Interesting. I have now done some research on sponge filters, and I'm actually somewhat perplexed. I hope one someone can help me to understand the differences better.
It seems to me that sponge filters work the same way as a UGF -- using air or water to push water out of a tub, causing tank water to flow into the space through a filter medium. The only real difference seems to be that a UGF uses the gravel in the tank as the filter medium, and a sponge filter uses a sponge as the filter medium.
Since they seem to work on the same concept, it is not obvious to me that the sponge filter couldn't cause the current that is of concern with long-finned betas, especially if you have two of them in a tank. In fact, it seems to me that a sponge filter might cause a more powerful current since the water is being pulled through a smaller area (the area of the sponge) rather than the large area of a filter plate. But the flow pattern would be somewhat similar. In fact, it seems to me that the strength of the current for either would be determined by the size and power of the airstone.
The flow pattern around the tank for a UGF would be controlled by the location of the riser tube -- and on most of the smaller tanks I have seen, the tube is in the middle of the tank, not a corner, so the water flow would not be down one side, across the top, and up the opposite side. It would be primarily down towards the entire filter plate, up through the riser tube, and then out across the entire top of the water.
In the 1 gallon tank I had with a UGF, this seemed to be shown to be very gentle, since a drop of Quick-Cure (it was a rescue and we were desperate), which is a very distinctive color, diffused slowly throughout the tank. It did not rapidly sink, which is what a strong current would have done to it.
So this does not square with the above explanations. Maybe I am just very confused.