Welp, good news and bad news I believe... I got home last night and after unpacking, went straight to the bar and got to sampling. It turned out to be calamitous. Apparently either my regulator is dying or I bumped something, because the gas pressure was set to about 3psi. As a result, the beers were drastically undercarbed. I reset the regulator and will see what it looks like tonight.
Another issue, which is the more important one- I think* my 68 degree sample picked up an infection somewhere

Either that, or it miraculously picked up some very distinct off-flavors in a short amount of time.
The 60 and 64 samples tasted very similar, with the 60 sample being separated from the 64 as more floral on the nose and more subdued in hop bitterness. The difference is very slight, however, and I doubt an average Joe Q. Drinker could tell the difference.
Now onto the 68 sample... It poured as one would expect, but it picked up a little bit of a fruity nose when chilled. Tasting revealed something completely different from initial appearance and prior flat tastings. Upon first sip, I immediately pulled a prominent-yet-not-overpowering apple/pear/stonefruit taste. It only got more pronounced as the sample warmed. It was almost as if the beer was cut with 90% beer/10% apple juice. It wasn't tart like any brett infection I've tasted, but just very fruity in nature. I won't lie- the sample wasn't terrible, but it definitely wasn't true to style... I honestly have no idea where it came from. I'm not sure if it's an infection or some sort of ester profile that gets exaggerated with chilling or what... But all I know is that is was not like the other two samples at all. I've made fruit beers that had less fruit taste, not gonna lie.
I really am pretty convinced it isn't the yeast, especially because I didn't pick anything up in prior samples. I wonder if I didn't sanitize my corny well enough- it was a new-to-me keg that hadn't been racked into yet...
Anyways, I'm going to give it more time to carb up at proper serving pressures, and re-test. I just fear that I may have lost one of the samples to accurately compare to, unfortunately.