NeverMind is an interface and application designed to support human memory. We combine the memory palace memorization method with augmented reality technology to create a tool to help anyone memorize more effectively. Early experiments conducted with a prototype of NeverMind suggest that the long-term memory recall accuracy of sequences of items is nearly tripled compared to paper-based memorization tasks. With this project, we hope to make the memory palace method accessible to novices and demonstrate one way augmented reality can support learning. Read more here. NeverMind won the Best Poster Award at ACM’s USIT in Tokyo, Japan.
Hit play to watch how NeverMind works.
To memorize with NeverMind, the user takes a walk through a building they are familiar with. With this method, the architectural spaces are the equivalent to the imaginary rooms of the original memory palace. At each room, the user queries the app to display an image. This allows for a specific image to be mentally bound to an architectural location, that way during recall, each location will trigger a specific memory.
Then, in order to memorize a sequence of items, the user physically walks to the next room and queries the interface for the next image to be displayed. Because augmented reality allows for the user to be aware of the spaces they are in, association and embedding of an image to a specific location becomes experiential instead of imaginary.
With NeverMind, the user only needs to see images once in order to memorize effectively. Experiments show that three seconds of exposure is enough for content to remain in memory. Because the user is only seeing the images once, I will refer to this process as one-shot memorization, inspired by Josh Tenenbaum’s concept of one-shot learning.
The contributions of this project are twofold: first, I developed NeverMind, a tool to facilitate memorization through a single exposure by biasing our minds into using episodic memory. When studying, we tend to use semantic memory and encoding through repetition; however, by using augmented reality interfaces we can manipulate how our brain encodes information and memorize long term content with a single exposure, making a memory champion technique accessible to anyone. Second, I provide an open-source platform for researchers to conduct high-level experiments on episodic memory and spatial navigation. In this thesis I suggest that digital user interfaces can be used as a tool to gather insights on how human memory works.
Project done in collaboration with Marc Exposito.