The Method of Loci

Written by Christy Maver, VP of Marketing

In our recent manuscript, titled Why Does the Neocortex Have Layers and Columns, A Theory of Learning the 3D Structure of the World, we introduced a new theory of how the brain makes predictive models of objects. We proposed that the brain pairs sensory data with location data to perform object recognition. After reading our manuscript, a friend of the company, Hans-Poul Veldhuyzen van Zanten, emailed us with the observation that our theory was compatible with an ancient memorization system known as the “method of loci.”

The method of loci is a memory enhancement technique that combines visualization with spatial memory of familiar environments in order to quickly recall information. It works like this:

Imagine you have a list of items you want to memorize. The list could be anything: numbers, words, faces, etc. To use the method of loci, you would think of a familiar space — something where you can easily draw up the layout of the space in your mind. Maybe it’s your house, or your favorite street of shops. For this example, we’ll use a house. To commit your list of items to memory, you imagine walking through your house and stopping at various locations. At each stop, you mentally assign an item on the list to a location by forming an image of the item and some feature of that location. Then when you want to retrieve this list, you would imagine walking through your house, and as you move from room to room, you recall the items on the list.

This technique dates back to ancient Greek and Roman times, but many memory champions still use this method today in contests that require recalling long lists of unrelated items. While it may sound like something best reserved for those who want to impress with a cool memory “trick,” there’s an underlying concept here that applies to everyone.

Yes, even for those of us who are not memory champs, we are still able to remember and recall enormous amounts of information, seemingly without effort. If I asked you to think about your house right now, you could imagine it instantly. If I asked you to describe your kitchen, you could tell me how it’s laid out, where you store the cups, which drawer contains the silverware, where the refrigerator is located. You could describe to me, in great detail, this environment that you know very well. But what you may not realize is that your knowledge of this space is associated with a location, and it’s the association with a location that allows you to remember and recall large amount of information more easily. When I ask you to picture your refrigerator, you do not recall an isolated image of your refrigerator, as if it came up in an online product search. No, you recall it as it exists, where it exists, in your kitchen.

If you’ve read our latest manuscript, you’ll recall that location was the “missing ingredient” that explained how the brain integrates movement with sensory input. The brain always pairs sensory data with location data. This is how we learn and recognize objects. It’s how we build a model of the world. It’s why we can learn and remember vast amounts of information very quickly. And when we explore what this means, we reach a surprising yet fascinating conclusion: The vast majority of knowledge we have about objects in the world only exists in the context of a location.


Originally published at numenta.com on August 29, 2017.