Technically, you don't need to know. All you really need to know is the method of training that produces results.
Did humanity only start exercising our muscles now that we know a bunch of stuff about the mechanics via modern science? No. What we had was eons of efforts, successes, failures, refinement, adaptation, and evolution, all of which combined to produce a variety of working training methodologies (not to be confused with mcdojos) from long before modern science got a crack at pinpointing the exact mechanics. Even after determining very specific mechanics, modern exercise didn't actually change a whole lot, despite a bunch of products being marketed as "this product changes everything!". We do now have a bunch of products that assist us, but push ups are still push ups, and stomach crunches are still stomach crunches, and running/jogging is still running/jogging.
The same thing goes for metaphysical workings, minus the whole science figuring out the specific mechanics so far. The best thing to look forward to when science determines the specific mechanics of psi will be in the refined training regime that excludes a bunch of extraneous things while including things that may currently be undersold, along with a much greater amount of certainty that people can place in their training, and the acknowledgement of multiple forms of training (ie, you don't "have" to do push ups. You can do other arm exercises instead. But you can still do push ups, if that works for you, or if you're too poor to afford multi thousand dollar exercise machines >_> ).
You don't need to know "how" or "why" of the mechanics. You just need to have a working system that you can follow. But where do working systems come from? From people who have achieved these things. Or people who lie and claim they have. How does a lay person tell the difference? Difficultly. If the person producing the system can verify it to the person who wants to learn, that goes a long way towards the learner being able to trust the system while they start doing it. If the person who created the system isn't "present" to prove it, then others who have been through the system might be able to. If they aren't present either, then the learner must try the system out for themself, see if it works for them, and if it doesn't then go onto the next system and painstakingly repeat the process over and over again until they find a system that works or they give up.