From the article:
"The researchers trained 14 sighted and 12 blind people for between two and three hours twice a week over 10 weeks. They started by teaching participants to produce mouth clicks, then trained them on three tasks. The first two involved judging the size or orientation of objects. The third involved navigating virtual mazes, which participants moved through with the help of simulated click-plus-echo sounds tied to their positions."
Agreed. There's a lot of stuff that reminds me of teeth in this one, maybe that's why?