🧠 Neuroscience: An Introduction

What you’ll learn

1. Two Men, One Nobel, and a Fight About Whether the Brain Has PartsExplain the neuron doctrine and the Golgi–Cajal dispute, state the sourced neuron count, and identify the explanatory gap between levels that organises the course.Golgi and Cajal shared the 1906 Nobel while publicly disagreeing over whether the brain is a fused net or a population of separate cells; Cajal settled it using Golgi's own stain, which blackens a random ~1% of neurons completely. The neuron doctrine — discrete cells with dendrites, an axon, and directional flow — is the field's founding statement. Azevedo et al. (2009) replaced the long-repeated '100 billion' figure with a measured average of about 86 billion neurons, most of them in the cerebellum.
2. The SpikeDescribe the resting potential and the all-or-none action potential, and explain why neural conduction is roughly a millionfold slower than electricity in a wire.A neuron holds about −70 mV using ATP-driven ion pumps that never stop, which is much of why the brain costs ~20% of resting energy on 2% of body mass. At threshold (~−55 mV) voltage-gated sodium channels open and a self-regenerating wave sweeps the axon — identical every time, so all information lives in firing rate and timing. That wave moves at roughly 0.5 m/s unmyelinated and up to about 120 m/s with myelin, vastly slower than a wire but immune to attenuation because it is rebuilt at every step.
3. The Gap Is the PointExplain chemical synaptic transmission and why converting signal to chemistry and back is what makes computation, plasticity and pharmacology possible.Neurons approach within 20–40 nm and do not fuse; Sherrington named the junction the synapse in 1897 on functional grounds. Most synapses convert the arriving spike into neurotransmitter release and back into voltage — slower and costlier than a direct electrical junction. That chemical step is what permits inhibition as well as excitation, summation of thousands of inputs into a single threshold decision, adjustable strength (the physical basis of memory), and the molecular locks that every psychoactive drug acts on.
4. Neurotransmitters, and the Stories We Tell About ThemCorrect the 'one chemical per emotion' framing using dopamine's actual role as a prediction-error signal, and state the serotonin dispute accurately without offering medical advice.Schultz's recordings showed dopamine neurons firing to unexpected rewards, shifting to predictive cues, and going silent for fully predicted rewards — a reward prediction error, not a pleasure signal. The 'chemical imbalance' account of depression was a 1960s hypothesis that escaped into public explanation; Moncrieff et al.'s 2022 umbrella review found no consistent evidence for it, and was itself sharply contested. The distinction both sides accept is that whether a drug works and whether its proposed mechanism is correct are separate empirical questions.
5. Inhibition Is the Main EventExplain why inhibition, not excitation, is the brain's central computational problem, and why 'a region lit up' is an ambiguous statement.A purely excitatory recurrent network has no stopping condition and saturates into a uniform roar — close to what a seizure is. Roughly one in five cortical neurons is a GABA-releasing interneuron, and they sharpen (lateral inhibition manufactures the edges and contrast you see), select (the basal ganglia hold a tonic brake that must be released for movement), and time (oscillations are largely interneurons pacing excitatory populations). Because a fifth of the machinery exists to enforce quiet, increased activity in a region can mean increased suppression rather than increased doing.
6. The Brain Has MapsDescribe cortical topographic maps from Penfield's stimulation studies and explain why the homunculus is distorted — and where the 'regions have jobs' framing stops being true.Penfield stimulated the exposed cortex of awake epilepsy patients across decades and found the body laid out in order along the cortical strip — adjacent body parts on adjacent brain. The resulting homunculus is wildly out of proportion because territory is allocated by innervation density and precision of use, not anatomical size, which you can verify through two-point discrimination on a fingertip versus your back. The same topographic logic runs through vision, hearing and movement — but it holds for primary sensory and motor cortex, and does not extend to cognitive or emotional 'centres'.
7. How Memory Rewires the MachineUse H.M. and long-term potentiation to explain how memory is multiple systems implemented in adjustable synapses — and separate established plasticity from what is sold as neuroplasticity.Henry Molaison's bilateral medial temporal lobectomy in 1953 destroyed his ability to form new episodic memories while leaving skill learning intact, proving memory is several dissociable systems. Bliss and Lømo demonstrated long-term potentiation in the rabbit hippocampus in 1973, and the NMDA receptor — blocked by magnesium until glutamate and depolarisation coincide — implements Hebb's rule as a molecular coincidence detector. Plasticity is real but specific: Maguire found enlarged posterior hippocampi in London taxi drivers, while Owen et al.'s 11,000-person 2010 Nature study found brain training improved trained tasks with no transfer to untrained abilities.
8. What a Brain Scan Actually MeasuresExplain what fMRI physically measures, why every scan is a thresholded subtraction, and identify reverse inference and the multiple-comparisons problem.fMRI measures the blood-oxygen-level-dependent signal, an indirect proxy that peaks roughly six seconds after firing and averages over about a million neurons per voxel. Published images are subtractions between chosen conditions, with arbitrary colours and a chosen statistical threshold — statistical maps in the costume of photographs. Poldrack (2006) named reverse inference, showing that inferring a mental process from a region's activation fails when the region is unselective, and Bennett et al.'s 2009 dead salmon found 16 'significant' voxels of 8,064 that vanished entirely under standard multiple-comparisons correction.
9. The Myths, Killed ProperlyRefute the 10%, left-brain/right-brain and learning-styles myths with the specific evidence, and recognise the common structure of the error.The 10% myth fails metabolically (20% of resting energy on 2% of mass), clinically (no silent 90% exists) and by imaging — fMRI needs subtraction precisely because the baseline brain is fully active. Hemispheric specialisation is real (Broca 1861; Sperry's split-brain work, Nobel 1981) but Nielsen et al.'s 2013 analysis of 1,011 people found no individuals with globally dominant hemispheres. Pashler et al. (2008) found almost no studies of the design required to test the meshing hypothesis, and those that existed contradicted it; format should match the material, not the learner.
10. The Part We Genuinely Do Not UnderstandState the hard problem of consciousness, outline the leading competing theories and their adversarial test, and restate the course's through-line about the explanatory gap.Chalmers (1995) distinguished the 'hard problem' — why there is subjective experience at all — from the 'easy' problems of attention, memory, report and behavioural control that the rest of neuroscience addresses. Global workspace theory holds that consciousness is a widespread cortical broadcast winning a competition for access, while integrated information theory proposes that consciousness is quantifiable integrated information present wherever the causal structure is right. A 2023 adversarial collaboration with pre-registered predictions from both camps returned mixed results, failing some predictions of each — progress in method rather than resolution.

Questions this course answers

Grounded in trusted sources

• Kandel, Schwartz, Jessell, Siegelbaum & Hudspeth — Principles of Neural Science, 5th ed. (2013)
• Azevedo et al. — Journal of Comparative Neurology 513(5) (2009)
• Russell A. Poldrack — Trends in Cognitive Sciences 10(2) (2006)
• Bennett, Baird, Miller & Wolford — Human Brain Mapping poster (2009)
• Pashler, McDaniel, Rohrer & Bjork — Psychological Science in the Public Interest 9(3) (2008)
• Nielsen et al. — PLOS ONE 8(8) (2013)
• Owen et al. — Nature 465 (2010)
• Bliss & Lømo — Journal of Physiology 232 (1973)

Every Wunder lesson is built from real, reputable sources — never invented.

Related Science courses