DFC. I think you are being a bit too clever by halves when you verbalise "Smoke coming from the engine" - then wait for the reaction. If one sees smoke coming from an engine say on a light twin, one would be silly not to immediately assume that where there is smoke there is fire - and that turning off the battery or master switch would be a case of the well known pissing into the wind, so to speak.
When you set such a scenario then you must explain quite clearly what sort of smoke - black as all hell or thick or whispy and so on. The victim under test can at least have a go at identifying the possible cause of the smoke and may modify his actions appropriately. In other words the instructor should avoid the risk of being labelled a smart-arse.
If in a single engine aircraft with no engine fire extinguisher, significant smoke is spotted coming from the engine, then there is little doubt that an immediate forced landing should be considered while pondering the unknown source of the smoke.
There have been fatal accidents where delay in initiating an emergency descent has been the direct cause of incapacitation due smoke inhalation and it's all over Red Rover.
An electrical fire forward of the engine firewall must be an awfully rare event and you cannot honestly blame your colleague for not second guessing what sort of smoke you are hanging on him.
When simulating any emergency such as that which you described in your post, it behoves the testing officer (or whoever he calls himself) to make your scenario crystal clear so that the person under test has no doubt as to what you are trying to simulate. Just hollering "Smoke in the engine" means zilch and is a poor way of signalling your intentions. Buy your long suffering student a beer...