Is memory redundant?
After all, the world is just a Google search away. Who needs to memorize facts, dates, terms and definitions when you can pull them up in less time than it takes to wrack your brain to remember them?
Some argue that just as teaching cursive or long division is becoming antiquated, so to is our scholastic obsession with memory. Why should we encourage people to memorize things if they’re so easy to look up?
Your Mind Outside Your Brain
An interesting idea from cognitive science is that of distributed cognition . Basically this suggests that your mind, as a system that processes information, may extend beyond the reaches of your skull.
This sounds crazy, but it’s actually rather simple.
Consider a calculator. When you want to multiply two long numbers, you could try to do this in your head using your internal cognitive tools for manipulating symbols and following the procedure you mastered in grade school.
Alternatively, you can pick up a calculator, put in the numbers and get the same answers. Although this calculator isn’t part of your brain, it could be seen as an extension of your mind—an externalized cognitive procedure for calculating things.
Similarly, notes function like externalized memories. In ages before paper, great orators mastered complex mnemonics to memorize great speeches verbatim. Now we just write them down, the paper substituting for our built-in memories.
Given this concept of externalized cognition, it’s worthwhile to ask why we bother trying to remember so much in the first place? If we can write it down or look it up, doesn’t that effectively mean the same thing as if you remember it personally?
Spreading Activation, Modularization and Why Brains Still Matter
The problem with this line of reasoning is that while a calculator does function as a piece of external cognition, it’s much more separated from our brain than other brain regions are from each other. In computer science parlance, we can say that the calculator and your brain are two very separate modules that communicate via a much more limited interface (the buttons on the calculator and the calculator output stream).
While the brain is likely modularized to some extent, it has a lot more interconnections than this, and therefore each module likely has much richer representation and feedback to other modules than you do with your calculator.
In terms of memory, this can have big implications.
A central idea in how brains store memories is that of spreading activation . This occurs when one idea suggests another, which suggests another, which suggests another. This spreading out (which may not be entirely conscious) enables us to locate memories.
However, if your memories are located behind a Google search, they can’t participate in this network effect.
The problem, therefore, isn’t that using a Google search to look up a fact is inherently wrong, but that very often you won’t even think to look something up. Not being in your mind, the correct question to search for may not even form, so there’s nothing for Google to reply with.
What Does it Mean to Understand?
There’s less consensus as to what it means to “understand” something, in strictly neurological terms. Part of this is because what we mean by understanding itself is somewhat ambiguous. As the problem of explanatory depth  shows, we often mean very different things by “understanding” depending on the context.
Regardless of whether a specific definition of understanding can be found, we can at least say that to understand something means to have some kind of flexible representation of it in your mind. You should be able to ask many questions and get a good answer about a thing you understand, versus something you simply memorized.
Here, however, gets to the heart of why remembering things in your brain still matters. While Google can substitute for poorly understood, memorized facts (what date was the Magna Carta signed?), it cannot easily substitute for the flexibly represented ideas that have deep understanding (why was the Magna Carta signed?).
As Google searches and easily accessible translations, definitions, calculations and algorithmic capacities increase, we’ll need less memorization. But, what’s left over is more understanding, deep knowledge that requires the spreading activation between many different concepts stored in memory in flexible ways.
That deeper understanding, however, is built off of a lot of memories actually stored in your brain. Ironically, the changes wrought by modern technology may require us to remember more, not less, as shallow recitation of facts gets replaced by deeply remembered ideas.