This question was a lesson in logic for me. The researchers that concluded prehistoric birds are warm-blooded were, in effect, saying:

If dense blood vessels → warm-blooded

In order to wreck this argument, E negates this conditional. E says, no, having dense blood vessels is NOT sufficient to know an animal is warm-blooded. The proof being that:

Animals with dense blood vessels ←s→ not warm blooded.

This goes back to JYs lessons about how to negate conditional statements. To negate them, you negate the relationship. You say that thing thought to be sufficient is actually not sufficient by showing that there are some cases of the sufficient term that happen WITHOUT the necessary term (and if the sufficient can happen without the necessary its not so necessary is it?) This negation of the conditional relationship is exactly what E does.

I had this same question when my study partner and I started meeting, but we've found a process that works!! First we pick a PT to both take under normal timed conditions. Then we each blind review on our own except we don't look at the answers after. Then, when we meet, we compare answers. On any questions where our answers diverge or we had questions, we discuss the question and argue for our answer/against the others. This is a kind of second blind review where we often change our answers again. and don't look at the answers.. This is a kind of second blind review process, and most of the time after each of us has argued for our answer we end up agreeing on one answer. This gives added insight, and there have only been a couple instances where we got a question wrong after the second blind review. We've done it for 3 full PTs so far and have had success with it. Good luck!