For many years the annual trip to the optician, courtesy of my sub-optimal eyesight, was a routine chore that was always accompanied with a nagging sense of failure.
If you've ever had your eyes tested you probably know the feeling. Does option 1 make it clearer or darker, or option 2? Is A crisper or B? And no matter what you answer it always feels like, because you are conscious of your ocular faults, it feels like you're wrong.
When it comes to designing systems - be it functional design, or the user experience - there's always that same risk. The person answering the question feels like they are somehow "wrong" with their answers. And that makes them hesitant to elaborate.
At my latest eye exam - as it turns out for Lasik - I mentioned that awkward feeling to the optician and they told me something that I wish I'd heard on one of my first visits. "The answers are not right or wrong, they're just parts of the diagnostic process". Option1 or 2, A or B. It doesn't matter. What matters is the differential data it provides the questioner to help refine their understanding of the problem and deliver a correct solution.
My takeaway from this realization is simple. Before asking someone what they need from a system it is important to help them understand that there are no "right" or "wrong" answers, but any responses (and the weight that they give those responses) simply helps provide data to inform the decision. In fact, as a good diagnostician the answers will simply help guide the discovery process rather than just delivery a binary result.
Of course, that still means that the questioner is going to have a subtle, nagging sensation that they're not asking the right questions... but that's probably a good thing as that will encourage them to keep refining their approach, and continue to iterate, as they learn more, both from individual interviews but also repeated experiences.
Oh, and the Lasik operation ... amazing. Wish I'd stepped out of my comfort zone and done it many years ago.