It has been attempted, albeit with a relatively small group of subjects and mostly focused on aversive vs. positive reinforcement methods: https://journals.plos.org/plosone/article?id=10.1371/journal...

Maybe look at international competition, and see what kind of training everyone has converged on?

There’s no one training Koehler in high level IPG (IPO/schutzhund) or one of the ring sports.

I’m also not aware of actual LE/Mil doing anything Koehler. They’re all using all four quadrants of the operant conditioning.

The biggest challenge comes with the fact that you're often dealing with rescue dogs that are so wildly different from one another that you'd need a huge population and extremely consistent methods to show a meaningful result.

It's more doable if things are controlled (e.g. take a bunch of German Shepherd puppies given to very experienced trainers at birth), but those circumstances don't necessarily tell you what the "best" method is in a generalizable fashion.