear and emotional salience, sits at the anterior tip of the temporal lobe.

• Occipital lobe — visual processing. The primary visual cortex (V1) receives input from the retina via the lateral geniculate nucleus of the thalamus. Subsequent visual areas process colour, motion, form, and depth.

Beneath the cortex lies a set of subcortical structures collectively called the limbic system — though the term is somewhat imprecise and neuroanatomists debate its boundaries. Key structures include the hippocampus (forming new explicit memories; spatial navigation), the amygdala (emotional processing, especially fear and threat detection), the thalamus (a relay and integration hub for nearly all sensory information heading to the cortex), the hypothalamus (regulating hunger, thirst, body temperature, and the hormonal system via the pituitary gland), and the basal ganglia (a group of structures including the caudate, putamen, and globus pallidus, involved in action selection, habit formation, and motor control).

The brainstem (medulla, pons, and midbrain) controls basic life-sustaining functions and serves as the conduit for all ascending sensory and descending motor pathways between the brain and spinal cord. The reticular formation within the brainstem regulates consciousness and arousal — damage here can produce coma.

Methods of studying the brain

Understanding how we study the brain is as important as understanding what we have found. Each method has distinct strengths and limitations.

Structural imaging reveals anatomy. CT (computed tomography) uses X-rays to produce cross-sectional images and is excellent for detecting bleeding, tumours, and acute structural damage. MRI (magnetic resonance imaging) uses strong magnetic fields and radio waves to produce detailed images of soft tissue, offering far better resolution than CT for brain structure.

Functional imaging attempts to measure neural activity indirectly. fMRI (functional MRI) detects changes in blood oxygenation — the blood-oxygen-level-dependent (BOLD) signal. The logic is that active brain regions consume more oxygen, and oxygenated and deoxygenated haemoglobin have different magnetic properties. fMRI has good spatial resolution (2-3 millimetres) but poor temporal resolution (seconds), because the haemodynamic response lags behind neural firing by several seconds. Crucially, the BOLD signal measures blood flow, not neural activity directly. It is an inference.

EEG (electroencephalography) records electrical activity from electrodes placed on the scalp. It has excellent temporal resolution (milliseconds) but poor spatial resolution, because the skull and scalp scatter the electrical signals. EEG is valuable for studying sleep stages, seizure disorders, and the timing of cognitive processes (event-related potentials, or ERPs).

PET (positron emission tomography) uses radioactive tracers to measure metabolic activity, neurotransmitter receptor density, or other molecular targets. PET can answer questions that fMRI cannot — for instance, mapping dopamine receptor distribution — but involves radiation exposure and is expensive.

Lesion studies examine what happens when a specific brain region is damaged. Historically crucial (Broca's and Wernicke's discoveries of language areas came from post-mortem examination of patients with specific deficits), lesion studies remain important but have limitations: brain damage is rarely confined to clean anatomical boundaries, and compensatory reorganisation can obscure the original function of the damaged area.

TMS (transcranial magnetic stimulation) uses magnetic fields to induce electrical currents in cortical neurons, temporarily activating or disrupting activity in a targeted region. When TMS briefly disrupts a function, it provides causal evidence that the stimulated region contributes to that function — a stronger inference than the correlational evidence from fMRI alone. However, TMS can only reach superficial cortical areas and its spatial precision is limited.

Each method illuminates different aspects of brain function, and converging evidence from multiple methods is always more convincing than any single approach. The temptation to overinterpret brain scans — showing a coloured blob on a brain image and declaring that a complex mental function has been "located" — is widespread. This is the core of the "neuromania" critique explored in later sections.

Neuroplasticity

The brain is not a fixed machine. Neuroplasticity refers to the brain's capacity to change its structure and function in response to experience. This was not always accepted. Through much of the twentieth century, the dominant view held that the adult brain was essentially static — neurons could die but not regenerate, and synaptic connections were fixed after a critical developmental period.

Evidence accumulated against this view. In the 1960s and 1970s, David Hubel and Torsten Wiesel showed that visual cortical organisation in kittens was profoundly shaped by early visual experience — suturing one eye shut during a critical period permanently altered cortical representation. In the 1980s and 1990s, Michael Merzenich and colleagues demonstrated that adult monkey somatosensory cortex reorganises after amputation or training: the cortical map of the hand shifts depending on which fingers are most used.

Long-term potentiation (LTP) and long-term depression (LTD) — persistent strengthening and weakening of synaptic connections following specific patterns of neural activity — were identified as candidate cellular mechanisms for learning and memory. Bliss and Lomo first described LTP in the rabbit hippocampus in 1973. The phenomenon is input-specific (only active synapses are strengthened), associative (weak and strong inputs to the same cell can cooperate), and persistent (lasting hours to days in vitro, potentially a lifetime in vivo).

Adult neurogenesis — the birth of new neurons — has been confirmed in specific brain regions (notably the dentate gyrus of the hippocampus and the olfactory bulb), though the extent and functional significance of adult neurogenesis in humans remains debated.

Plasticity is a double-edged sword. The same mechanisms that allow learning and recovery also underlie pathological changes: chronic pain, addiction (reward circuits remodelled by drug exposure), and phantom limb sensations (cortical reorganisation after amputation generating false signals).

Worked example Beginner

Consider a patient, "Patient DF," studied by Milner and Goodale (1995). DF suffered carbon monoxide poisoning that damaged her lateral occipital cortex, impairing conscious visual perception of object shape. Asked to describe the orientation of a slot in a letterbox-like apparatus, she could not report it. Asked to post a card through the slot, her hand oriented correctly as she reached.

This dissociation — between impaired conscious perception and intact visually guided action — illustrates the modular organisation of visual processing. The ventral stream ("what" pathway, running from V1 into the temporal lobe) supports conscious object recognition and was damaged in DF. The dorsal stream ("where/how" pathway, running from V1 into the parietal lobe) supports spatial processing and visually guided actions and remained intact.

The case shows that "seeing" is not a single, unified process. Different brain networks extract different kinds of information from the same visual input, and damage to one network can spare the other.

Check your understanding Beginner

Formal definition Intermediate+

The Hodgkin-Huxley model

The action potential was quantitatively characterised by Alan Hodgkin and Andrew Huxley in 1952 using the giant squid axon. Their model describes the membrane potential $V_m$ as a function of ionic conductances:

$$C_m\frac{d{V}_m}{dt}=-g_{\text{Na}}{m}^{3}h(V_m-E_{\text{Na}})-g_{\text{K}}{n}^4(V_m-E_{\text{K}})-g_L(V_m-E_L)+I_{\text{ext}}$$

where $C_m$ is membrane capacitance, $g_{\text{Na}}$, $g_{\text{K}}$, and $g_L$ are the maximal conductances for sodium, potassium, and leak channels respectively, $E_{\text{Na}}$, $E_{\text{K}}$, and $E_L$ are the reversal (Nernst) potentials for each ion, and $I_{\text{ext}}$ is any externally applied current. The variables $m$, $h$, and $n$ are dimensionless gating variables (ranging from 0 to 1) that evolve according to first-order kinetics:

$$\frac{dx}{dt}={\alpha }_x(V_m)(1-x)-{\beta }_x(V_m)x$$

where $x\in \{m,h,n\}$ and ${\alpha }_x$, ${\beta }_x$ are voltage-dependent rate constants determined empirically by Hodgkin and Huxley. The variable $m$ represents sodium activation, $h$ represents sodium inactivation, and $n$ represents potassium activation. The cubic and quartic exponents ($m^{3}h$, $n^4$) reflect the number of independent gating particles hypothesised to control each channel.

The resting membrane potential can be derived from the Goldman-Hodgkin-Katz equation, which accounts for the relative permeabilities of multiple ions:

$$V_m=\frac{RT}{F}\ln \frac{P_{\text{K}}[{\text{K}}^{+}{]}_{\text{out}}+P_{\text{Na}}[{\text{Na}}^{+}{]}_{\text{out}}+P_{\text{Cl}}[{\text{Cl}}^{-}{]}_{\text{in}}}{P_{\text{K}}[{\text{K}}^{+}{]}_{\text{in}}+P_{\text{Na}}[{\text{Na}}^{+}{]}_{\text{in}}+P_{\text{Cl}}[{\text{Cl}}^{-}{]}_{\text{out}}}$$

where $R$ is the gas constant, $T$ is temperature, $F$ is Faraday's constant, and $P_x$ denotes the membrane permeability to ion $x$.

Synaptic integration and neural coding

A single neuron receives inputs from thousands of synapses. The postsynaptic membrane integrates these inputs both spatially (summing simultaneous inputs from different synapses on the dendritic tree) and temporally (summing inputs arriving in rapid succession before earlier EPSPs or IPSPs have decayed). Whether the neuron fires depends on whether this integrated input drives the membrane potential at the axon hillock above threshold.

Neural coding — how information is represented in patterns of neural activity — is a central question. Rate coding posits that information is carried by the firing rate (spikes per second). Temporal coding argues that the precise timing of individual spikes carries information beyond what the rate alone captures. Population coding holds that information is distributed across large ensembles of neurons, with individual cells being broadly tuned and accuracy emerging from the collective pattern.

Key mechanisms: LTP and synaptic plasticity

Hebbian plasticity, summarised by the maxim "neurons that fire together, wire together," was formalised in models of long-term potentiation. The best-characterised mechanism involves NMDA receptors — glutamate receptors that function as coincidence detectors. NMDA receptors are blocked by a magnesium ion at resting potential. They open only when two conditions are simultaneously met: glutamate is bound (presynaptic activity) and the postsynaptic membrane is already depolarised (postsynaptic activity). This dual requirement implements a molecular AND gate for detecting correlated activity.

When NMDA receptors open, calcium enters the postsynaptic cell, triggering biochemical cascades (involving CaMKII, PKC, and other kinases) that strengthen the synapse. One mechanism is the insertion of additional AMPA receptors into the postsynaptic membrane, increasing the synapse's responsiveness to future glutamate release. Conversely, weak or uncorrelated activity can lead to LTD through phosphatase-mediated removal of AMPA receptors.

Split-brain research

Some of the most striking evidence for functional lateralisation comes from split-brain patients — individuals whose corpus callosum (the massive bundle of axons connecting the two cerebral hemispheres) has been surgically severed, usually as a treatment for severe epilepsy.

Roger Sperry and Michael Gazzaniga studied these patients from the 1960s onward. Because sensory pathways are largely contralateral (the left visual field projects to the right hemisphere and vice versa; the left hand is controlled by the right hemisphere and vice versa), information presented exclusively to one hemisphere of a split-brain patient cannot be transferred to the other.

The findings were dramatic. A word presented in the left visual field (right hemisphere) could not be verbally reported, because speech production (Broca's area) is typically in the left hemisphere, and the right hemisphere's visual input cannot cross to the left. But the patient's left hand (controlled by the right hemisphere) could select the corresponding object. The right hemisphere knew the word but could not say it.

This work revealed that the two hemispheres are not symmetrical in function. The left hemisphere (in most right-handed individuals) is dominant for language, sequential logic, and analytical processing. The right hemisphere contributes to spatial processing, face recognition, and certain aspects of emotional processing. The popular "left brain = logical, right brain = creative" simplification is a caricature — both hemispheres contribute to most functions — but the lateralisation itself is real.

Split-brain research also raised deep questions about consciousness and the unity of the self. If the two hemispheres can hold different information and generate different responses, does a split-brain patient have two streams of consciousness? Gazzaniga has argued for a "left-hemisphere interpreter" — a module in the left hemisphere that constructs narratives to explain behaviour, even when the true cause is inaccessible. In one striking demonstration, the right hemisphere was commanded to wave (via a left visual field instruction). When the patient was asked why they waved, the speaking left hemisphere — which had no access to the instruction — confabulated: "I saw someone I know" or "I was stretching."

Counterexamples to common slips

• "We only use 10% of our brains." False. Brain imaging shows that virtually all of the brain is active, though not all simultaneously. Energy consumption is remarkably high for an organ of its size: the brain accounts for roughly 2% of body weight but consumes 20% of the body's glucose and oxygen.

• "Brain scans show exactly what a person is thinking." Misleading. fMRI measures blood oxygenation changes correlated with neural activity. The relationship between BOLD signal and cognitive function is inferential, not direct. Statistical analysis of fMRI data involves multiple comparisons (tens of thousands of voxels) and requires careful correction — a point explored in the critique section.

• "The brain is hardwired." False. Neuroplasticity is now well-established. Cortical maps reorganise with experience, new synapses form, and even adult brains generate new neurons in limited regions.

• "Specific functions are localised to single brain areas." Oversimplified. Most cognitive functions emerge from networks of distributed brain regions. Damage to a single area can impair a function without eliminating it entirely, because other regions often compensate.

Exercises Intermediate+

Advanced topics Master

The nature-nurture debate in neuroscience

The question of whether behaviour and mental traits are determined by genes or by environment is one of the oldest in psychology. Neuroscience has transformed this debate, but it has not resolved it — because the question, framed as an either/or, is badly posed.

The genetic side. Behavioural genetics uses twin studies, adoption studies, and genome-wide association studies (GWAS) to estimate the heritability of traits. Heritability ($h^2$) represents the proportion of phenotypic variance in a population attributable to genetic variance. For many psychological traits — intelligence, personality dimensions, risk of mental illness — twin studies yield moderate to high heritability estimates (typically 0.3 to 0.8). GWAS have identified specific genetic variants associated with conditions like schizophrenia, depression, and autism spectrum disorder.

However, heritability estimates have critical limitations. They are population-level statistics: they say nothing about any individual. They are environment-dependent: heritability of height is high in a population with adequate nutrition but low in a population where malnutrition stunts growth. The same trait can have radically different heritability estimates in different populations or historical periods. Heritability does not imply immutability — highly heritable traits (like phenylketonuria) can be modified by environmental intervention.

GWAS findings have also revealed that most psychological traits are highly polygenic — influenced by hundreds or thousands of genetic variants, each with a tiny effect. No single "gene for" intelligence, schizophrenia, or personality exists. This polygenic architecture means that genetic prediction is probabilistic and incomplete.

The environmental side. Environmental effects on brain development are profound. Malnutrition, toxin exposure (lead, mercury), chronic stress, and early deprivation all produce measurable effects on brain structure and function. The Romanian orphanage studies — following children raised in severely deprived institutional conditions — documented dramatic effects on cognitive development, attachment, and brain growth, many of which persisted even after children were placed in nurturing foster care.

Animal models provide causal evidence for environmental effects. Meaney and Szyf's work with rats showed that maternal licking and grooming behaviour alters DNA methylation at the glucocorticoid receptor gene promoter in the hippocampus. Pups of high-licking mothers have more glucocorticoid receptors and a more tightly regulated stress response. Cross-fostering experiments — placing pups of low-licking mothers with high-licking mothers — reversed the effect, demonstrating that the mechanism is environmental, not genetic.

Epigenetics — the study of heritable changes in gene expression that do not involve changes to the DNA sequence itself — has become the key framework for understanding gene-environment interaction. DNA methylation, histone modification, and non-coding RNA regulation can all be influenced by experience, and some epigenetic modifications can be transmitted across generations. Epigenetics does not resolve nature vs nurture so much as dissolve the distinction: genes are not static blueprints but dynamic systems whose expression is regulated by the environment.

The emerging consensus is one of gene-environment interaction (GxE). The same genotype can produce different phenotypes in different environments, and the same environment can affect individuals differently depending on their genotype. A variant of the serotonin transporter gene (5-HTTLPR), for example, appears to moderate the effect of stressful life events on depression risk — individuals carrying the short allele are more vulnerable to depression following adversity, though the evidence is mixed and this finding has been contested.

The gut-brain axis

One of the most rapidly evolving areas of neuroscience concerns the communication between the gastrointestinal system and the brain — the gut-brain axis. This is not a new idea (the term "gut feeling" has a long folk history), but the mechanistic details have only recently begun to emerge.

The gut contains its own enteric nervous system — sometimes called the "second brain" — with roughly 500 million neurons, more than the spinal cord. The vagus nerve provides a major communication pathway between the gut and the brainstem, carrying roughly 80% of its fibres as afferent (gut-to-brain) rather than efferent (brain-to-gut) signals.

The gut microbiota — the trillions of bacteria, archaea, fungi, and viruses inhabiting the gastrointestinal tract — influence brain function through multiple pathways: (1) neural (vagal afferent signalling), (2) immune (microbial metabolites modulating inflammatory cytokines that affect the brain), (3) endocrine (microbial production or modulation of neurotransmitters, including serotonin — approximately 95% of the body's serotonin is produced in the gut), and (4) metabolic (short-chain fatty acids produced by bacterial fermentation of dietary fibre have neuroactive properties).

Germ-free mice (raised without any microbiota) show altered stress responses, impaired memory, and changes in anxiety-like behaviour compared to conventionally raised mice. Transferring gut microbiota from humans with depression to rats can induce depressive-like behaviours in the animals. These findings are provocative but must be interpreted cautiously: mouse models of psychiatric conditions have limited validity, and the leap from rodent behaviour to human mental health is substantial.

The gut-brain axis also has important limitations as a framework. Most human studies are correlational. The complexity of the microbiome (hundreds of species, interacting with diet, medication, and host genetics) makes causal inference extremely difficult. Popular claims about "healing your gut" to cure depression or anxiety vastly outpace the evidence. The field is exciting but young, and replication failures are common.

Consciousness and the brain

Consciousness poses what many consider the deepest problem in neuroscience. How and why does subjective experience arise from physical processes in the brain? Why does it feel like something to see red, taste coffee, or be in pain?

The hard problem of consciousness (distinguished from "easy" problems of explaining cognitive functions by David Chalmers in 1995) asks why neural processing is accompanied by phenomenal experience at all. Why are we not "philosophical zombies" — beings that process information and behave appropriately but have no inner experience?

Neuroscience has made progress on the "easy" problems — the neural correlates of consciousness (NCC). These are the minimal brain states that correlate with conscious experience. Several candidates have been proposed:

• Global Workspace Theory (GWT), proposed by Bernard Baars, holds that consciousness arises when information is broadcast widely across the brain via a "global workspace" — a network involving prefrontal and parietal cortices. Information processed locally and unconsciously becomes conscious when it enters this broadcast, making it available to multiple cognitive systems simultaneously.

• Integrated Information Theory (IIT), proposed by Giulio Tononi, attempts to quantify consciousness using a measure called $\Phi$ (phi), representing the amount of information generated by a system above and beyond its parts. On this view, consciousness is a fundamental property of systems with high integrated information, and any such system — biological or artificial — has some degree of consciousness. IIT makes specific, controversial predictions (it implies, for instance, that the cerebellum, despite having four times more neurons than the cerebral cortex, contributes little to consciousness because its parallel architecture generates low $\Phi$).

• Higher-Order Theories hold that a mental state is conscious when it is the object of a higher-order representation — when the brain represents itself as being in that state. On this view, consciousness is fundamentally about self-monitoring.

These theories are not mutually exclusive in all respects, and no consensus exists. The field is further complicated by the difficulty of studying consciousness empirically. The most productive experimental approach uses contrastive analysis: comparing brain activity when a participant is conscious of a stimulus versus when the same stimulus is presented but remains unconscious (as in masking paradigms, binocular rivalry, or the attentional blink). The differences between these conditions isolate the neural correlates of conscious awareness.

A critical perspective is necessary here. The neuromania critique applies with special force to consciousness research. Colourful brain images of people meditating, falling in love, or making moral decisions are visually compelling but tell us very little about why these experiences feel the way they do. The hard problem remains hard precisely because it may not be solvable by the methods of natural science — a position articulated by philosophers like Thomas Nagel and Colin McGinn, and hotly contested by neuroscientists like Daniel Dennett and Christof Koch.

Connections Master

Neuroscience to philosophy of mind

Neuroscience intersects philosophy of mind at every level. The mind-body problem — how mental states relate to physical brain states — is directly informed by neuroscientific evidence. Identity theory (mental states are brain states), functionalism (mental states are defined by their functional roles, not their physical substrate), and eliminative materialism (our folk-psychological vocabulary of "beliefs" and "desires" will eventually be replaced by neuroscientific descriptions) all draw on neuroscience for support. The hard problem of consciousness connects directly to the philosophy unit on consciousness and qualia (20.06.01).

Neuroscience to biology

The unit on cellular electrophysiology (17.09.01) provides the biophysical foundation for everything in this neuroscience unit. The Hodgkin-Huxley model is not unique to neurons — cardiac myocytes use analogous voltage-gated ion channel mechanisms, producing the cardiac action potential with its characteristically long plateau phase. The principle that electrical excitability arises from voltage-dependent membrane permeability is universal across excitable cells. Genetics and epigenetics link to evolutionary biology and the unit on the modern synthesis.

Neuroscience to ethics

Neuroscience raises ethical questions that ethics must address. Neuroethics encompasses both the ethics of neuroscience (e.g., the permissibility of cognitive enhancement, the privacy implications of brain imaging) and the neuroscience of ethics (e.g., what brain imaging reveals about moral decision-making, free will, and responsibility). The finding that moral judgments involve emotional brain regions (Greene et al., 2001) has been used to argue both for and against the rationality of deontological ethics — a use that arguably overinterprets the imaging data.

Neuroscience to methodology and statistics

The statistical challenges of fMRI research illustrate broader issues in scientific methodology. The "dead salmon" study by Bennett et al. (2009) — which found "significant" brain activity in a dead Atlantic salmon when proper multiple-comparison correction was not applied — became a landmark demonstration of why statistical rigour matters. The replication crisis in psychology intersects with neuroscience: many brain-imaging findings have small sample sizes, flexible analysis pipelines, and publication bias favouring positive results.

Historical and philosophical context Master

Phrenology and the localisation debate

The quest to map mental functions onto brain structures has a long and problematic history. Phrenology, developed by Franz Joseph Gall in the early nineteenth century, held that personality traits and cognitive abilities could be read from the shape of the skull — the idea being that well-developed brain "organs" would produce corresponding skull protrusions. Phrenology was pseudoscience: the measurement methods were subjective, the proposed brain-behaviour mappings were based on anecdotal evidence, and the entire enterprise was used to justify racist and classist hierarchies.

But phrenology's core intuition — that different brain regions serve different functions — was not entirely wrong. Paul Broca's 1861 discovery that damage to the left inferior frontal gyrus produces speech production deficits (Broca's aphasia) provided the first rigorous evidence for functional localisation. The tension between localisationist and holistic views of brain function has persisted: the truth lies somewhere between, with functions being both regionally specialised and network-distributed.

The lobotomy era

The history of neuroscience includes one of the most disturbing chapters in medical history: the lobotomy. Developed by Portuguese neurologist Egas Moniz in 1935 and popularised in the United States by Walter Freeman, the prefrontal lobotomy involved severing connections between the prefrontal cortex and the rest of the brain. Freeman's "ice-pick lobotomy" — hammering an instrument through the eye socket and sweeping it side to side — was performed on an estimated 40,000-50,000 Americans between the late 1930s and the 1960s.

The procedure was presented as a "treatment" for mental illness, including schizophrenia, depression, anxiety, and even behavioural problems in children. Some patients became calmer. Many were left profoundly disabled — emotionally blunted, cognitively impaired, unable to function independently. Howard Dully, lobotomised at age 12 at Freeman's hands simply for being "difficult," documented his experience decades later.

Moniz received the Nobel Prize in Medicine in 1949 for developing the procedure. The Nobel has never been revoked. The lobotomy era illustrates several recurring patterns: the seduction of a seemingly simple biological "fix" for complex problems, the willingness to inflict irreversible harm on vulnerable populations (women, children, racial minorities, and the institutionalised were disproportionately lobotomised), and the lag between enthusiastic adoption and critical evaluation.

Modern psychosurgery — deep brain stimulation (DBS), cingulotomy for refractory obsessive-compulsive disorder — is far more precise and is subject to rigorous ethical oversight. But the lobotomy history serves as a caution: the capacity of neuroscience to do harm in the name of treatment is not merely historical.

Animal research and its discontents

Much of neuroscience relies on animal models. Hodgkin and Huxley used the giant squid axon. Hubel and Wiesel used cats and monkeys. Modern studies of neural circuits, genetic manipulation of neural activity (optogenetics, chemogenetics), and pharmacology depend heavily on rodent models. The ethical status of animal research is contested. Animals cannot consent, and the procedures — brain surgery, genetic modification, behavioural testing under stress — are invasive. The scientific justification is that animal models provide causal evidence impossible to obtain in humans. The ethical question is whether this justification is sufficient. Reasonable people disagree.

The "neuromania" critique

The neuroscientist and philosopher Max Bennett and the philosopher Peter Hacker, in their 2003 book Philosophical Foundations of Neuroscience, levelled a comprehensive critique at the field. Their central argument is that many neuroscientific claims commit a mereological fallacy — attributing to parts of the brain (neurons, regions, circuits) properties that can only meaningfully be attributed to the whole person. The brain does not "see," "decide," or "believe" — the person does, using their brain.

Sally Satel and Scott Lilienfeld, in Brainwashed (2013), targeted the popular appropriation of neuroscience: the use of brain images to lend false authority to claims about consumer behaviour, political attitudes, criminal responsibility, and educational methods. Brain images are visually compelling and create an illusion of explanatory depth — a phenomenon called the "seductive allure of neuroscience explanations." Studies by Weisberg et al. (2008) showed that even irrelevant neuroscience information makes explanations of psychological phenomena seem more convincing to non-experts.

The critique is not that neuroscience is invalid. It is that neuroscience is one level of explanation among several, and reducing complex human phenomena to brain activity — without also addressing the psychological, social, cultural, and historical levels — produces a distorted picture.

WEIRD populations and generalisability

Most neuroscience research is conducted on a narrow subset of humanity: Western, Educated, Industrialised, Rich, Democratic (WEIRD) populations. Henrich, Heine, and Norenzayan (2010) demonstrated that WEIRD populations are among the least representative of humanity as a whole on many psychological dimensions. Brain imaging studies compound this bias with additional constraints: participants must be able to lie still in a scanner, have no metal implants, and tolerate the loud, confined environment — conditions that exclude many populations.

Neuroscience findings that are presented as universal truths about "the human brain" may in fact describe patterns specific to a small, unrepresentative slice of humanity. Cross-cultural neuroscience is a growing but still underdeveloped field. The assumption that neural mechanisms identified in WEIRD undergraduates generalise to all humans is exactly that — an assumption, and one that deserves more scrutiny than it typically receives.

Bibliography Master

Foundational papers:

• Hodgkin, A. L. and Huxley, A. F. — "A quantitative description of membrane current and its application to conduction and excitation in nerve," Journal of Physiology 117, 500-544 (1952). The quantitative model of the action potential.
• Hubel, D. H. and Wiesel, T. N. — "Receptive fields of single neurones in the cat's striate cortex," Journal of Physiology 148, 574-591 (1959). Originator work on visual cortical organisation.
• Sperry, R. W. — "Cerebral organization and behavior," Science 133, 1749-1757 (1961). Summary of split-brain research.
• Bliss, T. V. P. and Lomo, T. — "Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path," Journal of Physiology 232, 331-356 (1973). Discovery of LTP.

Split-brain and consciousness:

• Gazzaniga, M. S. — "The split-brain in man," Scientific American 217, 24-29 (1967).
• Gazzaniga, M. S. — The Consciousness Instinct: Unraveling the Mystery of How the Brain Makes the Mind (Farrar, Straus and Giroux, 2018).
• Nagel, T. — "What is it like to be a bat?", Philosophical Review 83, 435-450 (1974).
• Chalmers, D. J. — "Facing up to the problem of consciousness," Journal of Consciousness Studies 2, 200-219 (1995).
• Tononi, G. — "An information integration theory of consciousness," BMC Neuroscience 5, 42 (2004).

Neuroplasticity:

• Merzenich, M. M., Nelson, R. J., Stryker, M. P., Cynader, M. S., Schoppmann, A., and Zook, J. M. — "Somatosensory cortical map changes following digit amputation in adult monkeys," Journal of Comparative Neurology 224, 591-605 (1984).
• Draganski, B., Gaser, C., Busch, V., Schuierer, G., Bogdahn, U., and May, A. — "Changes in grey matter induced by training," Journal of Neuroscience 24, 2749-2751 (2004).

Nature-nurture and epigenetics:

• Meaney, M. J. and Szyf, M. — "Environmental programming of stress responses through DNA methylation: life at the interface between a dynamic environment and a fixed genome," Dialogues in Clinical Neuroscience 7, 103-123 (2005).
• Caspi, A., Sugden, K., Moffitt, T. E., Taylor, A., Craig, I. W., Harrington, H., McClay, J., Mill, J., Martin, J., Braithwaite, A., and Poulton, R. — "Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene," Science 301, 386-389 (2003).
• Rutter, M. — "Gene-environment interdependence," Developmental Science 10, 1-12 (2007).

Gut-brain axis:

• Cryan, J. F. et al. — "The microbiota-gut-brain axis," Physiological Reviews 99, 1877-2013 (2019).
• Mayer, E. A., Knight, R., Mazmanian, S. K., Cryan, J. F., and Tillisch, K. — "Gut microbes and the brain: paradigm shift in neuroscience," Journal of Neuroscience 34, 15490-15496 (2014).

History and ethics:

• Valenstein, E. S. — Great and Desperate Cures: The Rise and Decline of Psychosurgery and Other Radical Treatments for Mental Illness (Basic Books, 1986).
• Dully, H. and Fleming, C. — My Lobotomy: A Memoir (Crown, 2007).
• Bennett, M. R. and Hacker, P. M. S. — Philosophical Foundations of Neuroscience (Blackwell, 2003).
• Satel, S. and Lilienfeld, S. O. — Brainwashed: The Seductive Appeal of Mindless Neuroscience (Basic Books, 2013).

Statistics and methodology:

• Bennett, C. M., Baird, A. A., Miller, M. B., and Wolford, G. L. — "Neural correlates of interspecies perspective taking in the post-mortem Atlantic salmon: an argument for proper multiple comparisons correction," Journal of Serendipitous and Unexpected Results 1, 1-5 (2009).
• Button, K. S., Ioannidis, J. P. A., Mokrysz, C., Nosek, B. A., Flint, J., Robinson, E. S. J., and Munafo, M. R. — "Power failure: why small sample size undermines the reliability of neuroscience," Nature Reviews Neuroscience 14, 365-376 (2013).
• Henrich, J., Heine, S. J., and Norenzayan, A. — "The weirdest people in the world?", Behavioral and Brain Sciences 33, 61-83 (2010).

Textbooks:

• 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).
• Bear, M. F., Connors, B. W., and Paradiso, M. A. — Neuroscience: Exploring the Brain, 4th ed. (Wolters Kluwer, 2016).
• Gazzaniga, M. S., Ivry, R. B., and Mangun, G. R. — Cognitive Neuroscience: The Biology of the Mind, 5th ed. (Norton, 2019).
• Purves, D., Augustine, G. J., Fitzpatrick, D., Hall, W. C., LaMantia, A.-S., McNamara, J. O., and Williams, S. M. — Neuroscience, 6th ed. (Sinauer, 2018).