Teach the dog to do A.
Teach the dog to do B.
Teach the dog to do C.
Teach the dog to do D.
Ask for C.
Reward C by allowing (or telling) her to do D.
Reward D.
Ask for B.
Reward B by allowing her to do C.
Reward C by allowing her to do D.
Reward D.
Ask for A.
Reward A by allowing her to do B.
Reward B by allowing her to do C.
Reward C by allowing her to do D.
Reward D.
You now have a chain of behaviours which was backchained, where each behaviour gives the dog the right to do the next behaviour (or to be asked to do the next behaviour). Premack - the behaviour that came second rewards the behaviour that came first.
Official recall.
I tell the dog to Stay and walk to the other side of the room, turn, and face her. If she doesn't Stay, aw, too bad.
If she Stays, I call her. If she doesn't come, or comes before I call her, aw, too bad.
If she comes, I stand still so she can give me a Front. If she doesn't Front, aw, too bad.
If she Fronts, I build a little tension by widening my eyes and then ask for a Finish. If she doesn't Finish, or Finishes too soon, aw, too bad.
If she Finishes, PAR-TAY!
Traditional trainers spend a LOT of time correcting the dog for anticipating the pieces of this chain. MUCH easier to remember that being ALLOWED to Finish is the reward for sticking the Front. Lose the Front, I'll just walk away for a moment.
Sue.Eh