Store structures as circuits
For every structure your course names, answer four questions. If you can't, you have a name and not a fact.
- What projects into it? Inputs.
- Where does it project to? Outputs.
- What computation or role does it perform? In one plain sentence.
- What happens when it's damaged? The deficit — this is what the exam asks about.
The fourth is the highest-value and the most neglected. Lesion questions dominate neuroscience exams because they test whether you know what a structure does rather than where it sits, and a student who has only memorised locations can't answer them at all.
Learn pathways as journeys
Sensory and motor pathways are the backbone of most courses, and they're learnable as a small number of journeys with three facts each: where it crosses the midline, where it synapses, and what a lesion above or below each point produces.
- 1
Draw each major pathway from receptor to cortex, from memory
Spinothalamic, dorsal column, corticospinal, and whichever else your course emphasises. Weekly, on blank paper. The drawing is the revision.
- 2
Mark the decussation explicitly
Where it crosses is the single most examinable fact about a pathway, because it determines whether a deficit is ipsilateral or contralateral. Getting the side wrong loses the whole question.
- 3
Lesion it at three levels and state the deficit
Above the crossing, at it, below it. Three sentences per pathway, and they cover a large fraction of the clinical questions you'll see.
- 4
Combine two pathways in one lesion
Real lesions don't respect pathway boundaries — a spinal hemisection hits several at once, which is why it's an exam favourite. Practise the combinations, not just the isolated tracts.
- 5
Then locate from the deficit, backwards
Given a pattern of loss, where is the lesion? This is the actual exam question and it's a different retrieval direction from the one you practised drawing.
Neurotransmitters: system, then molecule
| Learn first | Then | Finally |
|---|---|---|
| The major systems and where their cell bodies live | What each system modulates broadly | Receptor subtypes and their signalling |
| Which diseases involve which system | Which drugs act where | The pharmacology detail |
Starting with receptor subtypes — which is how the material is often presented — means memorising a table with nothing to attach it to. Starting with "dopaminergic cell bodies are in these two places, and they project to these targets, and losing one projection causes this disease" gives every subsequent detail a home.
Cellular neuroscience is a different subject
Most courses combine systems neuroscience with cellular and molecular material — membrane potentials, channel kinetics, synaptic plasticity — and these need genuinely different methods. The cellular half is quantitative and mechanistic, closer to physiology or physics than to anatomy.
- Draw the action potential with axes and ion movements, from memory, and be able to say which channel is doing what at each point.
- Perturb it: what happens if you block this channel, change this concentration, cool the axon? These are the exam questions.
- Work the equations rather than memorising them. Nernst and Goldman are derivable and understanding what each term does is worth more than the formula — see how to memorise formulas.
- Keep plasticity mechanisms separate from plasticity phenomena. Students routinely conflate the molecular cascade with the behavioural finding, and questions test the link between them.
Reading the literature
Neuroscience courses lean on primary papers more than most, and this is where a lot of time goes unproductively. You are almost never expected to know a paper in detail — you're expected to know what it showed, what technique it used, and what it doesn't establish.
So read for four things: the question, the method, the result, and the limitation. Four lines per paper. The limitation is the one that earns marks in essays, because it's what separates describing a study from evaluating it. How to read academic papers covers the general skill; the neuroscience-specific addition is that the method usually determines the limitation — correlational imaging can't establish causation, lesion studies can't isolate function cleanly, and examiners want to see you know that.
Images, and why yours should be ugly
Neuroanatomy is spatial and you do need to recognise structures on sections and scans. But recognising a beautiful textbook figure is not the skill — recognising a real, badly-oriented, poorly-contrasted section is.
So work from your department's own images and from unlabelled sections, and draw your own diagrams rather than studying polished ones. A drawing you produced badly from memory is worth several you looked at, which is the generation effect applied to something visual. Keep the polished atlas for checking, not for studying.