Transcript Consistency of 9-11 Memories

Consistency of 9-11 Memories
Kristen Sager, Advisor: Dr. Arnold L. Glass
On September 13, 2001 a class of cognition students completed a questionnaire that focused on his/her individual circumstances during the initial hearing of the attacks on the World Trade Center. The students were phoned two times
after this initial report. The first call, more than three years later, asked the students the same questions from the questionnaire in order to gage the consistency of their memories. Reminiscence and recognition were assessed in order to
test whether inconsistent memories were due to retrieval error or a complete loss of the memory. The major finding was that half of the subjects reported inconsistent memories, suggesting flashbulb memories are prone to deterioration.
Most likely, these inconsistencies were due to retrieval error.
What is a Flashbulb Memory?
Participants/Procedure
Example of an Inconsistent Subject
Brown and Kulik argued personal circumstances during an unexpected, emotionalarousing event are encoded and stored in memory different than the memory
itself. This type of event triggers a special neurophysiological mechanism
which is activated for a short duration, which creates a flashbulb memory.
Flashbulb memories are accurate, “live,” rich with visual representations,
consistent over time and retrieved with confidence in its accuracy. The
flashbulb memories are characterized by information upon first hearing of the
emotional-arousing event including: place, ongoing event, informant, own
affect, aftermath, and Idiosyncratic information.
A cognition class of 189 students at Rutgers University was originally chosen to complete this study. Out
of these, 54 were able to complete at least one phone call to provide data regarding flashbulb memory.
Of the 54, 33 completed 2 phone calls, which addressed reminiscence and recall/recognition data.
The initial memory collection was on September 13, 2001, 2 days after the attack on the World Trade
Center. 189 students received a questionnaire which asked the subject to:“Please describe in as much
detail as possible how you first heard of the attack on the World Trade Center. What day and time was
it? Where were you? Did you hear from a friend, television, radio, etc.? What were your thoughts at the
time? How distressed has this made you: very, moderately, mildly, not at all?”
The second and third memory collections were between the time interval of October 2004 and March 2005.
These collections were completed over the phone and asked the same questions as the questionnaire.
To test for reminiscence the subject was called 2 times. The time inbetween the 2 phone calls provided
more opportunity for the subject to retrieve the memory.
To test recognition the subject was given 2 distracter lines in addition to one line pulled directly out of what
they wrote in their original recognition report. The subject was read all 3 lines and was told to pick out
their actual memory.
Original report: “I awoke around 11 and when I left the bedroom a
roommate/friend informed me that we were under attack.”
Results
87
74
80
84
71
50
60
Percent Consistent
40
20
rs
ro
In
for
m
an
t
0
pa
ny
2. Recall and Recognition: In a recognition test the subject is presented with a
copy of the information to be found in their memory, this would aid in retrieval
of the memory. The subjects are given a recognition question in the 2nd phone
call.
100
Co
m
1. Reminiscence: The remembering again of the forgotten without re-learning or
a gradual process of improvement in the capacity to revive past experiences.
After time to think, new generation strategies will lead to different activation
levels of the targets.
Example:
Original report: “After class was over, I found out the classes were canceled for
the rest of the day. I called my husband and he gave me all the details.”
1st phone report: “Then about the time that class ended it was a second period
class, they canceled school and on the way home I was listening to the radio in
the car and that is when I found out.”
2nd phone report: “...3rd period was canceled, they said that everyone can go
home, on my way towards the car I heard people talking about another plane
hitting the WTC, eventually I was able to get in touch with my husband and he
was able to tell me what was going on.”
These results show that this subject did have a consistent representation of her
memory, but needed more time to retrieve it. However, since only 2 subjects
showed this phenomena it is not a significant finding.
Recall/Recognition Results
•The 100% consistency in time recall can be attributed to hearing news reports
of the plane crash repeatedly.
100
Ti
me
In this study 2 memory aspects will test these reasons for inconsistencies:
2 out of the 33 subjects demonstrated reminiscence (6%)
Discussion
Percent Consistent
Reasons For Inconsistencies
There are 2 different reasons for inconsistencies. First, no consistent
representation of the memory is available for retrieval. Second, there is a
consistent representation of the memory, but retrieval failure.
Reminiscence Results
•15 out of 19 (79%)subjects were inconsistent in recall, but accurate in
recognition
Er
- In another study, by Neisser and Harsch(1992) it was found that memories are
likely to remain accurate for up to a year, but diminish after 2 and a half years.
2nd phone report: “…And we were sleeping and his roommate came into the
room and said we were being attacked by terrorists.”
•28 out of 32 (88%)subjects were correct in their recognition
No
- One example of a study was completed by Talarico and Rubin (2003) who tested
subjects 1, 6, and 32 weeks after the September 11th attacks and found that
flashbulb memories were not special in their accuracy, only in their perceived
accuracy.
For each of the 5 information categories the total number of subjects who provided an answer were totaled.
Not all subjects provided an answer for each category of information, since it was free recall. By
separating the categories one can see which category has the most consistent information. The graph
below represents the percentage of subjects consistent in each category along with the total number of
subjects with no errors.
Le
ve
l
- With a new test-retest model accuracy of memories were addressed by comparing
the memory shortly after the event to memory after a long retention interval.
These studies mostly demonstrate that flashbulb memory is NOT immune to
forgetting.
To assess the consistency of the memories between the initial memory report and the memory report 3 years
later, the following 5 different pieces of information were compared:
1- Time
2- Who subject was with (company)
3- How subject heard (informant)
4- Where subject was (place)
5- Distress level
es
s
Subsequent Studies Addressed Accuracy
Analysis
Di
str
Subjects were asked about their memory regarding the assassination of President
John F. Kennedy 13 years after the tragedy. 79 out of 80 of the subjects were
found to have flashbulb memories. The subjects had a flashbulb memory if
they (a). Answered, “yes” to the question, “Do you recall the circumstances in
which you first heard that event?” and (b). Provided information in at least one
of the six canonical information categories in reporting these circumstances.
There was a problem with this study. Brown and Kulik claimed flashbulb
memories are regarded as accurate and immune to forgetting, but the study did
not test the accuracy of the subject’s memories.
Pl
ac
e
The Original Flashbulb Memory Study, Brown and Kulik (1977)
1st phone report: “I woke up to his roommate coming into the room saying we
were being attacked by terrorists.”
•4 out of 32 subjects who completed the recognition test were incorrect
suggesting that their memory was completely gone, hence it was not a retrieval
problem. However, this is not a significant finding.
•15 out of 19 subjects were inconsistent in their recall BUT correctly
recognized their memories. This suggests that their failure to recall consistently
was due to retrieval error. However, in order for these results to be strong, one
distracter must contain the same false information which the subject reported
inconsistently. This study did not do this.
•According to Brown and Kulik, flashbulb memories should NOT change
overtime. However, flashbulb studies, including this one, find that memory
does change overtime. This is most likely the result of retrieval failure,
caused by an excess of post-event information!