Rule #5 in John Medina’s Brain Rules is, “Repeat to remember.” He’s talking about short-term memory and the ways we work with information. Working backward from what people have learned, we know there are at least two types of memory.

Bertie Wooster, the genial imbecile of the P. G. Wodehouse stories, demonstrates declarative memory — facts we are consciously aware of. (His man Jeeves demonstrates a higher level of skill with regard to declarative memory.)

“Newts, Jeeves. Mr. Fink-Nottle has a strong newt complex. You must have heard of newts. Those little sort of lizard things that charge about in ponds.”

“Oh, yes, sir. The aquatic members of the family Salamandridae which constitute the genus Molge.”

Nondeclarative memory, logically enough, refers to what we’ve learned that we are not consciously aware of. When Bertie is tooling about in his motorcar, he’s usually unaware of how he knows when to steer, when to shift, and so on. You may know how to ride a bike, but you can’t consciously recall the details, because that’s not in declarative memory.

Medina discusses for processes involved in memory: encoding, storage, retrieval, and forgetting. One point he makes is that our memories are not stored like tape recordings. You can’t just press play and stream them out again.

He cites one striking example. A stroke victim lost the ability to use written vowels. If you asked her to write, “Your dog chased the cat,” this is what she’d write:

Y r d g ch s d th c t.

She not only got all the consonants right, she left room for the vowels she couldn’t write. Consonants seemed to be stored in a different part of the brain from the one affected by her stroke, which did affect the region dealing with written vowels.

This phenomenon illustrates the binding problem — how does the brain connects the widely scattered elements of memory? The problem is even more complex when you consider we encode it in different ways. For example:

• Semantic encoding, or the meaning of what we’ve learned.
• Phonemic encoding, or the sound of what we’ve learned.
• Structural encoding, or the shape or arrangement of what we’ve learned.

There’s also automatic encoding, which all of us have experienced. We seem able at times to build rich memories that we can easily retrieve in great detail with almost no effort.

We’ve all experienced the challenge of deliberate encoding when we want to remember something but can’t drag it out of storage. What IS the password for this website? What is my wife’s Social Security number (which I have looked up at least 50 times)? What are some implications for learning?

Elaborate encoding means stronger memory.

Paradoxically, we can learn better when presented with great detail. It does seem that the detail needs to be relevant rather than distracting, though. Imagine two groups of people given a list of words to study. The first group is asked to focus on words containing the letters I or E. The second group is asked to rate each word and indicate whether they like it on a scale of one to 10. When asked to recall the words later, the “like” group recalls two to three times as many words.

One obvious aid to learning: real-world examples. Rather than focusing solely on principles or theory, we’d learn better with specific examples that provide connections or relationships to what’s already in our memory. It’s one thing to say, “Use striking visuals.” It’s another to see three examples in context.

Retrieve the way you stored.

Research suggests that memory works best if the conditions at retrieval mimic those at the time of storage (initial learning) .

One way to capitalize on this principle is to close the gap between “training” and on-the-job performance— the dilemma known as transfer. Making training or learning part of the job, rather than and auxiliary activity, something you do when you’re not working, helps ensure that the conditions for storage replicate the conditions for retrieval.

Tape recorder photo by AlphaDelta / Peter.