Yes very fair point SASless. I was looking at it from a stuck/fixed governor perspective where the first thing you will see is a TQ mismatch caused by lever inputs, as opposed to run away up/down leading to an Nr change without collective input highlighting there is no one size fits all approach, but your Nr first (and comparison against high/low engine) does make absolute sense.
To diagnose any situation accurately requires a comprehensive understanding of the potential variables which is why I support Jelico’s curiosity, particularly as he has identified he is looking for ‘generalisations’. Whilst we can’t give a ‘this is what you do’, I think it is useful to hypothesise on some basic considerations which may make the RFM instructions easier to adhere to, and form part of the discussion with the instructor who is teaching you the type specific considerations.