What really motivates us?

Common sense tells us that in order to get someone to do something, or to get them to do it better (faster, more thoroughly, more carefully), you might offer them a reward - or if the offer is already there, increase its size. All kinds of clever-sounding people hold this view - principally economists and management consultants, but more or less anyone else too, as it seems to be so basic and so widespread an idea as to not merit further inspection.

Daniel Pink's book Drive introduces us to the research that seems to turn much of this "common sense" on its head. It turns out the rewards such as financial incentives usually make people perform worse. It's not limited to financial incentives (although Pink focuses on these), and not even limited to people.
In place of the carrots-and-sticks theory, Pink introduces us to a new theory, which is about 40 years old and yet still isn't all that well known. According to MAP theory (not the only name), the most important components of motivation are mastery, autonomy, and purpose. The author explains to us in more detail quite what each of these mean, but in short, people are motivated to do things when they can apply their existing expertise and develop it further; choose what they get to do, when, how, and with whom; and do something that they think is important, and/or connects them to other people.

One of the reasons that it's taken so long for the ideas to catch on is the prevalence not only of the "rational agent model" of economics, but also of the psychological field of behaviourism. Behaviourism, along with psychoanalysis (a very different and mostly incompatible worldview), was the leading theory in psychology in the first half of the 20th century, and remains influential today. It turns out that behaviourists aren't completely wrong that rewards can motivate behaviour, but it also turns out that rats pushing buttons isn't the same as humans doing creative work.