Crashing Memory Paradigm - UCI Webfiles

Memory Distortions: Individual Differences
and Paradigm Comparisons
Memory distortions have been studied
extensively over the last three decades, and several
branches of research, or paradigms, have been
Misinformation Paradigm
Misleading post-event information can distort
our memory for the original event (Loftus, 2005).
Crashing Memory Paradigm
Following the suggestion that footage of a
news-event exists, when it actually does not exist,
30-40% of people reported having seeing the
footage (Crombag et al., 1996). The event is
typically upsetting, such as an assassination or
Imagination Inflation Paradigm
Using guided imagery of a fictitious event later
produces a false memory due to source confusion
between the imagination and real events (Garry et
al., 1996).
Memory for Felt Emotion Inconsistency
Although not traditionally considered a false
memory paradigm, research has also suggested that
memory for felt emotion is reconstructed and is
changed according to current appraisals (Levine,
In this study, each subject participated in all of
these memory paradigms. We explored whether
susceptibility in one paradigm would predict
susceptibility in others.
Lawrence Patihis, Steven J. Frenda, and Elizabeth F. Loftus
Department of Psychology and Social Behavior, University of California, Irvine
Before comparing paradigms, we confirmed that all
four paradigms produced memory distortions/
inconsistencies comparable to previous research:
Misinformation effect. 47% of participants robustly
indicated at least one false memory. Experimental vs.
control groups were reliably different on false memory
endorsement (p’s < .0001).
Crashing memory. Of those familiar with the United
93 crash, 42% indicated they had seen the crash footage
in the questionnaire, but only 23% did so in the more indepth interview.
Imagination inflation. 41% changed from saying
“no,” they had not seen the footage, to “maybe” (36%) or
“yes” (5%) after the imagination exercise.
Memory for emotion. The second time they were
asked, participants remembered feeling negative emotion
less frequently after 911, and this was modified by
thinking about a plane crash beforehand, consistent with
affective adaptation and appraisal theories.
Except for one small negative correlation,
performance in each paradigm did not predict
the others. Moreover, individual differences
also did not predict the same paradigms.
However, our findings may not generalize to
all paradigms, because Otgaar et al. (2012)
found that memory distortions in word list
recall (Deese-Roediger-McDermott; DRM)
predicted rich false memories in children (ŋ²p =
The present study suggests different
mechanisms in those paradigms we looked at.
The cognitive mechanisms involved in
remembering mild slideshows from 1 hour ago
are likely different from remembering
disturbing plane crashes from 10 years ago.
393 undergraduates completed the study, (M =
20; 75% female; 50% Asian, 21% Caucasian).
All subjects completed a 2 (misinformation
effect) x 3 (crashing memory) experiment.
Misinformation effect. 2 groups served as both
experimental and control conditions for each other
on different items. Materials consisted of photo
slideshows of two simulated crimes, narratives
with embedded misleading statement 40 minutes
later, and a memory test 20 minutes after that.
Crashing memory. 3 groups: “there is footage
of the United 93 crash on September 11, 2001,” a
“no footage” control, and a neutral 2001 newsevent control. A computer questionnaire was later
followed up by an interview to test the
questionnaire. Note that due to the age of the
participants, the target event represents a
potentially unpleasant childhood memory.
Imagination inflation. In the interview, those
who initially said they had not seen the United 93
footage were taken through an exercise of
imagining the crash footage, to see if they would
then flip and report seeing the nonexistent footage.
Memory for emotion The same questions were
asked, one week apart, about how often they felt
various negative emotions in the week after
September 11, 2001 (911).
r = .09
Various individual differences questionnaires
(see Table 1) were completed on a computer in lab,
over 2 sessions, one week apart. A sleep log was
kept in the week between sessions.
memory for
r = -.11
p = .03
r = -.06
These results could imply that we would not
be able to strongly predict whether someone
may have false memories of an upsetting scene
in childhood (like our manipulation involving
nonexistent footage of a plane crash) by
measuring them on a different false memory
test, like a classic misinformation experiment.
r = -.04
r = -.02
Crombag, H. F. M., Wagenaar, W. A., & van Koppen, P. J. (1996).
Crashing memories and the problem of source monitoring.
Applied Cognitive Psychology, 10, 95-104.
Garry, M., Manning, C. G., Loftus, E. F., & Sherman, S. J. (1996).
Imagination inflation: Imagining a childhood event inflates
confidence that it occurred. Psychonomic Bulletin and
Review, 3, 208–214.
Levine, L. J. (1997). Reconstructing memory for emotions.
Journal of Experimental Psychology: General, 126, 165-177.
Loftus, E. F. (2005). Planting misinformation in the human mind:
A 30-year investigation into the malleability of memory.
Learning & Memory, 12, 361-366.
Otgaar, H., Verschuere, B.,Meijer, E. H., van Oorsouw, K., (2012).
The origin of children's implanted false memories: Memory
traces or compliance? Acta Psychologica, 139, 397-403.
with this study
Figure 1. Correlations between paradigms. Illustration of the small amount of
variance, 1% or less, that is explained by one paradigm on another. ns indicates
memory distortion/inconsistency in one paradigm does not predict the other. Memory
for emotion inconsistency predicts less susceptibility to the misinformation effect,
although the effect is small.
Our study had some limitations: the
imagination inflation manipulation had a short
1 minute interval between exercise and test, as
well as having no control group. Three of our
paradigms were related to 911, and future
research could try comparing these paradigms
with unrelated target events.
This research was helped in part by a grant from UC Irvine’s
Undergraduate Research Opportunity Program, and by a National
Science Foundation graduate fellowship to Lawrence Patihis.
We extend our thanks to the undergraduate research assistants
who helped conduct the study: Matthew Miller, Lavina Ho,
Stephany Debski, Stephanie Martinez, Neda Bozorgkhan, Hoang
Nguyen, Patricia Place, Gina Machiaverna, Alex Chindri, Monica
Aguilar, Paddy Asgari, Yesenia Orozco, Luis Garcia, Natalie Ross,
Samuel Cretcher, Devan Duenas, and Hillary Patton.
We are grateful to the Loftus graduate students who helped
with ideas during study development, including Ian Tingen and
Rebecca Nichols.

similar documents