In 1950s, I think, there were extensive research about behavior and how it changes in response to external influences, run chiefly by F. B. Skinner. The overall

In 1950s, I think, there were extensive research about behavior and how it changes in response to external influences, run chiefly by F. B. Skinner. The overall conclusion of that was that punishing the undesirable behavior does not work; that is, it will likely change the behavior, but not in the desired way. The only reliable way to change the behavior was to reward the desirable behavior; there's also some details on how to dispense the rewards (first regularly and then over irregular intervals). Skinner even wrote an utopian-genre book Walden Two about the future society that engineers its own culture using behaviorist method. (For example, quite a few diseases can be prevented or detected early simply by undergoing regular medical observations; yet not that many people do this regularly. What if we could change our culture to shape the desired behavior? Wouldn't it result in much more efficient health maintenance?) For quite some time I regarded this as a humanist approach, but then I changed my mind and consider rewards used in this way not much different from punishments. Even if you don't try to use it manipulatively (which is obviously evil) but actually care about another person, it's still too dangerous and, I believe, eventually detrimental to the person's well-being. Besides, it's very easy to deceive yourself that you're not trying to manipulate while you actually are. Consider that, for example, many video games follow the Skinner's model to the letter: they reward the desired behavior (playing) by dispensing some virtual rewards from time to time. Not regularly, so it's not boring, but often enough so that you get the boost and continue; and from time to time they give "better" rewards (more rare or in exchange to a number of smaller ones, so they look subjectively more expensive). It's easy to see people addicted to these games.