All language subtitles for Lecture04 Scientific Secrets for Powerful Memory

Forgetting can be defined as the absence of memory.

Failure to remember something formation, the concept of forgetting,

which makes me think of looking at a picture of myself
fishing with my grandfather when I was three or four years old.

I recognize that it's me in the photo.

I recognize the place where the photo was taken,
but I have no memory of the experience.

I'm confident that the picture is real, that I had
the sensory experiences of someone fishing with their grandfather.

Why don't I have that memory? Where did that information go?

This session is about forgetting.

This course is aimed at teaching you about how human memory works

and about how you can use that information to enhance the accuracy
and detail of your own memory.

It's worth
noting that we could state the same goals from the opposite direction.

Forgetting is in essence a failure of the memory system.

If we understand how to reduce forgetting,
then we will know how to improve memory.

If you ever study for exams or know someone who does,
this session is an important one in terms of tips

about how to commit things to memory
as part of taking a course in school.

This session will have some very directly useful information.

I often talk with students who spent many, many hours studying something
only to get back disappointing scores on an exam.

This session will cover
how those people could have learned much more in much less time

by matching their study schedules
to the characteristics of their human brain.

I'm going to start today
by discussing something called the Ebbinghaus Forgetting function.

Named for an early psychologist named Hermann Ebbinghaus Ebbinghaus.

Like most people notice that we tend to forget things.

He wondered
if this forgetting was governed by any specific principles or laws.

Ebbinghaus did his research

in the late 1800s and early 1900s
when cognitive science just didn't exist as a discipline.

Yet science was being used to study
a lot of different things things like physics and chemistry and biology.

But the idea of applying science to people was still a very new idea.

His primary research subject was himself.

Ebbinghaus generated an enormous library of three letter
nonsense syllables.

He would then randomly pick out some set of these syllables
and attempt to memorize them.

His thinking here was that if he'd used actual words
instead of these nonsense syllables,

then he would have to deal with the interaction
between knowledge and memory by using random letter.

Try Grams. He called them.

He aimed to study memory without that interaction,

like studying memory in isolation, separate from the processes
associated with existing knowledge.

Having watched

the first three sessions of this course,
I hope you know this was actually a bit of a bad experimental choice.

Memory for most real world
information is inherently related to your existing knowledge.

One might even argue that the very way that we remember new things
is by linking those new things with other things that we already know.

It's much harder to get an experimental handle on existing knowledge
to say, experimentally control it

because it varies so much from person to person.

But to throw this out
throws out a whole bunch of important aspects of memory.

Nonetheless, Ebbinghaus discovered two different principles
that do seem to apply to human memory in general,

not just to memory for three letter nonsense syllables,
but to almost everything that's ever been studied.

Ebbinghaus had an enormous effect on the study of memory.

In a way, all of the memory studies that have come since have been
greatly affected by the scientific rigor that applied to this topic.

Ebbinghaus used a lot of different methods,

but the most basic involves selecting some number of these random
three letter nonsense syllables to try grams and then memorizing them.

He would run through the list and then test his memory for them.

He would repeat this process over and over
until he reached some criterion of performance.

So, for instance, a typical one is he might repeat until he could
correctly recall the whole list correctly two times in a row.

This first phase of any study was referred to as the learning phase.

As soon as he was finished with the learning phase,
he entered into a forgetting phase.

Although was involved
here, was waiting for some time after that forgetting time, Ebbinghaus

would then test himself again
and see how many of the try grams he could still remember.

Pretty simple stuff so far, right?

The genius of having housework was how carefully he varied
how much he studied,

how long he waited, and and also just the sheer amount of data
he collected and analyzed.

One of the first things he discovered
is that forgetting is not a linear process.

We don't forget an equal amount of information across every hour.

In fact, most of the forgetting that's going to take place
after you've studied something happens relatively quickly,

right after you've finished studying.
An initial period of rapid information loss.

After that period, the rate of forgetting drops
and eventually levels off.

When Ebbinghaus plotted all of his data, a logarithmic curve emerged.

The amount of information retained follows a very predictable function
one that's subject to a law of diminishing returns.

In retrospect, I suppose this shouldn't have been so surprising.

The more you've forgotten, the less you have left to forget,
and so the slower the forgetting takes place.

What's surprising here is just how consistent this function is.

If you learn the material better, that is, if you study it more,
you will forget more slowly.

But the shape of the function will be the same.

It might be stretched to the right, reflecting slower forgetting,
but still that same basic logarithmic shape will emerge.

This work with random lists of try grabs, but it also works
with vocabulary words, pictures, movies, even calculus.

This ebbinghaus forgetting function seems to describe
something very fundamental about how the human brain encodes

and then slowly loses information.

Many researchers have proposed that this curve describes a general decay
process that happens within the human memory for any information.

When you learn something new,
whenever you experience things that you will later

remember, it's because there are structural changes
made to the pattern of connections in your brain.

This network of hundreds of millions of neurons and trillions
of synaptic connections between those neurons is always changing.

When new experiences come in through your eyes, ears and other senses.

The pattern of neuronal activation can be compared
to patterns of activation that were produced in the past.

If you recognize an image as something you've seen before,

this is a result of comparing the new information
to the previously encoded information.

Similarly, if you recall some piece of information, say,
the atomic weight of helium or the name of the capital of France,

you are somehow reading that information
out of the pattern of activity in your own brain.

This pattern of activity results from stimulation
from some areas of your brain, perhaps the areas that become active.

When you think of the words capital and France.

Any areas of your brain with information about names?

The fact that the term capital of France activates Paris and not London
is because of the elegant and precise patterns of connections

between the neurons up here in your brain.

New experience is encoded by changing the patterns of connectivity.

Any single, any single experience might result in slight changes
in the strengths of connections between dozens,

hundreds, even thousands of different neurons.

It's presumed that the complexity involved
in getting these connections right is part of the reason

that we need to study things repeatedly to get the information
encoded in a concrete, reliable fashion.

Now, these neurons in your brain are physical cells.

Sometimes they die off. Sometimes they get fatigued.

The strength of their connections
is subject to a fair amount of random variation over time.

The more time that passes, the more the neuronal networks in your brain
will change in what is essentially a random fashion.

These random changes are thought to produce a slow,
consistent decay in the information stored in our long term memories.

It's thought to be responsible for the consistent pattern of decline
associated with the Ebbinghaus forgetting curve.

Ebbinghaus wasn't finished at this point, however.

So far I've described how he memorized those try grams
until he reached a criterion,

then waited for some period of time, that forgetting phase,
and then tested his memory.

That's where these initial forgetting curves emerged.

The next thing Ebbinghaus would do is study his chosen set of try
Grams again until he reached that memorization criterion again.

The first thing Ebbinghaus noted was that there is a tremendous savings
in the RI learning time.

That's obvious, I think, for the cases
where he had only waited a short time and forgotten almost nothing.

But consider the situation where he waited for a long time, maybe
a few days until he had forgotten half of the tri grams on his list.

You might think you would need about half as much time
to relearn the ones that he'd forgotten.

That's not the case, however,

when he restarted, even when he couldn't recall most of the list.

A very brief additional study
time would push him back up to that criterion level.

This is referred to as a savings in relearning and like the forgetting
curves, it seems to be a near universal in terms of human memory.

Researchers have tried this with material ranging from math and language
learning to poetry memorization.

If you've learned something in the past,
even if you feel as if you've all but completely forgotten it,

traces of that memory are still present up there in your brain

and with a little refreshing.

Those traces can be greatly bolstered in strength.

Ebbinghaus has discovered
one more really important and fascinating thing with these experiments.

With some of the studies, he would learn to try Grams
then forget to try grams, then relearn the try grams.

And then here's the new thing here.

Track the forgetting that took place after the relearning,
the forgetting process still took place, of course.

Just because you've studied
something twice doesn't make it stick perfectly in your memory forever.

But the forgetting took place more slowly, in some cases,
really significantly more slowly.

And if you repeat this whole process,
wait for some forgetting to take place and then study it again.

A third time. A fourth time and more. Now,
the forgetting could be really, really slow.

I think every student should be taught

this as early as they are capable of understanding it,
maybe even in the third or fourth grade.

Certainly kids should be taught this as soon as they reach any stage
of their education where they're being asked to memorize anything.

If you want to be a good learner, you should really be
given some sort of a user's manual, some instructions on using it right.

The reason these particular experiments are so important,
however, is that there are some inherent studying guidelines here.

There's information
not just about how to study, but specifically when to study.

Say you want to learn some new set of information.

Could be anything memorized.

Poem list of presidents, anything, really.

Let's use the example of memorizing a list of the countries
in the European Union.

There are 27 countries there
alphabetically from Austria to the United Kingdom.

The first thing you need to do is read over the list
and then start practicing recalling it from memory.

This would be a hard task, of course, probably one that should inspire a

consideration of other aspects of learning and memory.

But let's just focus on timing for the moment.

Imagine, then, that you have succeeded in memorizing the lists
at least two times.

You've managed to list all of the all of those countries in Europe,
Austria, Belgium, Bulgaria, Cyprus, etc.

etc., etc. Spain, Sweden and the United Kingdom.

Time to take a break and do something else

Now. As you're doing something else, this memory will begin to fade.

Within a short period of time.

You will have started losing a few countries here
and there since you've just learned this list for the first time.

We're in the steepest part of that Ebbinghaus forgetting function.

If you want to remember the list
perfectly later on, there's only one solution. Study it again.

Probably within about an hour of that initial learning.

It would be a good idea to test yourself again.

Almost certainly you'll make a few mistakes,
but you can correct them, refresh your memory

and within a few rounds of practice
you'll be again back to that level of perfection.

Note that this second study session will probably be a quick one
since as Ebbinghaus first

discovered
the RE learning process is much faster than the initial learning.

Time to do something else again.

The forgetting will start happening again, of course, but it will happen
more slowly this time rather than studying again after an hour.

Probably two or 3 hours of forgetting.

Time is more appropriate again.

The information will get relearned even more quickly
and the subject of subsequent forgetting will be even slower

for the next cycle.

This process, you can probably double the time again to about 6 hours.

Maybe now you're into the next day
when you even more quickly relearned your way to perfection.

The next cycle can probably wait for 12 hours,
maybe right before bed the next day.

The next study session can be after 24 hours.

Then 48 hours. The 96.

I should note that all of these study
sessions are intended to be relatively brief.

By the fifth or sixth session, we're talking about spending just a few
minutes probably describing what might seem like a lot of studying.

But when you add it all up,
it might only be about an hour over the course of a week of studying.

I should emphasize here that the result is a very durable
and essentially perfectly accurate encoding.

It's worth contrasting this with the study habits
that I see with many of my undergraduate students,

most of whom have never heard of Ebbinghaus.

As I should say, when they arrive in my classroom.

Most students read their textbook before coming to class.

Some of them read their textbooks before coming to class.

They then listen to the session and participate in in the discussions.

For the majority of students, unless I I've given them some assignment
to force them to do otherwise, they then put away their notes.

And that particular section of the text until the night before the exam.

For some of these students,
the study sessions the night before can be truly epic.

Herculean all night endeavors.

It's not that these students lack commitment or devotion to the goal
of learning the material and getting a good grade in the class.

The problem is that they invest their study time inefficiently.
Really horribly. Inefficiently.

If they just learn to take

those hours of exam night studying and spread them out
over the course of the semester, they can remember more.

Much, much more.

With the same amount of effort,
in most cases, actually, with much less effort.

I think it's clear how one might schedule their studying for a course
with Ebbinghaus and his forgetting curves in mind.

They should review their notes and the text as soon as possible
after a class meeting, ideally within an hour or so.

They should plan to review it again several hours later and then
a day later, and then two days later and then four days and so on.

There are a lot of study sessions here, but as I've mentioned,
the relatively short and within a few days, very brief.

This stuff works. It works really well.

Ultimately, of course, the hope isn't just to free up more time
for students to spend relaxing, although that's fine too.

The real fruit of applying this method is that students
can learn more information in more detail with greater accuracy.

Undergraduate students can do it and so can we.

Okay,

so we've talked about forgetting in general
and about how to avoid it with repeated study sessions.

I also talked about one theory about why forgetting takes place decay.

I described how memories are stored as changes
to the patterns of interconnections between neurons in the brain.

That precise pattern breaks down over time, leading to forgetting.

But decay isn't the only theory about why forgetting takes place.

There are many theories
at least several of which are likely to be correct.

I've talked about how memory functions as a series of steps.

Let me review those
and then talk about some ways in which break downs in those processing.

In the processing at those steps can lead to forgetting.

So the steps I outlined in Session three involve three stages.

First, somethings perceived and transferred into short term memory.

Second, during the time that the information is in your short term
memory, it's encoded into the more durable long term memory.

And then the final stage, the memory is retrieved from your long term
storage back into short term or working memory,

at which point you're able to use the information

perhaps for answering your question on an exam, deciding
where you parked your car, and a thousand other tasks.

Forgetting or more specifically
a failure of memory can happen at all three of these stages.

Let's consider this with an example.

Say remembering
where you left your keys last night when you arrived home.

I have this problem a lot myself, actually. I've gotten much better
remembering it, however.

I'll tell you at least two good techniques for remembering things
like this after we consider how we failed to do so.

At times.

So the first stage in any memory is perceiving
the event to be remembered.

With the keys,

we may think that we've forgotten where we left them, but that presumes
that we actually perceived where we left them to begin with.

For instance,

maybe when I arrived home last night with my keys in my hand,
I was met by my beautiful daughter who eagerly ran up to give me a hug.

This is a wonderful thing,
actually far more important than my where my keys ever are.

So understandably,
I pick her up, swinger around and ask her how her day was.

If I'm lucky, she has some stories to tell me.

Maybe something to show me in a room. In the midst of all that,
she might hand me a piece of paper with a drawing on it.

I set my keys down in order to take the paper
and then devote my attention to the drawing and our discussion about it.

At some point, we move on to another room and voila,
my keys are left behind.

The next morning, as I hurry around the house looking desperately for
those keys, I'll be kicking myself for forgetting where I left them.

In a sense,
I have forgotten, but I really never had a chance in this situation.

You can't remember something that you never perceived.

This type of forgetting is usually referred to as a failure
to encode a failure to encode a memory.

The heart of the problem is the difference between our belief that
we should have a memory of something and the reality of the situation.

Failure to encode errors often arise
when people talk about errors by eyewitnesses, for instance, failures

by eyewitnesses to identify perpetrators of crimes.

For instance, at an armed robbery, as soon as the robber pulls out
a gun, people are understandably drawn to focus their attention on it.

So much of their looking and thinking is aimed at the gun that witnesses
may never actually look at the face of the armed robber.

Many hours later, when a police detective

is asking the witness to describe the face of the robber,
they may try, but will often be very inaccurate.

They'll feel as if they should remember the face.
How could they not? But it's not their fault.

They just failed to perceive it in the first place.

Okay. So that's forgetting at the first stage of the memory process.

How about that second stage in which we transfer things
from short term memory into long term memory?

It's pretty pretty easy to think of examples here
where a failure could occur.

I put my keys down on the table next to my bed and think to myself,
Remember the keys, Remember the keys?

Remember the keys? Maybe say it out loud three times like that.

I had the information in my short term memory,
but it never made it to my long term memory.

As soon as my attention turned to something else,
it faded from my short term memory and was gone.

Okay, so imagine I did encode the information
about my keys in my memory, well into my long term memory.

I perceived the location of the keys.

I thought of some way to relate that location
to things that I already know.

Maybe even General generated some sort of a good image,
the way we talked about.

Maybe I imagined giant keys sitting on my pillow,
tucked into bed saying the key to happiness is a good night's sleep.

So I'm all ready to recall the key location tomorrow morning.

But still, there's a chance I might forget how

we've already talked about the process of decay
in the context of the Ebbinghaus forgetting function,

the specific changes in the neuronal networks of my brain

light decayed over the course of the night,
such that I will fail to remember the key location.

That's unlikely, actually, given the short amount of time.

But there are two other types of errors

that can lead to a failure to retrieve information
even after it's been encoded properly in my long term memory.

The first is called retroactive interference.

That is when you learn new things.

Those new things can interfere
with the information that you've learned in the past.

My favorite example of this is when you try to remember the phone number
of a knew of a house where you used to live.

You might have remembered that phone number for years and recalled it
many, many times.

Once you moved to a new house and learned the new phone number,

the new phone number that goes with the new address.

That information seems to overwrite in a way
that old original information.

The human brain represents information in an associative fashion.

We recall information by having a cue that connects with it.

If I ask for a home phone number,
you have a strong association to your current number.

When you move from one house to another.
That associative connection gets remade to the number.

So learning new information can impact your memory.

For older information, four can impact your ability to retrieve it.

With my key remembering example, I can imagine how this might apply.

Perhaps my wife talked to me about the location of her keys
and asked me to help remember them.

So I mentally made the association of key location
to bowl on the kitchen counter where her keys were.

Its possible that this new association

will compete with or completely block my existing association
between key location and bedside table where I left them.

That would be retroactive interference, proactive interference.

This is a little more complicated to describe in the abstract, but
I think it's the most likely culprit here if I forget where my keys are.

The basic idea of proactive interference
is that information you've learned in the past

can interfere with the learning of new information in the future.

There are many examples of this from everyday learning.

There's good evidence, for instance, that college students retain
more of the information they learn from their first year classes

than from their second and third year classes.

Something about the new context of college
seems to provide a new wave of fresh associative connections

as they learn more and more in college, however, that existing knowledge
seems to interfere with the acquisition of additional knowledge,

as the location based associations have to be spread
more and more thinly.

With my keys, if I forget where they are in the morning,
proactive interference is the most likely culprit.

I've encoded where I left them on one particular evening,
but I've also done that dozens, hundreds of other times

using different locations.

The concept of key location is associated already with kitchen tables,
counters, coat pockets, work bags and many other locations.

Even the lock of the front door itself, the learning of those
associations in the past makes it extra hard to learn a new association.

That's proactive interference.

All of this leads to
what is a really simple way of remembering where your keys are.

Always put them in the same place.

For instance, a hook or tray near your front door.

In this case, the interference of prior memories will actually become
a benefit as you repeatedly encode the same location for your keys.

There's one other phenomenon of forgetting
that's worth discussing here called the tip of the tongue phenomenon.

It's related to retroactive and proactive interference,
perhaps even caused by it.

But it's a particular type of forgetting

that's specially associated specifically associated with a failure
to retrieve the information in that third stage.

The very salient part of the tip of the tongue phenomenon

is that you can't recall some piece of information,
but you feel very strongly that you do know it.

It seems to be there in your head.

It's like it's just out of reach as it feels, as if you can almost say
the answer to the memory question you have, but just not quite.

Colloquially,
this is described as having something on the tip of your tongue.

Several researchers have explored this phenomenon
and found evidence that it's quite real.

First, the participants essentially never experienced this phenomenon.

For information
they haven't learned, they do every so often. But it's very, very rare.

Extremely rare.

Second, for facts that were reported as being on the tip

of their tongues, participants are likely to be able
to recall that information eventually after some delay.

So if you think you know something,
even if you know that you can't recall it at the moment,

there's a pretty good chance that the information is up there
and that it will eventually be retrievable.

So what should we do in that situation?

How can you get around one of these failure to retrieve mental blocks?

One of the best things to do, actually, is that just wait.

Think about something else when you return to the question
a few minutes later, the new mental set

that you've created may enable you to access that information.

So with all of this in

mind, how should we remember things like where we put our keys,
our medicine bottles, our books, other things?

Are there good ways we can apply our knowledge
of what human memory is good at to enhance our memory performance here?

My favorite tip here is to enhance
the encoding of the location of your keys or whatever the other item is.

With a good mental image at the time you set them down.

My keys, my key ring has always looked at me a bit like a flower
with the keys forming the petals and the ring the center

as I set them down somewhere, especially if I'm concerned
about remembering where they are in the future.

I like to think of it as planting a flower in that location.

In my vivid mental image here, the flower grows while I'm away, lifting
up on a long, stout stem and leaning towards the nearest sunny window.

The human brain is good at locations in images, right?

If you develop your own key image,
I can't promise you will never lose your keys again.

But I can assure you that you'll lose them less often, especially on
occasions when you take the time to perceive and encode their location.

This session has been about forgetting failures
to encode decay, retroactive interference and proactive of interference.

We also talked about some of the timing of the forgetting process
as first studied by Hermann Ebbinghaus.

And the way to schedule study time around those timing characteristics.

This session is very much the
the sort of story that I hope to convey with this course

as you get to know the operating characteristics
of this information processor that you have between your ears.

Both the strengths and limitations of that information processor,
you can make better and better use of it.

Hopefully your memories about forgetting
help you to forget future memories less.

I'll see you next time.