Idiot Brain (6 page)

Encoding is something the brain can do fairly quickly, but not immediately, hence short-term memory relies on less permanent but more rapid patterns of activity to store information. It doesn't form new synapses; it just triggers a bunch of essentially multipurpose ones. Rehearsing something in short-term memory keeps it “active” long enough to give the long-term memory time to encode it.

But this “rehearsing something until I remember it” method isn't the only way we remember things, and we clearly don't do it for
everything
we can remember. We don't need to. There's strong evidence to suggest that nearly everything we experience is stored in the long-term memory in some form.

All of the information from our senses and the associated emotional and cognitive aspects is relayed to the hippocampus in the temporal lobe. The hippocampus is a highly active brain region that is constantly combining the never-ending streams of sensory information into “individual” memories. According to a great wealth of experimental evidence, the hippocampus is the place that the actual encoding happens. People with a damaged hippocampus can't seem to encode new memories; those who have constantly to learn
and remember new information have surprisingly large hippocampi (like taxi drivers having enlarged hippocampal regions that process spatial memory and navigation, as we'll see later), suggesting greater dependence and activity. Some experiments have even “tagged” newly formed memories (a complex process involving injecting detectable versions of proteins used in neuronal formation) and found that they are concentrated at the hippocampus.
⁵
This isn't even including all the newer scanning experiments that can be used to investigate hippocampal activity in real time.

New memories are laid down by the hippocampus and slowly move out into the cortex as new memories form “behind” them, gradually nudging them along. This gradual reinforcing and shoring up of encoded memories is known as “consolidation.” So the short-term-memory approach of repeating something until it's remembered isn't
essential
for making new long-term memories, but it is often crucial for making sure that
a specific arrangement of information
is encoded.

Say it's a phone number. This is just a sequence of numbers that are already in the long-term memory. Why would it need to encode them again? By repeating the phone number, it flags up that this particular
sequence
of numbers is important and requires a dedicated memory to be retained long term. The repetition is the short-term memory equivalent of taking a bit of information, sticking on a label marked “
Urgent!
” then sending it to the filing team.

So, if the long-term memory remembers everything, how do we still end up forgetting things? Good question.

The general consensus is that forgotten long-term memories are still technically there in the brain, barring some
trauma in which they're physically destroyed (at which point being unable to remember a friend's birthday will not seem so important). But long-term memories have to go through three stages in order to be useful: they need to be made (encoded); they need to be effectively stored (in the hippocampus and then the cortex); and they need to be retrieved. If you can't retrieve a memory, it's as good as not being there at all. It's like when you can't find your gloves; you still
have
gloves, they still exist, but you have cold hands anyway.

Some memories are easily retrieved because they are more salient (more prominent, relevant, intense). For example, memories for something with a great degree of emotional attachment, such as your wedding day or first kiss or that time you got two bags of chips out of the vending machine when you only paid for one, are usually very easily recalled. As well as the event itself, there are also all the emotions and thoughts and sensations going on at the same time. All of these create more and more links in the brain to this specific memory, which means the aforementioned consolidation process attaches a lot more importance to it and adds more links to it, making it much easier to retrieve. In contrast, memories with minimal or no important associations (for instance, the 473rd uneventful commute to work) get the bare minimum of consolidation, so they're a lot harder to retrieve.

The brain even uses this as something of a survival strategy—albeit a distressing one. Victims of traumatic events often end up suffering from “flashbulb” memories, where the memory of the car accident or gruesome crime is vivid and it keeps recurring long after the event (see
Chapter 8
). The sensations at the time of the trauma were so intense, with the brain and body flooded with adrenalin causing heightened senses and
awareness, that the memory lodges powerfully and remains raw and visceral. It's as if the brain took stock of the awful things happening and said, “This right here, this is awful; do
not
forget this, we
do not
want to have to go through this again.” The trouble is, the memory can be so vivid it becomes disruptive.

But no memory is formed in isolation, so even in more mundane scenarios the context in which the memory was acquired can also be used as a “trigger” to help retrieve it, as some bizarre studies have revealed.

In one example, scientists got two groups of subjects to learn some information. One group learned it in a standard room; the other group learned it while underwater, wearing full scuba suits.
⁶
They were later tested on the information they were told to learn, either in the same situation or the alternative one. Those who studied and were tested in the same situation performed significantly better than those who studied and were tested in different ones. Those who studied underwater and did the test underwater got much better scores than those who studied underwater but did the test in a normal room.

Being underwater had nothing to do with the information being learned, but it was the
context
in which the information was learned, and this is a big help in accessing memory. Much of the memory for where information is learned involves the context at the time, so putting someone in the same context essentially “activates” part of the memory, making it easier to retrieve it, like revealing several letters in a game of hangman.

At this point, it's important to point out that memories for things that happen to us are not the only types of memories. These are called episodic memories, or “autobiographical”
memories, which should be self-explanatory. But we also have “semantic” memories, which are for information essentially without the context: you remember light travels faster than sound, but not the specific physics lesson where you learned this. Remembering that the capital of France is Paris is a semantic memory, remembering the time you vomited off the Eiffel Tower is an episodic memory.

And these are the long-term memories we're consciously aware of. There's a whole swathe of long-term memories that we
don't need to be aware of
like abilities we have without thinking about it, such as driving a car or riding a bike. These things are called “procedural” memories, and we won't go into them any further because you'll start thinking about them, and that might make it harder to use them.

In summary, short-term memory is fast, manipulative and fleeting, whereas long-term memory is persistent, enduring and capacious. This is why a funny thing that happened while in school can be something you remember forever, and yet still decide to go into a room but, if distracted even slightly, forget why by the time you get there.

Hey, it's . . . you! From . . . the thing . . . that time

(The mechanisms of why we remember faces before names)

“You know that girl you went to school with?”

“Can you narrow it down?”

“You know, the tall girl. Dark blond hair but I think she was dyeing it, between you and me. She used to live in the street
next to us before her parents divorced and her mother moved into the apartment that the Jones family lived in before they moved to Australia. Her sister was friends with your cousin before she got pregnant with that boy from town—that was a bit of a scandal. Always wore a red coat but it didn't really suit her. You know who I mean?”

“What's her name?”

“No idea.”

I've had countless conversations like this, with my mother, gran or other family members. Clearly, there's nothing wrong with their memory or grasp of detail; they can provide personal data about someone that would put a Wikipedia page to shame. But so many people say they struggle with names, even when they're looking directly at the person whose name they're trying to recall. I've done this myself. It makes for a very awkward wedding ceremony.

Why does this happen? Why can we recognize someone's face but not their name? Surely both are equally valid ways of identifying someone? We need to delve a bit deeper into how human memory works to grasp what's really going on.

Firstly, faces are very informative. Expressions, eye contact, mouth movements, these are all fundamental ways humans communicate.
⁷
Facial features also reveal a lot about a person: eye color, hair color, bone structure, teeth arrangement; all things that can be used to recognize a person. So much so that the human brain has seemingly evolved several features to aid and enhance facial recognition and processing, such as pattern recognition and a general predisposition to pick out faces in random images, as we'll see in
Chapter 5
.

Compared to all this, what does someone's name have to offer? Potentially some clues as to their background or
cultural origin, but in general it's just a couple of words, a sequence of arbitrary syllables, a brief series of noises that you're informed belong to a specific face. But so what?

As we have seen, for a random piece of conscious information to go from short-term memory to long-term memory, it usually has to be repeated and rehearsed. However, you can sometimes skip this step, particularly if the information is attached to something deeply important or stimulating, meaning an episodic memory is formed. If you meet someone and they're the most beautiful person you've ever seen and you fall instantly in love, you'd be whispering the object of your affection's name to yourself for weeks.

This doesn't usually happen when you meet someone (thankfully), so if you wish to learn someone's name, the only guaranteed way to remember it is to rehearse it while it's still in your short-term memory. The trouble is, this approach takes time and uses mental resources. And as we saw from the “Why did I just come in here?” example, something you're thinking about can be easily overwritten or replaced by the next thing you encounter and have to process. When you first meet someone, it's extremely rare for them to tell you their name and nothing else. You're invariably going to be involved in a conversation about where you're from, what you do for work, hobbies, what they arrested you for, that sort of thing. Social etiquette insists we exchange pleasantries on first meeting (even if we're not really interested), but every pleasantry we engage in with a person increases the odds of the person's name being pushed out of short-term memory before we can encode it.

Most people know dozens of names and don't find it takes considerable effort each time you need to learn a new one.
This is because your memory associates the name you hear with the person you're interacting with, so a connection is formed in your brain between person and name. As you extend your interaction, more and more connections with the person and their name are formed, so conscious rehearsing isn't needed; it happens at a more subconscious level due to your prolonged experience of engaging with the person.

The brain has many strategies for making the most of short-term memory, and one of these is that if you are provided with a lot of details in one go, the brain's memory systems tend to emphasize the first thing you hear and the last thing you hear (known as the “primacy effect” and “recency effect,” respectively),
⁸
so a person's name will probably get more weight in general introductions if it's the first thing you hear (and it usually is).

There's more. One difference between short- and long-term memory not discussed so far is that they both have different overall preferences for the
type
of information they process. Short-term memory is largely
aural,
focusing on processing information in the form of words and specific sounds. This is why you have an internal monologue, and think using sentences and language, rather than a series of images like a film. Someone's name is an example of aural information; you hear the words, and think of it in terms of the sounds that form them.

In contrast to this, the long-term memory also relies heavily on vision and semantic qualities (the
meaning
of words, rather than the sounds that form them).
⁹
So a rich visual stimulus, like, say, someone's face, is more likely to be remembered long term than some random aural stimulus, like an unfamiliar name.

In a purely objective sense, a person's face and name are,
by and large, unrelated. You might hear people say, “You look like a Martin” (on learning someone's name is Martin), but in truth it's borderline impossible to predict accurately a name just by looking at a face—unless that name is tattooed on his or her forehead (a striking visual feature that is very hard to forget).

Let's say that both someone's name and face have been successfully stored in the long-term memory. Great, well done. But that's only half the battle; now you need to access this information when needed. And that, unfortunately, can prove difficult.

The brain is a terrifyingly complex tangle of connections and links, like a ball of Christmas-tree lights the size of the known universe. Long-term memories are made up of these connections, these synapses. A single neuron can have tens of thousands of synapses with other neurons, and the brain has many billions of neurons, but these synapses mean there is a link between a specific memory and the more “executive” areas (the bits that do all the rationalization and decision-making) such as the frontal cortex that requires the information in the memory. These links are what allows the thinking parts of your brain to “get at” memories, so to speak.