There is lots of research on the general effect of one-time reinforcers in
the behavioral psychology literature. I am not up on that literature any
more so I can't tell you if there has been any good work on effect size.
However, from the general studies, one thing can surely be expected: That
whatever the size of the effect (of a one-time bonus), it will lead to
some level of under-performance after the bonus is paid. This used to be
called the "post-reinforcement pause." You get the same effect on any
fixed (in time) schedule of rewards -- that is, little additional activity
until shortly before the scheduled time, followed by a lot of activity in
a brief time immediately before the reward is to given, followed by an
immediate cessation of activity for a period of time after the reward is
received.
There was a brilliant article written, in the 1970's, analyzing the
behavior of the U.S. Congress on just these grounds. It showed a
repeated, very predictable pattern of passing laws (or not passing many)
based entirely on the schedule for the end of each session.
Also, as a behavioral psychologist, I feel I must say the effect of paying
a one-time training bonus are due to the "credible promise" of paying the
bonus. That is, it is in anticipation of receiving the bonus that people
behave differently -- not as an affect of having received the bonus. I
only mention it because it means that the effectiveness of such a bonus
has a lot to do with the credibility and attractiveness of the promise
(and how it is made). One cannot simply make a bonus available, not
publicize it, not make it sound attractive, not make it sound like "the
thing to do," and expect it to have highly positive effects.
--"John W. Gunkler" <jgunkler@sprintmail.com>
Learning-org -- Hosted by Rick Karash <rkarash@karash.com> Public Dialog on Learning Organizations -- <http://www.learning-org.com>