I'm curious how one would even do control for this. The whole point of this treatment is to deliver sensations. How would you pretend delivering sensations in a way that can't be easily told from actual sensations?

You could pick something like shocking the fingertips. A sensation, but not one expected to quiet the chronic ringing.

It might be a little much, but maybe you could numb the area? (although this might eliminate the effect...) or maybe use something like laughing gas to put the patient semi-out (you probably wouldn't want general anesthesia, since that's risky)?

If I could get a prescription for some quality placebos, then I'd be on board as long as I get relief.

I just sent you some, they were administered wirelessly, you're welcome.

Not saying this is the same, but often medical practices cannot be tested this way. Would it be ethical to perform surgeries on a bunch of people, but not actually do the surgery on a portion of them?