Research

In this line of work, we investigate the inputs into memory – what is it about certain images that makes us remember or forget them? In fact, we are highly consistent in the images we remember: in spite of our individual differences, people tend to remember and forget the same images (Bainbridge, 2019. Psychology of Learning and Memory.). Images can be thought of as having an intrinsic and quantifiable memorability, and our brains are sensitive to this measure during late visual processing (Bainbridge et al., 2017. NeuroImage). Memorability operates at the level of dynamic face identities beyond just images (Bainbridge, 2017. Journal of Experimental Psychology: Learning, Memory, and Cognition), has meaningful implications for understanding memory disorders (Bainbridge et al., 2019. Alzheimers & Dementia: Diagnosis, Assessment, & Disease Monitoring), and is separate from phenomena of bottom-up attention, top-down attention, and priming (Bainbridge, in review).

In our current work, we are investigating this concept of memorability from both computational and neuroscientific angles: How is memorability calculated in the brain? Can we create computational models that accurately quantify and manipulate image memorability? How do memorability effects operate for more dynamic, complex stimuli, and how robust are they in the real world?

Here, we investigate the outputs of memory – how can we understand the visual representations of a memory? We ask participants to make drawings of real-world scenes, and then through large-scale online crowdsourced scoring, we quantify the amount of detail present within these memories (Bainbridge et al., 2019. Nature Communications). Beyond what might have been expected from verbal memory studies, we find people on average draw hundreds of details from memory, in surprisingly accurate locations. We can then use this method to probe the nature of perception and memory in populations with unique cognitive experiences, such as aphantasics, individuals with impaired visual imagery (Bainbridge et al., 2021. Cortex). From these drawing studies, we have also been able to revisit well-accepted memory phenomena, such as boundary extension, and discovered that the opposite effect is also very common (Bainbridge and Baker, 2020. Current Biology).

In our current work, we are expanding these metrics of visual recall to see how memory representations evolve over the lifetime of a memory and an individual.