Embracing New Techniques in Deep Learning for Estimating Image Memorability

Various work has suggested that the memorability of an image is consistent across people, and thus can be treated as an intrinsic property of an image. Using computer vision models, we can make specific predictions about what people will remember or forget. While older work has used now-outdated deep learning architectures to predict image memorability, innovations in the field have given us new techniques to apply to this problem. Here, we propose and evaluate five alternative deep learning models which exploit developments in the field from the last five years, largely the introduction of residual neural networks, which are intended to allow the model to use semantic information in the memorability estimation process. These new models were tested against the prior state of the art with a combined dataset built to optimize both within-category and across-category predictions. Our findings suggest that the key prior memorability network had overstated its generalizability and was overfit on its training set. Our new models outperform this prior model, leading us to conclude that Residual Networks outperform simpler convolutional neural networks in memorability regression. We make our new state-of-the-art model readily available to the research community, allowing memory researchers to make predictions about memorability on a wider range of images.

Shared memories driven by the intrinsic memorability of items

When we experience an event, it feels like our previous experiences, our interpretations of that event (e.g., aesthetics, emotions), and our current state will determine how we will remember it. However, recent work has revealed a strong sway of the visual world itself in influencing what we remember and forget. Certain items -- including certain faces, words, images, and movements -- are intrinsically memorable or forgettable across observers, regardless of individual differences. Further, neuroimaging research has revealed that the brain is sensitive to memorability both rapidly and automatically during late perception. These strong consistencies in memory across people may reflect the broad organizational principles of our sensory environment, and may reveal how the brain prioritizes information before encoding items into memory. In this chapter, I will discuss our current state-of-the-art understanding of memorability for visual information, and what these findings imply about how we perceive and remember visual events.