Anderson, S. J., Cohen, G., & Taylor, S. (2000). Rewriting the Past: Some Factors Affecting the Variability of Personal Memories. Applied Cognitive Psychology, 14, 435-454.

The purpose of this study is to test the assumption that autobiographical memory is the

same memory when re-recalled.

*Witnesses may change their reports

*Repeated questions elicit new information

*Repeated closed questions may cause children to change their original reports.

*Importance

*Consequences

*Emotionality

Are memories Reconstructive or Reproductive copies?

2 X 2 within subjects design.

Independent Variables:

Participants: younger (N=28; M=28) or older adults (N=28; M=72)

Memory: high and low frequency memories

Participants wrote a short narrative description of two specific memories, a memory

frequently recalled and a memory that was recalled infrequently. After a 2-month

interval, participants were asked to reproduce these two memories. This study compared

initial version of memories with the re-recalled version.

A total of 67 memories were used in the analysis including:

Older adults: 15 high frequency and 17 low-frequency memories

Younger adults: 18 high frequency and 17 low-frequency memories

Repeated facts were greater for older adults and percentage of repeated facts was

greater for older adults.

Conclusion: Stability was related to age of participant and age of memory. Older adults

produced more stable memories across recall sessions and older memories were more stable than newer memories.

Across time, the general mood of memories remained stable and there were no

inconsistent or contradictory memory accounts. However, in regard to details, only 46%

remained stable in both memory versions for younger adults and only 58% remained

stable in both memory versions for older adults.

Repeated memories often included more detail which suggests that participants in some

way felt that their original accounts of events was unsatisfactory. Experiment 2 attempts

to control for this variable.

Independent Variables:

*Age of Participants: younger adults (N=15, M=37) and older adults (N=15;

M=72)

*Age of Memories: 1 year ago, 5 years ago, or 10 years ago.

*Initial vs. Delayed (8 weeks) recall.

Participants were asked to recall a specific personal memory and list details of that

memory from a designated lifetime period in response to a cue word given to them by the

experimenter. Six cues were assigned to each of the lifetime periods, for a total of 18

memories from each participant. Each participant provided a title for each memory. After

recall, participants rated personal importance and frequency of rehearsal for each

memory. After an eight-week time period, participants were unexpectedly asked to recall

details from each of their original 18 memories. In recall 2, participants were provided

with each memory title and the original cue word.

The data analyzed the memories of 13 younger and 14 older adults.

examination, it was found that the quality of the details was different. The

memories of the older adults contained fewer specific details than the memories

of the younger adults. It appears that older adults begin by recalling specific

details then resort to more theme-based information, while younger adults only

provide specific details

*More details were supplied in Recall Session 1

*The older group produced a higher percentage of repeated memory units(64%)

than the younger group (44%)

*The Memory Age Factor approached significance. Older memories tended to

lose details

*No significant effects were observed for personal importance, emotion, and

frequency of rehearsals.

Conclusion: Older people's memories are more stable across repeated recalls and older

memories are more stable than relatively recent memories.

The purpose for Experiment 3 waste extend and replicate the findings from Experiment 2 with a focus on the age differences in retrieval by examining the timing of the output of memory details and monitor memory variability over a larger number of repeated recalls.

Independent Variables:

Participants: younger adults (N=15,M=36.4) and older adults (N=15,M=69.6)

Recall Sessions: Initial, After 4 weeks, After 8 weeks.

Dependent Variables:

Details listed within a designated time period

Percentage of overall repeated details

Participants were told that the purpose of the experiment was to test the speed with which memory details were reproduced. Participants were asked to recall a specific memory in response to 16 cue words. After 15 seconds, the experimenter noted which detail the participant was working on and after 30 seconds the participant was instructed to finish the detail and stop. At the completion of all 16 memories, the participant was asked to rate the memory for personal importance, emotion, and rehearsal frequency. In Recall Session 1, participants provided a title for each memory. In Recall 2 and Recall 3, participants were supplied with the 16 titles and asked to reproduce as many details as possible in 30 seconds using the same procedure as before.

*Older adults produced a higher percentage of repeated details than the younger adults

did.

*For older adults, 68% of the total memory output was recorded in the first 15 seconds of

retrieval, and for the younger adults, 40% of their total memory output was recorded in

the first 15 seconds of retrieval. This finding may be interpreted as suggesting that

younger people's memory retrieval process is more dynamic and requires memories to be

reconstructed. The initial production is slow because the memory is in a process of

assembly. Once the memory is assembled, the production rate becomes much faster.

*Older participants produced fewer overall memories than the younger participants did.

This result differs from Experiments 1 and 2. This may be interpreted as suggesting that

because recall was timed. Older participants did not have the time to add general, them-

based details as was found in Experiments 1 and 2.

*Older participants produced a higher percentage of repeated details. This suggests that

their memory content is more stable and less likely to change. If their memories are, in

fact, reproduced rather than reconstructed, output order should be more stable. Out put

order was assessed by examining the position where details appeared in re-recalls.

*The order for details was more stable for older people than the output order for younger

people.

Conclusion: The memory of older people for common life events is much more stable than the memory of younger people. There are age-related differences in the structure of personal memory and in retrieval processes. Older people produce initial details rapidly but then slow down. The order of their details is more consistent. Younger people initially produce memories slower and the order is less consistent.

The results of the present study suggest that personal memories are variable. When people re-call the same memory, the same details may not always be recalled.

*Older memories are more stable than newer memories.

*Emotionality, Importance, and Rehearsal Frequency showed no effect on stability in the current study.

*Older people recall memories in a more stable way than younger people do.

This may be due to a higher level of schematization in older people (their

memories were more generic). This may also result from a less evolving

self-concept of older people.

*The retrieval mechanism for older people may differ from the mechanism used by younger people.

Understanding the variability of personal memories is important because of applied under-

standing in the areas of witness testimony, medical history taking, and circumstances where people are asked to supply information at interview.