Oral exams
Stations written from one board's own notes, sat out loud, and marked by the examiner you set for the subject.
The Oral tab sits on every board, beside Cards, Exam and Lecture. It holds that board's stations and the record of every attempt you have made at them. Nothing on it is generated until you ask for it.
Stations come from the board#
A station is one question, with the answer the marker is listening for. Making one takes two steps, and you see the result of the first before committing to the second.
Suggest stations reads the board — all of it, or only the sections you tick — and proposes a handful of questions, each with a line on what it examines. That is the cheap pass: it writes titles and angles, nothing more. The proposals are not saved as stations; they sit on the tab until you do something with them, and they survive leaving the page and coming back.
Generate writes one proposal out in full: the examiner's opening question, the material behind it, a suggested time to speak for, and the must-not-miss core it will be marked against.
The part that matters is where the material comes from. A station is written from the notes on this board — the sections you wrote, imported, pasted or had drafted, plus the text read out of any pictures on it. It is not written from the board's name. Ask for stations on a board about melanoma and you get questions about what your board says about melanoma, including the parts you got wrong in it.
There is no shared station library any more. An earlier version had one, generated from a topic's name and pooled across boards, which meant a board full of your own notes examined you on somebody else's material. Station-making now lives on the board it draws from, and only there.
A board's stations end up looking like this:
Recorded#
Start on a station — or Sit again, if you have sat it before — opens the recorded viva. The station's opening question is on screen, the microphone is the only control, and a timer runs against the time the station suggests.
When you stop, four things happen in order: the recording is uploaded, it is transcribed, the transcript is marked, and the feedback is read back to you in the examiner's voice. The audio, the transcript and the marking are all saved together.
The marking is your examiner's. The criteria, what they weigh and the words they use come from the examiner set for the subject, so the same answer marked by a different examiner gets a different sheet. That is deliberate, and it is why an examiner has to be set before the tab will write you anything. Each criterion carries a score, the marker's reasoning, and the one thing that would have earned the rest.
Examiner
A 45-year-old woman has a pigmented lesion on her back that her partner says has changed over six months. Talk me through your approach.
You asked about change, and about family history. You did not ask about previous skin cancers or immunosuppression, and both change the pre-test probability here.
ImproveAdd prior keratinocyte cancers, transplant or immunosuppressive therapy, and cumulative sun exposure to the standing set of questions.
The ABCDE run-through was ordered and complete, and you examined the rest of the skin and the draining nodes rather than only the lesion.
ImproveSay what you are looking for dermoscopically — atypical network, blue-white veil, irregular vessels — rather than that you would use a dermatoscope.
Excision biopsy with a 2 mm margin was right, and you said why a partial biopsy risks understaging.
ImproveName the orientation and the depth: down to subcutaneous fat, along the lines of lymphatic drainage on a limb.
You said the word melanoma out loud rather than around it, and checked what she already suspected.
ImproveGive her the timeline she asked for — when the histopathology comes back, who calls, and what happens if it is benign.
Follow-up was mentioned but not arranged, and safety-netting was general.
ImproveSay who is responsible for the result and what she should do if she has not heard in two weeks.
A strong answerI would take a history focused on the change she has noticed, its speed, and any bleeding, itch or crusting. I would ask about prior skin cancers, family history of melanoma, immunosuppression and her cumulative sun exposure, including childhood burns. On examination I would describe the lesion by ABCDE, examine it dermoscopically for an atypical network, blue-white veil and irregular vessels, then examine the whole skin and the draining lymph nodes. My working diagnosis is melanoma until proven otherwise, with an atypical naevus and a pigmented basal cell carcinoma as differentials. I would arrange an excision biopsy with a 2 mm margin, down to subcutaneous fat and oriented along the lines of lymphatic drainage, and I would explain to her that the margin depends on what the histopathology shows.
The transcript is kept as spoken. The phrases that cost you marks are tinted in it, so the criticism sits on the words rather than in a paragraph underneath them.
So the first thing I would want to know is how quickly it has changed, and whether it has bled or become itchy. I would examine it and use a dermatoscope. Then I would arrange a biopsy and refer her on. And I would explain to her that we are checking for a skin cancer and that we will have an answer within a fortnight.
I would arrange a biopsy and refer her on
Which biopsy is the whole question. A partial or shave biopsy of a suspected melanoma can understage it, and the marker is listening for you to say so unprompted.
Say: Excision biopsy with a 2 mm margin, down to subcutaneous fat, oriented along the lines of lymphatic drainage — a partial biopsy risks understaging.The Must not miss core button beside Start opens the checklist the station is marked against, to read straight through before you sit. The same checklist is reachable from inside the sitting, and opening it there is counted on the attempt — so a score you read your way to stays distinguishable from one you did not.
Attempts#
Every attempt is kept. Nothing is overwritten by a later sitting and nothing expires.
Each station's row carries how many times you have sat it, your best, your average, your average speaking time against the station's target, and when you last sat it. Expanding the row lists the individual attempts; opening one gives you the full transcript, the marks criterion by criterion, and the spoken feedback again. The four tiles at the top of the tab are the same numbers for the board as a whole.
The reason to keep all of it is the resit. Sitting the same station again a week later is the point of the surface, not a fallback for a bad score: the second attempt is where you find out whether the feedback changed what you say, or whether you only agreed with it at the time. Best and average sit side by side for the same reason — a best you have not repeated is a different situation from an average that has moved.
Every attempt attributes to a station. There is no free-form mode that produces a score, because an attempt with no station has nothing to file itself under and nothing to compare itself to next week. If you want to be asked something the board has not produced yet, write the station first, then sit it.