Neurology has a fearsome reputation among candidates, and it is almost entirely undeserved. It is fearsome only if you revise it as a list of diseases with unfamiliar names, because there are a great many of them and the names do not help. Approached as the specialty actually approaches it, neurology is the most systematically derivable subject in the exam, because two questions, asked in the right order, generate the differential before you have recalled a single diagnosis. Where is the lesion, and how did it evolve. Everything else is downstream.
Key takeaways
- The exam is two papers of 100 best-of-five questions, three hours each, with a break, and no negative marking.
- Localise before you name anything, because the location constrains the differential more than any symptom does.
- Then apply the time course, because localisation and tempo together produce the diagnosis.
- Findings that contradict your localisation are the most valuable information in the question.
- Separate the four compartments: central, peripheral nerve, neuromuscular junction, and muscle.
Where before what
The single discipline that transforms neurology performance is refusing to name a disease until you have named a place.
Read the vignette and ask: which part of the nervous system, if damaged, would produce exactly this constellation? Cortex or subcortical white matter. Brainstem, and if so at which level, because the cranial nerve involved localises it precisely. Spinal cord, and at which level, and which tracts. Anterior horn cell. Nerve root. Plexus. Peripheral nerve, and is it a single nerve, multiple individual nerves, or a length-dependent process. Neuromuscular junction. Muscle.
The signs tell you this deterministically, and that is the beauty of the specialty. Upper motor neurone signs put you above the anterior horn cell. A sensory level puts you in the cord. A crossed pattern, ipsilateral face and contralateral body, puts you in the brainstem, and it does so with more certainty than any imaging finding a candidate could guess at. Fatigable weakness with normal reflexes and sensation points to the junction. Proximal symmetrical weakness with preserved sensation points to muscle.
Do this before you consider a single disease name, and the list of possible diseases will already have shrunk to something manageable.
Then the time course
Localisation alone gives you a place. Add tempo and you usually have the diagnosis.
Hyperacute, over seconds to minutes, is vascular until proven otherwise, and in some contexts seizure.
Acute to subacute, over hours to days, suggests inflammatory, infective, or vascular processes evolving.
Subacute over weeks suggests inflammation, infiltration or a compressive process.
Chronic and progressive over months to years suggests degenerative, neoplastic, hereditary or metabolic causes.
Relapsing and remitting is its own powerful category, and it points hard in specific directions.
Episodic with complete recovery between events points elsewhere again: seizure, migraine, transient ischaemia, or a channelopathy.
Now combine the two axes. A cord lesion evolving over hours is a different differential from a cord lesion evolving over two years, and neither requires you to have remembered anything. The grid produces the list.
The contradiction is the clue
Here is the refinement that separates strong candidates.
When you have localised, look deliberately for the finding that does not fit. It will usually be there, and it will usually be the point of the question.
Preserved reflexes in a patient who looks like they have a lower motor neurone lesion. Sensory involvement in a picture that otherwise looks like pure muscle disease. Sphincter involvement where you were thinking peripherally. A cognitive feature in an apparently spinal presentation. Fatigability where you had settled on a structural cause.
Candidates skim past these because they conflict with the tidy story they have constructed, which is precisely why the examiners put them there. A finding that contradicts your localisation means either that your localisation is wrong, or that the process is not confined to one place, and either conclusion is more valuable than the diagnosis you were about to select.
Keep the four compartments distinct
Weakness is the presenting problem in a large share of neurology questions, and four quite different compartments produce it. Keeping them separate is most of the battle.
Central, with upper motor neurone signs, spasticity, brisk reflexes and an extensor plantar.
Peripheral nerve, with lower motor neurone signs, wasting, fasciculation, reduced reflexes and, importantly, sensory involvement in most causes.
Neuromuscular junction, with fatigability, preserved reflexes and sensation, and characteristic patterns of ocular, bulbar or proximal involvement.
Muscle, with proximal symmetrical weakness, preserved sensation, and usually preserved reflexes until late.
Each compartment carries its own investigations and its own treatments, so a localisation error at this level propagates into every subsequent answer in the question.
Investigation follows localisation, not the other way round
A structural point about how questions are built.
Candidates asked which investigation to order frequently choose the most sensitive or the most modern imaging. The exam wants the investigation appropriate to the localisation you have made, and if you have not made one, you cannot answer.
If the lesion is in the cord, image the cord and image the right level. If the process is at the junction, imaging the brain is a wasted test. If it is a peripheral neuropathy, neurophysiology tells you what imaging cannot, and it tells you whether the process is axonal or demyelinating, which splits the differential again.
The question "which test" is really the question "where do you think this is", asked in disguise.
Where iatroX fits
iatroX's Neurology SCE bank is built around the localisation and time-course reasoning that this exam actually tests, rather than around disease recall. Missed questions can be opened in the Socratic Tutor, which asks you to localise and reason before it explains, and which names the sign you failed to weight or the contradiction you skimmed past, so you reconstruct the anatomical reasoning rather than being handed a diagnosis you could not have reached. The adaptive engine returns the same localisation principle in a different presentation, which tests whether the reasoning transferred. Try it with free sample questions at iatroX. For the review method that turns a near-miss into a durable rule, see the distractor audit.
Frequently asked questions
How should I approach a neurology question? Localise before you name anything. Work out which part of the nervous system would produce exactly this constellation of signs, then apply the time course. Localisation and tempo together generate the differential without requiring you to recall a disease first.
Why is the time course so important in neurology? Because the same lesion in the same place has a completely different differential depending on whether it evolved over minutes, days, weeks or years. Hyperacute is vascular; chronic and progressive is degenerative, neoplastic or hereditary.
What should I do with a finding that does not fit my localisation? Treat it as the most valuable information in the question. Examiners include contradictory findings deliberately. A finding that conflicts with your localisation means either that your localisation is wrong or that the process is not confined to one site.
How do I choose the right investigation in neurology? By localising first. The question of which test is really the question of where you think the lesion is. Imaging the wrong compartment, or the wrong level of the cord, is a wasted answer even when the modality is impeccable.