YouTube Transcript:
The neuroscience of memory - Ri Science Podcast with Charan Ranganath

[Music]
Chon thank you so much for joining us
today here at the Royal Institution and
congratulations on your new book why we
remember um for those of you listening
or watching uh By the time this is
released the book will be out so if you
haven't read it yet this is your
official reminder to do so um your book
and your research Focus particularly on
what's going on in our brains to
facilitate our memory all the way down
to the cellular level um and how we can
use it more
effectively what inspired you to get
into this area of research and write the
book well one of the things that really
inspired me was when I was in graduate
school I was doing Clinical Psychology
and many of the people that I was
working with had U were worried that
they had brain damage or they were
referred by a doctor or sometimes even a
lawyer and uh we would do testing with
these p patients and what I would find
is is that whatever the disorder it was
whether it was something like you know
pre-alzheimer's disease to traumatic
brain injury to even things like
clinical depression uh we would see that
memory was affected it was the first
thing to go and uh at the same time I
was doing clinical work in Psychotherapy
and finding that the patients that I was
working with uh they wanted to they had
the opposite problem of too much memory
and they had these traumatic memories
that they were trying to work through
and so it really became apparent to me
how important memory is and at the same
time uh I also felt that we didn't have
a very strong scientific basis for
understanding how to treat memory
problems and so at that time functional
MRI or functional magnetic resonance
imaging was this technique that was
being just getting off the ground as a
way of looking at people's brains and
being able to see what areas of the
brain are actually doing things in
relation to memory and so I jumped on
that
opportunity so you've already mentioned
like memory underlines basically
everything we do it's our Behavior
relationships careers um and we're all
learning from the experiences that we've
had and we're retaining that information
but it's also something that a lot of us
do take for granted you've mentioned
that some people have problems with it
and it's also quite complicated could
you give us a bit of a whistle stop tour
of our current understanding of how
memory works
so it's such a big topic I'll try to
reduce it to a few key points um but one
of the uh key ideas is is that uh
memories are thought to be stored so to
speak in uh through connected networks
of neurons which are the basic cells
that are doing the Computing in the
brain and so what we think happens is
when you speak when you feel something
when you see something there are groups
of neurons that are activated that allow
you to have these conscious experiences
or even unconscious experiences things
like your heart racing when you you
something scares you and so when we have
these experiences the changes um there
changes in the connections between these
neurons and so that's what you'd call
plasticity and those changes allow us
group of cells that were active during
an event to become more easily activated
again later even when that event isn't
taking place so if I could activate some
of the neurons in this cell assembly is
what we call them uh from this
conversation presumably through these
strengthened connections the other
neurons could be brought into play and
so it only takes me a small cue to be
able to pull up the entire network of
neurons that are part of this memory so
that's like kind of the very Whistle
Stop tour I guess but then you can
certainly go deeper than that into all
the different kinds of memory and and
the way it works I mean so we've got um
I'm going to keep it very basic even
though it's very complicated but we've
got short-term memory long-term memory
um and different types of long-term
memory could you give us a bit more
about that in memory research we like to
cut things up and say this is different
and that's different but in reality when
we remember an event for instance it's
all very connected um some of the key
one of the key distinctions that I've
really been working on throughout much
of my career is the distinction between
episodic memory and semantic memory so
semantic memory would be your general
knowledge about things that you've
acquired over the years and episodic
memory would be your ability to remember
a singular event whether it's an event
that's mundane like I'm in the kitchen
and I'm trying to figure out wait a
minute why am I here and we could talk
about other kinds of episodic memory
like where did I put my keys to I'm
vividly remembering my 16th birthday or
I'm trying to remember that later on I
have to take out the trash because it's
trash day so all of these kinds of
memory are related to episodic memory
and we know that there's a brain area
called the hippocampus that's very
important for episodic memory so people
with damage to this area they can still
walk talk think but they have an
exceptional problem being able to
remember things related to a time and a
place so um there's many famous patients
including uh some that have been uh um
described at University College London
who like can learn information but they
don't have this ability to re-experience
it and tie it to a context in the way
that somebody with a hippocampus can do
and so what that tells us is is that the
hippocampus seems to be organizing and
associating our experiences so it's
getting information about who was in an
event and where the event took place and
how things transpired what was the
situation what were the feelings that we
experienced at the time and it's
associating them all and it's just
putting it together just because at this
moment in time all of these things
happen at this moment in time I was
excited to tell you about memory and at
the same time you were here and at the
same time I'm sitting on these purple
chairs or whatever this color is right
and so all of those bits are being put
together by The Hip campus and what that
allows you to do is some other time if I
walked into this room that would be
enough to cue me to be able to allow the
hippocampus to pull out all the other
bits that happened in this past event
and so that's why for instance if you
end up in the kitchen and you're trying
to figure out why am I here you actually
have to pull up information about that
past context when you decided you were
going to go into the kitchen in the
first place and so I then essentially
engage in what uh endel tulving the
psychologist who came up with this term
he called this episod U he called
episodic memory is mental time travel
and the idea is that you have to go back
in time mentally and be able to say hey
wait a minute okay I'm here in my office
and I realize that I left my phone in
the kitchen and that's why I'm there and
so otherwise what happens and this often
happens to me is I end up in the kitchen
and I forget why I'm there and then I
end up just eating something and then I
come back and then I realize why I went
to go there at the first place but being
able to visualize that context the place
and time really helps and in fact we've
uh and many others have done research
where if you scan people's brains while
they're experiencing an event and then
you scan them while they remember that
event what you can see is this this
little bit of a reset in the patterns of
activity in the hippocampus such that it
looks more like what it did in the past
so these patterns change a lot over time
but when we remember you can reverse
those effects of time so it's almost
like being in a time machine essentially
you're going backwards in time and one
of the fascinating things about this is
that as a result we can actually change
the past when we remember so we tend to
think of memory being just a replay like
you would play a movie on your phone but
in fact the act of remembering a past
event can lead to changes and updating
in the memories and this has been shown
in everything from uh rats and very
simple kinds of memories to humans and
much more complex memories so for me
personally and I know it happens to a
lot of people there are certain um like
smells and songs that bring back very
specific memories that I didn't realize
I had um why does this happen well one
of the things that I think people don't
realize this is that sometimes a memory
isn't forgotten but we just don't have
the right way to access it and so when
we hear a song for instance that comes
out of the blue and it just pops this
memory of a particular time back in our
head often it's the case that that song
was a particular soundtrack for us and
uh during a particular period in our
life and it evoked a particular emotion
and so as a result they're baked into
the context of these episodic memories
and so they provide this beautiful way
of accessing memories for a time and a
place so sometimes we had sad periods in
our life and you listen to sad music or
you have periods of time where you just
decide to get into electronic dance
music and then you go through a
classical phase and so forth and those
songs that are associated with
particular points in our life tend to be
the ones that are the best retrieval
cues to pull out these once dormant
memories and uh but it's not just of
course music smells can do this being in
a particular place can do this as long
as there's some kind of a context uh
that is unique it will help us access
those memories so lots of memory tips
and tricks out there and one that I've
heard is that if you revise for a test
in the same place you're going to take
the test um those memories should come
flooding back to you
um does that work and if so why and if
not what are the other like memory
misconceptions that are out there well I
think it can work if you're not using
good study habits in the first place so
if you're really just have this unique
memory of having studied something once
then being back in that same seat and
looking at the same scenery can help you
pull up that information that you
studied just once but ideally what you
want to do is take advantage of the fact
that memories update when we remember in
different context so if you study or
even better if you test yourself on what
you've learned let's say in this room
and then you go home and you do it then
you go to a your favorite Cafe and you
do it what happens is every time you
recall this uh information that you
studied now what happens is is it
becomes updated so it's not tied to this
one place it can now generalize to
different places and at some point what
happens is is that the memory for the
information you need is accessible
anywhere and so ideally what you want is
you don't want to have to be sitting in
one place to remember everything that
you're trying to remember you want to be
able to get it from anywhere else and so
I think one you mentioned a memory
misconception I think one misconception
is is that if you study read the same
thing over and over over again and then
you take the test in the same spot or
whatever that you're going to retain
this information and it's just not true
actually you retain much more
information if you force your brain to
struggle and so by that I mean that um
you try to test yourself and sometimes
you can even test yourself before you
study and that can actually enhance
memory too and so this is a a method
that I talk about in the book called
error driven learning and the idea is is
that when you actually force yourself to
try to pull up information from memory
and challenging
circumstances what happens is when you
do get the right answer essentially
there's changes We Believe anyway uh
based on our computer models of the
brain that essentially there's parts of
a cell assembly that gives rise to a
memory that are not necessarily very
helpful and adaptive and there's some
that are very very effective right so
what happens is is that uh the memory
can get reconfigured and tuned up so
that the bad connections go and the good
connections are
strengthen and that only happens though
if we allow ourselves to be stress
tested in the first place so even though
there is a long way to go to
understanding Neuroscience um we have
come a remarkable way but there is a lot
of progress that we need to make um we
are also sat in the Royal Institution uh
in the theater where the Christmas
lectures are filmed so I thought I'd go
over a couple of Christmas lectures that
we've had in the past are based on
neuroscience and memory um so we've had
Richard Gregory who gave his lectures
the intelligent ey back in 1967 which
was the first Christmas lectures ever
broadcasting color and only the second
ever broadcast on bbc2 we've had Susan
Greenfield delivered uh the Christmas
lectures in 1994 titled the Journey to
the Center of the brain so that's quite
fun and then the most recent one is
Bruce Hood who in 20
delivered the series called meet your
brain how have you seen the
understanding of memory and Neuroscience
develop over your career well one of the
big developments that I've seen over my
career has been the ability to move away
from thinking that memory is just one
brain area to really thinking about
networks in the brain and how they
interact with each other and uh so
because before brain Imaging was really
big people would focus on specific areas
of the brain like the hippocampus but
now one of the things we appreciate is
that memory is more than just the
hippocampus it's just that some areas of
the brain it's harder to find disorders
that affect it so for instance there's a
whole network of brain areas called the
default mode Network and people used to
think this is just a network that comes
on when you're not doing anything uh and
now we realize it's actually
exceptionally important for memory and
if you look at the pathology in the
brain that happens over the course of
Alzheimer's disease the hippocampus is
part of the you know Ground Zero but
then soon soon after you start to see
pathology throughout the whole default
mode
Network um you mentioned Alzheimer's
disease um so obviously we've got
neurodegenerative diseases that are
marked by memory loss um what goes wrong
to cause these like memory
dysfunction well it's still under debate
but one of the ideas that's out there is
is that
brain areas aren't it's not like you can
go to any part of the brain and find
connections to every other part of the
brain right it's just like if I go to a
tube station there's certain lines that
you can get onto and there's other lines
which aren't really connected to each
other and then there's some lines that
are like uh there's some stations that
are hubs where you can transfer from one
station to another and you can think of
the brain in a similar kind of way that
there are these networks of Highly
interconnected neurons where if you're
on one part of the line it's very easy
to get to another part of the line but
then there are some networks where the
between network connections are very
weak and so what that means is it allows
the brain to both be specialized but at
the same time it gives some flexibility
to allow areas of the brain to talk to
each other now one of the consequences
of this is if there's pathology uh like
the buildup of uh toxic proteins like uh
towel for instance that what it can do
is you can get this uh spread from one
neuron to another to another and so that
can cause an entire network to basically
collapse when you have the have the
spread of a neurogenerative disease like
that and your team at UC Davis are
working on ways to catch signs of
Alzheimer's earlier um and try and
prevent damage before it starts how does
that work well so there's many different
approaches that are being tried out
right now to catch uh the pathology very
very early because one of the big
discoveries in Alzheimer's research uh
that was that you can actually start to
see buildup of heavy buildup of this
protein called amalo which is thought to
play a part in the whole Cascade of
events that causes the brain damage and
Alzheimer's uh and if people were trying
to do drug Discovery what they would do
is they would try out these drugs and
people after they've been diagnosed with
a clinical memory problem but now we
know that by the time someone has memory
problem that's easy to detect what they
find is is that there's already pretty
significant loss of neurons and you
can't really reverse that so if you want
to come up with a intervention you want
to be able to intervene before that
brain damage really kicks in and so now
we realize there's this long preclinical
period And so one of the uh one approach
that's being done is to look at blood
plasma uh tests and that that looks very
promising uh in our lab we're hoping to
develop behavioral tests that you know
any neurologist could give so that
there's different ways of identifying
people in these early stages and so one
of uh our lines of research is focused
on this interaction between the hip
campus and the default Network and we've
got good data now from a very large
group it was a um 346 people scanned
actually in the University of Cambridge
and we analyzed that data and found that
there are these certain points when
people people are say watching a movie
for instance where the hippocampus seems
to come online and that's related to
memory and uh and the default Network
too and these uh activations reduce as
people get older but what's interesting
is that independent of age people with
poorer memory performance show less of
this variability in the hippocampus and
so what that tells us is that we can use
these as biomarkers for memory problems
or bi ological markers so to speak so
keeping on the theme of looking into the
future and future research um we get
this question a lot on well artificial
intelligence is a big topic we've just
had the 2023 Christmas lectures on it um
and it's being integrated into society
uh more and more so how will artificial
intelligence
um impact our memory or is there any
concern around that well there's there's
many different kinds of concerns uh what
I would say is is uh certainly one issue
that I I worry about is uh what you can
see with a lot of the issues that come
up with technology in terms of our
thinking and our memory functions are
not really related to the technology per
se but the way we interact with it and
so one of the things that I think about
quite a bit is uh now for instance with
uh um this actually happened a few years
ago when I would start using email
programs they would use AI to finish my
sentences and about onethird of the time
I would say you know uh no this isn't
what I want and just type it myself
onethird of the time I'd say this is
exactly what I want just fill it in
onethird of the time I'll say that's
good enough let's just go with that
because I didn't want to type the rest
of the sentence and it's not what I
would have wanted to type but it goes
anyway but the brain as I said is very
plastic and it's constantly reorganizing
to be more and more efficient and so as
a result what happens is is that often
people can start to just learn to adapt
exceptionally well to whatever is being
suggested to them and the downside of
that is is that people can actually
become less creative and you might end
up just sort of seeing this like funnel
towards Mass mediocrity essentially if
everyone it's not that chat GPT would
write what you want it's more like
everyone would start writing like at GPT
I hope that doesn't
happen I hope so to I hope not do um so
you mentioned the potential lack of
creativity how does memory help us be
creative so memory is incredible in
humans because we do things that are
very different than typical generative
AI systems so generative AI they just
absorb massive amounts of data and then
try to find patterns in it humans
actually take in very weird kinds of
data like we don't even see the entire
world in front of us we just get little
Snippets of the World by moving our eyes
around and so as a result we actually
put together much more Rich kinds of
inferences for much less
information so you wander around the
world you meet people maybe in your when
you're in uni you lived in a dormatory
with uh friends who you wouldn't have
met if it just didn't happen that you
were living in the same place and these
kinds of unique experiences somebody
gives you a book for your birthday and
you read it and it's something
completely outside of your normal
experience you can start to build
connections between ideas and if you
look at some of the great artists some
of the great musicians poets there are
actually people who took a diverse range
of influences from their lived
experiences and just found unusual
connections between them and this is the
kind of thing that we can do because of
episodic memory episodic memory allows
us to take these unique moments in our
lived experiences and then recombine
them and put them together into new
things and synthesize them and this
would be something that is not present
in at least current AI systems because
they're designed to just take the common
elements of whatever they're trained
with as opposed to the distinctive
unique weird experiences that we have
and also in your book you mentioned to
me in the same thread uh imagination
um how does memory help us with that
well so one of the early memory
researchers uh that's been just a huge
influence in the field was Frederick
Bartlett um who was actually at the
University of Cambridge and Bartlett
proposed a really radical idea remains
radical to this day that essentially we
don't remember the past by replaying a
video in our heads even though it feels
like that we actually only get little
glimpses of those past experien es so we
don't activate the entire set of neurons
that we're part of an event we actually
just get little bits and pieces but then
we're able to use our knowledge and use
our goals and use our everything that we
have to build a story and imagine how
the past could have been and that's the
act of remembering and so this is why
there's this just amazing link between
memory and
Imagination uh so for instance a number
of researchers including Demis hbus who
started uh Google Deep Mind and uh
Elanor Maguire uh discovered that many
patients who had memory problems also
had a very impoverished imagination and
when people started to scan brains while
healthy people were just imagining or
healthy people were remembering
experiences from their own lives what
they found was is that basically the
same networks of brain areas in the
hippocampus and the default mode Network
come alive and so there seems to be this
element where imagination is the core of
memory but also memory is the core of
imagination that is if you have a memory
problem your ability to think about to
imagine how things could be in the
future or even Imagining the present and
I know that might sound a little weird
that we can imagine the present but you
know one thing that we're doing in this
conversation is we've got this common
ground because I have some understanding
of of what you are asking me that goes
beyond the simple words that you put out
right and memory is very important for
being able to connect those dots and
that kind of uh ability to use memory
flexibly to generate in things that
aren't in front of us is exceptionally
important for imagination and for
creativity but it also creates some huge
problems because we have a very hard
time telling the difference between
things that we imagine versus things
that actually happen so I don't know
about you but I often have cases where I
get an email and uh I think oh yeah I've
responded to that and then a week later
it comes back and it says urgent please
reply and I realized that I didn't
respond to it I just thought about it
and then I said oh this is how I should
reply and then I got interrupted and
distracted and I never did it and so we
have this real challenge of being able
to say is this something that I thought
about or is this something that actually
happened and sometimes people can
actually get so far out in this that
they can remember events that never
actually took place if they imagine them
too vividly so I'm sure that everyone uh
listening to this and watching this uh
podcast episode has one thing that they
can never quite remember um it's at this
point that I'm meant to say one thing
that I can't remember but ironically I
can't actually remember what I can't
remember so we're going to move on um
and one of the themes in your book is
how we can use our memory uh to our
advantage and more efficiently uh for
all our sakes especially mine uh what's
the
secret I think the secret is you have to
first accept the fact that memory is not
supposed to be an effortless process
where you can record every part of your
experience and replay it exactly as it
happened I think a lot of people start
with this assumption that it should be
effortless it should be complete and it
should be 100% accurate and when you
start with those assumptions you're
going to make mistakes and that's in
some sense normal but on in another
sense it won't help you remember the
things that you want to remember it well
um so we know that the information that
tends to get forgotten uh is suffering
because memories compete with one
another and this is a way that one of
the many ways in which the human brain
is different from machine intelligence
where we really
store Many Memories uh with
overlapping Assemblies of neurons and so
what happens is is that activating one
memory can actually compete with or
inhibit another memory so if you take
the kinds of things that happen every
day like you know I put down my keys and
then later I'm saying where did I put my
keys it's a very difficult problem
because you have memories of where I put
my keys the an hour ago versus 3 hours
ago versus one day ago versus two days
ago and so it's not a matter of just
remembering where I put my keys but
finding the right memory of where I put
my keys that's the issue so how do you
beat that interference one way you can
beat that interference is by first of
all capturing and paying attention to
what are the important features in your
environment and that sounds very easy
but it's not especially in the world we
live in you know people are always being
bothered by uh text alerts and so forth
we're often doing one thing but engaging
in all sorts of mental simulation and
planning for other things that we're
supposed to be doing and actually
engaging in a lot of memory processing
you know thinking about the future
thinking about the past and so that
really distracts us from the present
moment and makes it that we're not even
there in the first place when you put
down your keys um and we know that the
kinds of memories that really benefit um
and are able to come up at a moment's
notice are the ones where we have a lot
of sensory detail for instance um just
Vivid information about the sights and
the sounds and the smells and the
emotions that make a memory distinct
from other memories that they could be
competing with and so anything that you
can do to focus yourself on those moment
ments that make an event unique will
help you remember what you need to
remember and I think part of what we
need to do is actually go ahead and
think about well what do we want to reme
remember in the first place because if
we're not going to remember everything
then the question is what do you need to
remember what do you want to remember
and how do you want to remember it so
we've covered how we can access memories
with unique events and different senses
um but there are obviously some lify
lifestyle choices which lead to memory
loss um for one the consumption of
alcohol um why does this happen well
alcohol is one substance and there are
other drugs like this too like for
instance benzodiazapines like valum and
Xanax that actually can have an amnestic
effect because essentially there are
certain uh chemicals that are released
in the brain during particularly
interesting or new experiences for
instance or emotional experiences and
alcohol can block some of those uh
changes that happen when these chemicals
are released in the brain and so what
happens is is that it actually can block
the formation of a new memory as a
result so sometimes you can even see
after someone's had alcohol that they
have better memory for the stuff that
happened before they started drinking
but worse memory for the things that
happen while they were drinking because
what happens is that now you've lost all
this memory for while you were drunk and
so the stuff that happened before is now
not suffering from competition as a
result um alcohol has another
interesting effect which is that when
you are under the influence of alcohol
you're in a particular mental context so
I think many people after like three or
four drinks are in a different mental
state than they are in their daily life
hopefully anyway and so as a result you
can't access whatever memories you did
form because you'd have to get back into
that mental state again and there's some
evidence suggests that in fact if you
did form memories while you're under the
influence of alcohol you can get better
access to it if you're remembering under
the influence of alcohol which I don't
recommend but nonetheless that yeah and
there's other effects of alcohol so for
instance uh it can give you these
rebound things where it messes with your
sleep so even though alcohol can help
you get to sleep sleep quality is
terrible you wake up in the middle of
the night and all those factors can lead
you to have the next day um from all the
sleep deprivation among other things uh
your brain isn't going to function as
well in particular the prefrontal cortex
is an area of the brain that's important
for allowing us to focus on what we need
to learn and it allows us to be able to
reconstruct the context of past events
so that we can actually pull up the
right memories that we're looking for
and the prefrontal cortex is really shut
down when we're in a state of sleep
deprivation so there's many of factors
uh lifestyle factors and of course
alcohol is not the only drug that
affects memory there are many others uh
um THC for instance and marijuana can
affect memory um psychedelic drugs
actually have interesting effects on
memory we're still trying to unpack all
of them uh they do seem to change
plasticity in a way that can affect
memory for while you're under the
influence but I think the effects are
actually quite complex and interesting
uh but the effects on memory are
probably the reason why some uh
psychedelic drugs like MDMA seem to have
uh an effect on traumatic memories
because often what happens is people
retrieve these old memories but you're
also producing these changes in the
ability to form new memories while
you're retrieving these old memories and
so what can sometimes happen is the
memory can become updated in interesting
ways so I wish we could continue this
conversation all day but in true RI
fashion I'm going to put you on the spot
to finish the episode not that I haven't
been throughout this whole thing um so
if there is one thing you want people to
remember from this episode or the book
again everyone should read it uh what
would it be I would say that the one
thing I would want people to take away
is that memory is not about the past
it's about the present and the future
and that we're constantly using memory
to be able to tell ourselves everything
from where am I at the given moment what
time is it to things that are the big
choices that we want to make in life I
mean I know when I chose my career path
part of my key decisions were always
based on what are the memories that I
have of Past Times where I've been happy
what are the kinds of contexts in which
I function well in right and so uh I
really want people to think about memory
as this resource that they can draw upon
uh for making better decisions and for
being more present and for choosing the
kinds of Life uh paths that you want to
take because the thing is it's going to
influence your behavior and if you're
not aware of it then memor is in the
driver's seat but you really just want
it to be the
co-pilot excellent Sharon thank you so
much for joining us today um at thank
you for such an incredibly insightful
episode excellent thank you
[Music]