chose to play with the robot that had been forbidden to them earlier.
Freedman’s severe threat, which had been so successful six weeks before,
was almost totally unsuccessful when he was no longer able to back it
up with punishment.
But Freedman wasn’t finished yet. He changed his procedure slightly
with a second sample of boys. These boys, too, were initially shown the
array of five toys by Freedman and warned not to play with the robot
while he was briefly out of the room because “It is wrong to play with
the robot.” But this time, Freedman provided no strong threat to
frighten a boy into obedience. He simply left the room and observed
through the one-way mirror to see if his instruction against playing
with the forbidden toy was enough. It was. Just as with the other sample,
only one of the twenty-two boys touched the robot during the short
time Freedman was gone.
The real difference between the two samples of boys came six weeks
later, when they had a chance to play with the toys while Freedman
was no longer around. An astonishing thing happened with the boys
who had earlier been given no strong threat against playing with the
robot: When given the freedom to play with any toy they wished, most
avoided the robot, even though it was by far the most attractive of the
five toys available (the others were a cheap plastic submarine, a child’s
baseball glove without a ball, an unloaded toy rifle, and a toy tractor).
When these boys played with one of the five toys, only 33 percent chose
the robot.
Something dramatic had happened to both groups of boys. For the
first group, it was the severe threat they heard from Freedman to back
up his statement that playing with the robot was “wrong.” It had been
quite effective at first when Freedman could catch them should they
violate his rule. Later, though, when he was no longer present to observe
the boys’ behavior, his threat was impotent and his rule was, con-
sequently, ignored. It seems clear that the threat had not taught the
boys that operating the robot was wrong, only that it was unwise to do
so when the possibility of punishment existed.
For the other boys, the dramatic event had come from the inside, not
the outside. Freedman had instructed them, too, that playing with the
robot was wrong, but he had added no threat of punishment should
they disobey him. There were two important results. First, Freedman’s
instruction alone was enough to prevent the boys from operating the
robot while he was briefly out of the room. Second, the boys took per-
sonal responsibility for their choice to stay away from the robot during
that time. They decided that they hadn’t played with it because they
didn’t want to. After all, there were no strong punishments associated
with the toy to explain their behavior otherwise. Thus, weeks later,
Robert B. Cialdini Ph.D / 73