Transcript Informing misinformation. Gerkens, D. R., Cross, A., Kline, J

Implication and Exposure: Informing Misinformation
Methods
Hypotheses
There will be a significant misinformation effect on the recall test
regardless of either type of implication.
Both forms of implication will increase the size of the misinformation
effect on the recall test.
Participants will be able to identify that misinformation items were
presented during the post-event task given source monitoring instructions.
However, source monitoring instructions will only slightly reduce the
illusion that misinformation items had appeared on the original lists.
 Both forms of implication would adversely affect the ability of source
monitoring instructions to reduce misinformation acceptance.
Figure 1. Recall of Misleading Information as a function
of Implication Conditions
Mean Proportion Recall
To-Be Remembered
Event
Math
Problems
Priming Task,
No Implication
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Presented
Non-Presented
High Overlap High Overlap Low Overlap Low Overlap
/ Implication
/ No
/ Implication
/ No
Implication
Implication
Priming Task,
Implication
Proportion Overlap / Instruction
Cued Recall
Figure 2. Source Attribution as a Function of Instruction
Discussion
Recall Test
There was a large increase in misleading item recall when presented.
However, there was a three way interaction:
For the High Overlap condition the Implication Instruction seemed to
both enhance recall of misleading items and reduce semantic based
false recall.
For the Low Overlap condition the Implication Instruction did not
have a reliable effect on either misleading item recall or semantic
based false recall.
Therefore, the affect of implication on recall was only partially
supported.
Source Monitoring Task
Participants recognized that misleading items had been on the postevent task approximately 36% of the time.
However, they realized the item had not appeared on the original lists
only about 8% of the time.
Contrary to prediction, both forms of implication appear to have
improved source monitoring accuracy.
It should be noted that the accuracy of source attributions was poor
in even the best conditions (≈ 17%).
A post-hoc explanation is that both forms of implication made
participants more aware of the possibility of confusing sources during
the post-event task and consequently they paid more attention to
source.
Conclusions
Source
Monitoring
Task
Current Study
The current experiment disentangles exposure and implication allowing
for analysis of the independent contributions of each to recall and source
attributions in a categorized word list paradigm that parallels the standard
misinformation methodology. Exposure was manipulated by presenting
or not presenting plausible category members that had not been in the
original list learning task. Explicit implication was manipulated via
instruction for the post-list learning task and implicit implication was
manipulated by altering the proportion overlap of words in the list
learning task and the post-list learning task.
Results
Implication Instructions
Priming Task, No Implication
1
Implied
Not Impled
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
In the next experiment you will see some words on the
screen. Your task will be to give a pleasantness rating for
each of the words. For the pleasantness rating, please
indicate how these words make you feel on a scale of –3 to
+3 (-3 meaning very bad, and +3 meaning very good).
Each word will be presented briefly (2 seconds) followed by
a blank slide. Please pay attention and respond quickly.
Please rate every word.
Original
Pleasantness
Both
Source Attribution
The major source of episodic errors in recall (i.e., the misinformation
effect) seems to be exposure to the items. Explicit implication in
conjunction with high overlap between the context in which
misinformation is presented and the original event both increases source
errors and decreases semantic errors. However, if overlap between
misinformation presentation and the original event is low, explicit
implication has little effect.
Both forms of implication seem to have made participants more aware
of source during the post-event task. Although this improved source
monitoring accuracy, participants were still more likely to incorrectly claim
misleading items appeared in both sources rather than the post-event
alone.
Figure 3. Source Attribution as a Function of Overlap
References
Priming Task, Relational Implication
These instructions were identical to the no implication with
the following addition:
Also, note that some of the words that you will rate were
also on the category lists that you originally studied. For
those words, rate their pleasantness in relation to the other
items on the studied lists. Rate the item a +3 if it is as
pleasant as the most pleasant item on the list, a –3 if it is
as unpleasant as the most unpleasant item on the list, or a
number between that corresponds to the pleasantness of
the word with which it is most similar in pleasantness.
Mean Proportion Attributed .
The misinformation effect refers to incorrect recall or source attribution of
an item presented after a to-be-remembered event as having been
presented during the to-be-remembered event. The current popularity of
studying the misinformation effect can be traced back to 1974 when Loftus
and Palmer demonstrated the effect of misleading post-event questions on
memory. However, the bulk of research on misleading post-event
information confounds presentation of plausible details (exposure) and
implication that the details were present in the original event.
 Allen and Lindsay (1998) demonstrated that the misinformation effect
could be obtained without any implication that the post-event information had
been part of the original event. That is, memory accuracy is reduced by the
presentation of semantically related, but non-episodically related information.
However, the study did not examine the effect of implication on acceptance
of post-event information.
 A second method used to reduce the role of implication could be called
“warning studies.” Greene et al. (1982) found that warnings about
misinformation prior to its presentation reduced later acceptance of that
misinformation, whereas post-misinformation presentation warnings did not
help. Later warning studies have produced mixed results concerning the
efficacy of post-misinformation warnings (e.g., Wright, 1993; Chambers &
Zaragoza, 2001).
In misinformation studies implication may take two general forms. First,
there may be explicitly stated implication (e.g., this narrative describes the
event you just saw in the slide show). The second form is implicit
implication. That is, the rememberer may infer that post-event information is
an accurate recounting of the original event because there is significant
overlap of detail between the two.
Mean Propotion Attributed.
Introduction
David R. Gerkens, Joelle M. Kline, Amanda R. Cross
California State University, Fullerton
Jessica Wood
Mississippi State University
1
0.9
0.8
0.7
0.6
0.5
High Overlap
Low Overlap
0.4
0.3
0.2
0.1
0
Original
Pleasantness
Source Attribution
Both
•Allen, B.P. & Lindsay, D. S. (1998). Amalgamations of memories: Intrusion of
information from one event into reports of another, Applied Cognitive Psychology,
12, 277-285.
•Chambers, K. L., & Zaragoza, M. S. (2001). Intended and unintended effects of
explicit warnings on eyewitness suggestibility: Evidence from source identification
tests. Memory & Cognition, 29, 1120-1129.
•Greene, E., Flynn, M. B., & Loftus, E. F. (1982). Inducing resistance to misleading
information. Journal of Learning & Verbal Behavior, 21, 207-219.
•Loftus, E. F. & Palmer (1974). Reconstruction of an automobile destruction: An
example of the interaction between language and memory, Journal of Verbal
Learning and Verbal Behavior, 13, 585-589.
•Wright, D. B. (1993). Misinformation and warnings in eyewitness testimony: A
new testing procedure to differentiate explanations. Memory, 1, 153-166.