Still not quite perfect (Score 2, Insightful)

They've gone to great lengths to keep the first subject from marking the objects in any way to indicate which one was chosen, but this won't completely eliminate false positives.

The first subject still has to make an entirely subjective choice of objects. If the second subject knows the first subject extrememly well, it may still be possible for that person to guess which object was originally chosen just because he or she knows which object would grab the attention of the first subject.

More cynically, there's nothing to stop the subjects from creating some kind of heuristic before the test. For example, always choose the larger object or the one with the name that comes first alphabetically. Of course, I suppose you prevent this by refusing to reveal the details of the study to either participant before they are separated.