Isn't the smell fishy, as in primarily coming from trimethylamine from reduction of trimethylamine oxide? In that case, it would not be that closely linked to oxidation, though it might still correlate with it (as they both increase with storage).
I don't have a citation (well, unless "personal communcation" is accepted). It was stated in a presentation about stability of fish oils, but that is not solid enough for the confidence in my original post. I am sorry for that.
With regard to the manuscript you linked to about how the oxidation state of fish oils affects the lipid profiles, it is a small study (52 participants, 17-18 in each group) and it seems that they did not correct for multiple comparisons (but I might be missing that). Looking at their numbers, it seems that the largest effect is that the group that got the good oil started out with higher blood pressure (systolic and diastolic), lower blood glucose and higher cholesterol, and than normalized on these measures, except for cholesterol, where they more than normalized. I am not even sure this meats the "hypothesis-generating" state.