Autobiographical memories

Posted by Eva Bonning on 17 December 2023

I would like to express my gratitude to the HOPE Foundation for supporting my research.

This research aims to investigate how older adults share memories of their personally experienced life events (their autobiographical memories). More specifically, recent evidence has suggested that people share their autobiographical memories using similar emotional trajectories as found in Western fiction. That is, both fictional stories—and people’s memory reports—demonstrate changes in positive and negative emotion at specific points in the story, as it unfolds from beginning to end. However, so far, we only know how these trajectories behave in younger adults, and we know that older adults generally describe their memories differently to younger adults. For example, older adults can demonstrate a change in narrative style when describing their autobiographical memories that focusses less on specific details, and more on imparting meaning. Or, compared with younger adults, older adults often remember events more positively, and report greater wellbeing. Therefore, is important to determine the extent to which emotional trajectories may replicate—or look different—in older adults. 

In the first year of my PhD, I learned to use natural language processing software to analyse the content of people’s reports for their most positive, negative and important memories. As I am nearing the end of my second year, my goal has been to further validate these software methods with two major projects (explained below), so that in my third year, I can begin to examine the use of emotional trajectories in older adults. For the first project, we are launching an online experiment where subjects will learn about emotional trajectories in Western fiction, then classify whether others' memory descriptions conform to one of those trajectories. For the second project, we have hired two research assistants (using pre-existing funding) to hand score the valence of individual sentences in the same subset of memory descriptions as for the first project. This process also involves validating a coding scheme for possible future use in other experiments. As we come to the end of 2023, we are close to completing these two major projects. Their findings will directly inform our method for investigating emotional trajectories in older adults’ retellings of their autobiographical memories.  

Therefore, my plan for next year is to complete the first study I discussed in my research proposal for the HOPE Foundation. More specifically, we will ask approximately 100 older and younger adults to describe their most positive, negative and important memories, and determine which trajectories are most common for each group of people. We will use our findings from this year’s two main projects to guide our analyses on this experiment. We will also use a subset of the collected data from this project to further validate the natural language processing software for older adults. We will do this validation in a similar way to the second project discussed above. That is—we will hire research assistants to hand score the valence of individual sentences in a subset of memory descriptions. We will then compare this hand scored subset, to those scored by software.  

In terms of conference attendances, in August 2023 I was able to travel to Japan and attend a biannual memory and cognition conference—SARMAC. This experience was a great opportunity to learn from others, network, and come in close contact with experts in the field. 

Once again, thank you to the HOPE Foundation’s continued support.  

You can also help by

  • Spreading the word about what we do / share this newsletter
  • Donating your time to the Friends Supporters to help with fundraising and committee work
  • Encouraging your children and grandchildren to invest in their futures by donating time and money (a baby girl born today has a 1 in 3 chance of living to 100 , a boy 1 in 4 and is likely to be fitter and healthier–think about the implications of that)
  • Consider a bequest