To obtain the best compromise between the speed of acquisition of a desired response and resistance ...

To obtain the best compromise between the speed of acquisition of a desired response and resistance to its extinction, which reinforcement strategy should be used in training? (In other words, how do we get the fastest learning combined with the strongest resistance to extinction?)
a. Continuous reinforcement
b. Begin with self-management and then go to modeling.
c. Begin with delayed and partial reinforcement and gradually go to immediate and continuous reinforcement.
d. Begin with immediate and continuous reinforcement and gradually go to delayed and partial reinforcement.
e. Begin with negative reinforcement and gradually change to continuous positive reinforcement.