3 ms·

Because when you can't win by building a better mouse-trap, you try to win by tricking the mice.

by tabtab 8y ago

AstralStorm 8y ago
If your "trick" does not actually work, you're wasting time and resources. Most of behavioral economy is not seriously tested, because the endpoint is not "tricking the mouse" but positioning yourself as a trickster educator, a place that cannot be blamed. One of classic cons.
- tabtab 8y ago
  Ideally you try different tricks in different areas as an experiment, and merge and adjust the best performers. "It's not random marketing BS, it's a genetic algorithm."
- lucas_membrane 8y ago
  > If your "trick" does not actually work, you're wasting time and resources. Many of the popularly-cited experiments both in behavioral economics and in psychology involve misleading or deceiving the experimental subjects, and many more simply conceal their objectives. The assumption that the suckers never caught on and the results should be interpreted accordingly is seldom questioned. In some cases, the assumption is verified by asking the subjects about their motivations post-mortem, ignoring the long-standing axiom of applied behavioral economics (business) that there are two reasons anyone does anything, the one they will tell you and the real one. The null hypotheses in these experiments ought to be (1) the subjects know or can figure out what the experimenters are looking for, and (2) the subjects are there because they want to help the experimenters. Consider the subject who is paid a small sum to participate and interacts with a either a researcher who must publish or perish or one who will put another batch of subjects through a similar rigmarole next year if they cannot get their dissertation accepted this year. What does motivational psychology say about the behavior of such a subject? Is such a search for truth better than a congressional committee?

Because when you can't win by building a better mouse-trap, you try to win by tricking the mice.

by tabtab 8y ago

Because when you can't win by building a better mouse-trap, you try to win by tricking the mice.

If your "trick" does not actually work, you're wasting time and resources. Most of behavioral economy is not seriously tested, because the endpoint is not "tricking the mouse" but positioning yourself as a trickster educator, a place that cannot be blamed. One of classic cons.

tabtab 8y ago
Ideally you try different tricks in different areas as an experiment, and merge and adjust the best performers. "It's not random marketing BS, it's a genetic algorithm."
lucas_membrane 8y ago
> If your "trick" does not actually work, you're wasting time and resources. Many of the popularly-cited experiments both in behavioral economics and in psychology involve misleading or deceiving the experimental subjects, and many more simply conceal their objectives. The assumption that the suckers never caught on and the results should be interpreted accordingly is seldom questioned. In some cases, the assumption is verified by asking the subjects about their motivations post-mortem, ignoring the long-standing axiom of applied behavioral economics (business) that there are two reasons anyone does anything, the one they will tell you and the real one. The null hypotheses in these experiments ought to be (1) the subjects know or can figure out what the experimenters are looking for, and (2) the subjects are there because they want to help the experimenters. Consider the subject who is paid a small sum to participate and interacts with a either a researcher who must publish or perish or one who will put another batch of subjects through a similar rigmarole next year if they cannot get their dissertation accepted this year. What does motivational psychology say about the behavior of such a subject? Is such a search for truth better than a congressional committee?