When reinforcement principles are applied properly, they can be surprisingly effective. Examples discussed include improving classroom behavior, a token economy to reduce injuries in coal mines, reducing drug use, and Lovaas’s treatment for autism. One limitation is that when rewards are discontinued, the behavior may extinuish. To encourge persistence, and thus allow more time for reinforcers in the natural environment to take over, strategies include partial reinforcement during training, reinforcing behavior in a variety of settings, and fading reinforcement gradually. Another potential problem is that rewards can undermine a child’s intrinsic motivation, especially if the child feels coerced. One way to minimize this danger is to start with the mildest reinforcement likely to be effective, ususally praise, and to reinforce achievement rather than obedience. Another strategy is to train children to control their own behavior. Skinner argues that apparent failures of will-power—for example, difficulty dieting—are actually due to reinforcement contingencies that favor immediate gratification over long-term interests. One way to combat this is to introduce more immediate reinforcement for the desired behavior; another is to learn coping responses such as distraction. These techniques have been very effective in helping people lose weight and improve studying.