Well, one last point and I'll leave this alone
I have a better idea of your experiment now. So a quick comment
Test 1: This is a toughie, I won't be surprised that most people would get a whole mess of this one.
Test 2: This is pretty much the same as the Foobar ABX. I personally like that tool, give it a go if you've not.
Test 3: Still tough. I probably would fail high quality lossy tests (e.g. APE vs API).
One should also keep in mind that listening fatigue is also an issue. When you are focussing on looking for difference, it is certainly more tiring. One could say then argue that there is no point in lossless since you probably don't pay enough attention to tell the difference (especially past a point).. but in general, I think that there is not much to gain from settling for anything less. Rip it once, have the freedom to transcode, know you have a perfect backup if you lose your damage your CD, etc.
Personally, I think it is best to start with Test 2. If people fail that one, then it's game over already. However, if someone can cope with Test 2 between lossy and lossless (again, this are the main thing I focus one), I am inclined to say they can hear a difference, even though it might be a tiny detail that won't be noticed unless you look for it. At that stage, they can decide whether it is worth to go lossless or not. In my case, I choose to pay extra space for my music and keep two copies of my collection. I will only need to do this once and it is one aspect I can know for sure is not limiting my system