I'm going to explain to you exactly what I do to start a SAR dog, which is a combination of both scent imprinting on humans and a drive building exercise in the form of a runaway. 

Remember my philosophy on definitions....depends on where you're standing at the time on what the word means.

9 times out of 10, I use food as a reward for a wilderness dog. This is how I start the dog:

1) The handler holds the dog. 
2) The runner has a Ziploc baggie of nice juicy stinky dog lovin' hot dogs. 
3) The runner stands in front of the dog and feeds the dog a couple
    pieces of hotdog. 
4) The runner then teases the dog with a piece and just as the dog goes
     to eat it, the runner runs away and hides behind an obstacle (tree etc). 
5) The dog runs to the runner and gets the jackpot...three or four pieces
    of hotdog. 

We repeat this exercise two more times. This set is repeated maybe three times at the most over a week and a half and then the dog doesn't see the runner leave. So all together, we have three training sessions. During those three runaways, we extended the length so the dog would have to put down his nose naturally to search for scent. Once we see that, which normally happens around the third session, the dog doesn't see the runner leave again.

By the fourth training session, the dog is running a "hot" trail without seeing the victim leave. Once the dog put his nose down, he has exhibited "hunt" drive and now we build from there. So actually, I spend very little time on runaways, even though that is how I started the dog.

Jonni Joyce