In brief, hypnosis, which is supposed to help people relive past experiences,
does not really work. The research on hypnosis and memory is consistent with
the idea that records of past experiences are not routinely maintained in mem-
ory, but must be reconstructed.
The Influence of Beliefs on Memory The idea that recollecting is reconstructing
suggests that we reconstruct a memory of our past from our current beliefs and
what we believe to be true about human personality in general (Ross, 1989).
One idea that people have about personality is that beliefs remain rather stable
over time. As a result, people tend to remember that their past beliefs were
similar to their currently held beliefs, even when their beliefs have, in fact,
changed over time. Let me provide a few experimental demonstrations.
In one study (Goethals & Reckman, 1973; see also Markus, 1986), high school
students filled out a survey asking them for their opinion on various topics,
including forced busing. About two weeks later, students met with a respected
high school senior who presented a carefully crafted and well-rehearsed argu-
ment to the students about busing that was the opposite of the students ’own
opinion. For example, students who were opposed to forced busing heard a
counterargument in favor of forced busing. Following the counterargument,
students were again asked their opinion on busing, and were also asked to try
to recall how they had filled out the survey two weeks earlier. The instructions
emphasizedtheimportanceofaccuraterecall.
The counterarguments were effective; students tended to reverse their opin-
ion about busing after hearing the counterargument. The result, consistent
with reconstruction, was that the students tended to remember that they origi-
nally filled out the survey question about busing in a way consistent with their
newly formed opinion and inconsistent with the way they actually had origi-
nally answered the busing question. For example, the students who were orig-
inally opposed to forced busing but heard a persuasive argument in favor of
forced busing tended to remember that they had been in favor of forced busing
all along. It was as if the students examined their current belief about busing,
assumed that attitudes remain stable over the short period of two weeks, and
so reconstructed that they must have held their current attitude two weeks
earlier.
Galotti (1995) studied the criteria students use when selecting a college.
Galotti asked students to recall the criteria that they had listed in a previously
filled-out questionnaire assessing the basis on which they decided where to go
to college. Galotti also asked the students to describe the ideal criteria that they
thought, in retrospect, they ought to have used. The questionnaires had been
filled out 8 to 20 months earlier. Subjects recalled about half of the criteria they
had used when originally making the decision about where to go to college.
But the overlap between what they recalled and the ideal criteria was substan-
tially greater than the overlap between what they recalled and the criteria they
had actually used. It was as if subjects used their current sense of the ideal de-
cision criteria to reconstruct a memory of the criteria they used when originally
making the college decision.
Many people, at least in our culture, believe that a woman’s mood is likely
to become more negative just before and during menstruation. It turns out,
338 R. Kim Guenther