It seems to me, from that graph, that the plant preferred ammonium until levels hit at or below 0.5, at which point it had no preference and took both in roughly equally....if we assume that ammonia levels were probably 0.1 or less, and not exactly 0.1. If we assume that ammonia was actually stable at 0.1, then it had nitrate preference by the time it got down to that level.

I think maybe the bottom line is that this graph doesn't provide enough data to make the assumptions that are being made. All it's telling us is that if you happen to have a tank with 2ppm of ammonia and 2ppm of nitrates, the plants will use much of the ammonia first. Maybe, this could be generalized to say that if you have a tank with equal amounts of ammonia and nitrates, this would occur. But what if we started with 0.5 ppm of ammonia and 20 ppm of nitrates (a more realistic situation in an aquarium)? Plants are very good at adapting to their circumstances, so my guess would be that you would see what could be interpreted as a definite nitrate preference under those circumstances....in other words, the ammonia wouldn't be reduced to near 0 before nitrate started being used, both would probably be used concurrently.

Therefore the real question is - is there a benefit to keeping nitrates low enough and reducing the biological filtration to cause the plants to take in a much larger percentage of their nitrogen as ammonia? Can plants actually take in ammonia significantly faster than nitrate? Do the plants benefit in any other way from taking in ammonia instead of nitrate? If not, there is no benefit in avoiding biological filtration (unless someone wanted to, for other reasons not having to do with the health of the tank).