Yeah, polishing the rulebook was a pretty last-minute thing so I'm sure that things could be explained better. That's probably Phase 1 is explained better than Phase 2, because we ran out of time! We playtested this way more than we have playtested any game in the past, so I suspect that we didn't explain things perfectly because we were already confident in how to play and had had to do a lot of balancing/tweaking. We'll take it away and look at it :)
In our playtests, we found that it was much more efficient to put the survey tokens down and leave them there (rather than digging up straight away) because they're worth more points at the end of the game and you can then move on to try and dig something different up - but good to get feedback from people that have played it differently.
Personally, I never took some else's research cards (and preferred to take from the table) but the other fellas used the "steal" more effectively to pinch cards that would stop someone solo-finding a Relic, or which you suspect might be one that you need to complete a set. It becomes useful particularly late-game because you might be able to work out who has the matching card that you're missing, or to stop triggering the final round(s).
Either way, useful feedback - cheers. It's by far the most complex jam game I've/we've made before, so still really proud of it!