Essentials: Understand & Improve Memory Using Science-Based Tools

The episode opens by framing memory in terms of perception. People are constantly exposed to touch, light, smells, tastes and sound waves, and the nervous system (the brain, the spinal cord and their two-way connections with the body's organs) converts all of these stimuli into electrical and chemical signals. Even so, people perceive only a small subset of them. Listeners can hear the voice speaking, for example, while most likely ignoring the feel of the surface they are sitting or standing on.

This selectivity is presented as necessary: without it, people would be overwhelmed by everything happening in a given moment. On that basis the episode offers a compact definition of memory:

Now, memory is simply a bias in which perceptions will be replayed again in the future.

That definition raises the episode's central question. If people experience things all the time and perceive only some of them, why do only some perceptions get "stamped down" as memories? The stated aim is to explain how that happens and how to use the process to remember chosen information better.

The episode also notes that every memory is linked to something else by a close, medium or very distant association. Memory programs and "phenoms" who can recall 50 names in a room often depend on association tricks that link a word's sound or meaning to something personally meaningful. The episode calls this impressive but says that most people can't do it, or can do it only with a lot of effort and training. It then turns to tools that work with the nervous system's natural biology instead.

The first tool is described as one that clearly works: repetition. The more often something is performed or recited, the more likely it is to be remembered. At the neural level, repetition fires particular chains of neurons in a circuit (neuron A, B, C, D in a set sequence) again and again, and more repetitions strengthen those connections further.

The limitation is practical. People may lack the patience or the time, sometimes because of a deadline, or they may simply want to remember things better and faster in general. The episode describes a typical learning curve in which a skill first takes a thousand repetitions, then gradually fewer, until it can be done correctly "the first time and every time." It claims this curve can be shifted so that the connections behind a memory or behavior strengthen more quickly.

The answer draws on several decades of experiments by James McGaugh and Larry Cahill. Their work looked at whether stress, and neurochemicals linked to stress, improve the ability to learn information of all kinds, not just emotional information.

The animal evidence starts with conditioned place aversion. A rat or mouse receives an electrical shock at one spot in an arena. The next day, with the shock device removed, it avoids that spot. This happens after a single trial, which the episode cites as an example of one-trial learning and describes as hippocampal-dependent.