-
![]()
-
![]()
Suppose you were asked to participate in a blind taste-test of five different brands
of strawberry jam. After tasting all of the jams, but before being asked to rate their
quality, you spend a couple of minutes (
I
1 ) down your reasons for liking and
disliking each jam. Then you rate each one on a scale from 1 to 9. How accurate
would your ratings be, assuming we judged accuracy by comparing your ratings with
those given bya panel of experts assembled by Consumer Reports magazine?
When psychologists Timothy Wilson and Jonathan Schooler conducted this
experiment with college students as their subjects, they found that the ratings the
students gave to the jams had almost no resemblance to , those given by the experts.
2
They should have been able to tell which ones were good and which ones were not
the jams varied widely in quality and included those ranked 1st, 11th, 24th, 32nd, and
44th best out of 45 that Consumer Reports had reviewed. Did the students have no
taste for jam? Did their preferences differ from the experts'? Not at all. In a
separate condition of the experiment, rather than writing the reasons they liked and
disliked each jam, each subject wrote about something entirely ( 4 ): their reasons
for choosing their college major. The subjects then rated the jams, and despite not
having thought about them at all after tasting them, they made ratings that were much
closer to those of the experts.
