> "To this day," Bjork says, "most people think about forgetting as decay, that memories are like footprints in the sand that gradually fade away. But that has been disproved by a lot of research. The memory appears to be gone because you can't recall it, but we can prove that it's still there. For instance, you can still recognize a 'forgotten' item in a group. Yes, without continued use, things become inaccessible. But they are not gone."
This was quite eye opening to me when I first heard of it. Apparently, our long term memory is virtually limitless. You never truly lose something you once had learned well.
Another fact about memory that I found equally interesting: memory is not a "thing" but a process. When you recall something, you are actually reconstructing the memory from various factors associated with it, pretty much like how a paleontologist might reconstruct the image of a dinosaur from fossil remains.
It seems to me that there are two gates, and they move with age, and the rate at which they move differs from person to person. The first gate is the time until losing something from associative memory. The second gate is losing the memory completely. Yes, this does happen, and it happens to everyone. But the quote still has some validity in that many things we think we forgot are still buried deep inside.
> This was quite eye opening to me when I first heard of it. Apparently, our long term memory is virtually limitless. You never truly lose something you once had learned well.
There are startling experiments in the area. Take a large set of pictures (it has been done with sets of up to 20000 at least). Take half those pictures and show them to subjects, one at a time, giving them a few seconds to look at each picture.
Later, take the set of pictures you showed them, and pair each with one of the pictures you did not show them. Shuffle that set of picture pairs, and then show the subject the pairs sequentially, asking them for each which picture of the pair they have seen before.
On a test that used 20000 pictures, with the recognition test a day and half later, subjects got 80% right. If instead people had just been asked to recall as many pictures as they could, most people only be able to recall a tiny fraction of them.
> Another fact about memory that I found equally interesting: memory is not a "thing" but a process. When you recall something, you are actually reconstructing the memory from various factors associated with it, pretty much like how a paleontologist might reconstruct the image of a dinosaur from fossil remains
This is sort of illustrated by a thing called the Baker/baker paradox. Show people a photograph of someone, with a label that says "BAKER" attached. Tell half the subjects that the label means that the person is a baker, and tell the other half that the label means his name is Baker.
Later, show the subjects the photo but with the label removed, and ask them what the missing label had said. The people given the job story will be much more successful at remembering this than the people given the name story.
The name "Baker" has very little associated with it in most people's minds, whereas the job "baker" brings up associations with all kinds of delicious treats, and maybe childhood memories of wonderful smells permeating the house as your mother baked in the kitchen.
So when you are told the photo is of a baker, all of those things that baking is associated with in your mind get associated with the photo, and you've got more to grab onto mentally when trying to recall the label. These may be weak associations, but they still beat out the name "baker" which for most people has no extra associations.
Anyway, we don't just reconstruct the memory to recall it. We also might edit it and save the changes. This has serious implications for the criminal justice system when dealing with human witnesses.
If an investigator interviewing the witness says anything that suggests events happened different than the way the witness remembers them, there is a danger the witness' memory will change to match the interviewer's version. The witness will not know this is happening, and genuinely believe that the updated memory is what really happened. These aren't always minor updates...they can be big, like in a confrontation changing who drew a gun and started shooting.
I don't have a link, but I've read of experiments where researchers got parents or older siblings of some adult subjects to mention at a family gathering some amusing/embarrassing incident from the subject's childhood--a totally made up incident. Later, the researchers found excuses to question the subjects about that incident, and the subjects "remembered" it, and not only the details that had been supplied in the made up story. They had filled in complete details about the incident and the things leading up to it and following it.
This was quite eye opening to me when I first heard of it. Apparently, our long term memory is virtually limitless. You never truly lose something you once had learned well.
Another fact about memory that I found equally interesting: memory is not a "thing" but a process. When you recall something, you are actually reconstructing the memory from various factors associated with it, pretty much like how a paleontologist might reconstruct the image of a dinosaur from fossil remains.