The beautiful mysterious science of how you hear Jim Hudspeth

Transcriber: Joseph Geni
Reviewer: Camille Martínez

Can you hear me OK?

Audience: Yes.

Jim Hudspeth: OK. Well, if you can,
it’s really amazing,

because my voice is changing
the air pressure where you sit

by just a few billionths
of the atmospheric level,

yet we take it for granted

that your ears can capture
that infinitesimal signal

and use it to signal to the brain
the full range of auditory experiences:

the human voice, music, the natural world.

How does your ear do that?

And the answer to that is:

through the cells that are the real hero
of this presentation –

the ear’s sensory receptors,

which are called “hair cells.”

Now, these hair cells
are unfortunately named,

because they have nothing at all
to do with the kind of hair

of which I have less and less.

These cells were originally named that
by early microscopists,

who noticed that emanating
from one end of the cell

was a little cluster of bristles.

With modern electron microscopy,
we can see much better

the nature of the special feature
that gives the hair cell its name.

That’s the hair bundle.

It’s this cluster of 20 to several
hundred fine cylindrical rods

that stand upright
at the top end of the cell.

And this apparatus is what is responsible
for your hearing me right this instant.

Now, I must say that I am
somewhat in love with these cells.

I’ve spent 45 years in their company –

(Laughter)

and part of the reason
is that they’re really beautiful.

There’s an aesthetic component to it.

Here, for example, are the cells

with which an ordinary chicken
conducts its hearing.

These are the cells
that a bat uses for its sonar.

We use these large hair cells from a frog
for many of our experiments.

Hair cells are found all the way down
to the most primitive of fishes,

and those of reptiles often have
this really beautiful,

almost crystalline, order.

But above and beyond its beauty,

the hair bundle is an antenna.

It’s a machine for converting
sound vibrations into electrical responses

that the brain can then interpret.

At the top of each hair bundle,
as you can see in this image,

there’s a fine filament
connecting each of the little hairs,

the stereocilia.

It’s here marked with
a little red triangle.

And this filament has at its base
a couple of ion channels,

which are proteins that span the membrane.

And here’s how it works.

This rat trap represents an ion channel.

It has a pore that passes
potassium ions and calcium ions.

It has a little molecular gate
that can be open, or it can be closed.

And its status is set by this elastic band
which represents that protein filament.

Now, imagine that this arm
represents one stereocilium

and this arm represents
the adjacent, shorter one

with the elastic band between them.

When sound energy impinges
upon the hair bundle,

it pushes it in the direction
towards its taller edge.

The sliding of the stereocilia
puts tension in the link

until the channels open
and ions rush into the cell.

When the hair bundle is pushed
in the opposite direction,

the channels close.

And, most importantly,

a back-and-forth motion
of the hair bundle,

as ensues during the application
of acoustic waves,

alternately opens and closes the channel,

and each opening admits millions
and millions of ions into the cell.

Those ions constitute
an electrical current

that excites the cell.

The excitation is passed to a nerve fiber,

and then propagates into the brain.

Notice that the intensity of the sound

is represented by
the magnitude of this response.

A louder sound pushes
the hair bundle farther,

opens the channel longer,

lets more ions in

and gives rise to a bigger response.

Now, this mode of operation
has the advantage of great speed.

Some of our senses, such as vision,

use chemical reactions that take time.

And as a consequence of that,

if I show you a series of pictures
at intervals of 20 or 30 per second,

you get the sense of a continuous image.

Because it doesn’t use reactions,

the hair cell is fully 1,000 times
faster than our other senses.

We can hear sounds at frequencies
as great as 20,000 cycles per second,

and some animals have ever faster ears.

The ears of bats and whales, for example,
can respond to their sonar pulses

at 150,000 cycles a second.

But this speed doesn’t entirely explain
why the ear performs so well.

And it turns out that our hearing
benefits from an amplifier,

something called the “active process.”

The active process enhances our hearing

and makes possible all the remarkable
features that I’ve already mentioned.

Let me tell you how it works.

First of all, the active process
amplifies sound,

so you can hear, at threshold,
sounds that move the hair bundle

by a distance of only about
three-tenths of a nanometer.

That’s the diameter of one water molecule.

It’s really astonishing.

The system can also operate

over an enormously wide dynamic range.

Why do we need this amplification?

The amplification,
in ancient times, was useful

because it was valuable for us to hear
the tiger before the tiger could hear us.

And these days, it’s essential
as a distant early warning system.

It’s valuable to be able
to hear fire alarms

or contemporary dangerous such as speeding
fire engines or police cars or the like.

When the amplification fails,
our hearing’s sensitivity plummets,

and an individual may then need
an electronic hearing aid

to supplant the damaged biological one.

This active process also enhances
our frequency selectivity.

Even an untrained individual
can distinguish two tones

that differ by only
two-tenths of a percent,

which is one-thirtieth of the difference
between two piano notes,

and a trained musician can do even better.

This fine discrimination is useful

in our ability to distinguish
different voices

and to understand the nuances of speech.

And, again, if the active
process deteriorates,

it becomes harder to carry out
verbal communication.

Finally, the active process is valuable
in setting the very broad range

of sound intensities
that our ears can tolerate,

from the very faintest sound
that you can hear, such as a dropped pen,

to the loudest sound that you can stand –

say, a jackhammer or a jet plane.

The amplitude of sounds
spans a range of one millionfold,

which is more than is encompassed
by any other sense

or by any man-made device
of which I’m aware.

And again, if this system deteriorates,

an affected individual
may have a hard time

hearing the very faintest sounds

or tolerating the very loudest ones.

Now, to understand
how the hair cell does its thing,

one has to situate it within
its environment within the ear.

We learn in school
that the organ of hearing

is the coiled, snail-shaped cochlea.

It’s an organ about
the size of a chickpea.

It’s embedded in the bone
on either side of the skull.

We also learn that an optical prism

can separate white light
into its constituent frequencies,

which we see as distinct colors.

In an analogous way,

the cochlea acts as sort of
an acoustic prism

that splits apart complex sounds
into their component frequencies.

So when a piano is sounded,

different notes blend
together into a chord.

The cochlea undoes that process.

It separates them and represents
each at a different position.

In this picture, you can see
where three notes –

middle C and the two
extreme notes on a piano –

are represented in the cochlea.

The lowest frequencies go all the way up
to the top of the cochlea.

The highest frequencies,
down to 20,000 Hz,

go all the way to the bottom
of the cochlea,

and every other frequency
is represented somewhere in between.

And, as this diagram shows,

successive musical tones are represented
a few tens of hair cells apart

along the cochlear surface.

Now, this separation of frequencies

is really key in our ability
to identify different sounds,

because very musical instrument,

every voice,

emits a distinct constellation of tones.

The cochlea separates those frequencies,

and the 16,000 hair cells
then report to the brain

how much of each frequency is present.

The brain can then compare
all the nerve signals

and decide what particular
tone is being heard.

But this doesn’t explain everything
that I want to explain.

Where’s the magic?

I told you already about the great things
that the hair cell can do.

How does it carry out the active process

and do all the remarkable features
that I mentioned at the outset?

The answer is instability.

We used to think that the hair bundle
was a passive object,

it just sat there, except
when it was stimulated.

But in fact, it’s an active machine.

It’s constantly using internal energy
to do mechanical work

and enhance our hearing.

So even at rest,
in the absence of any input,

an active hair bundle
is constantly trembling.

It’s constantly twitching back and forth.

But when even a weak sound
is applied to it,

it latches on to that sound
and begins to move very neatly

in a one-to-one way with it,

and by so doing, it amplifies the signal
about a thousand times.

This same instability also enhances
our frequency selectivity,

for a given hair cell
tends to oscillate best

at the frequency at which
it normally trembles

when it’s not being stimulated.

So, this apparatus not only
gives us our remarkably acute hearing,

but also gives us
the very sharp tuning.

I want to offer you a short demonstration

of something related to this.

I’ll ask the people who
are running the sound system

to turn up its sensitivity
at one specific frequency.

So just as a hair cell is tuned
to one frequency,

the amplifier will now enhance
a particular frequency in my voice.

Notice how specific tones emerge
more clearly from the background.

This is exactly what hair cells do.

Each hair cell amplifies and reports
one specific frequency

and ignores all the others.

And the whole set of hair cells,
as a group, can then report to the brain

exactly what frequencies
are present in a given sound,

and the brain can determine
what melody is being heard

or what speech is being intended.

Now, an amplifier such as
the public address system

can also cause problems.

If the amplification is turned up too far,

it goes unstable and begins to howl

or emit sounds.

And one wonders why the active process
doesn’t do the same thing.

Why don’t our ears beam out sounds?

And the answer is that they do.

In a suitably quiet environment,
70 percent of normal people

will have one or more sounds
coming out of their ears.

(Laughter)

I’ll give you an example of this.

You will hear two emissions
at high frequencies

coming from a normal human ear.

You may also be able to discern
background noise,

like the microphone’s hiss,

the gurgling of a stomach,
the heartbeat, the rustling of clothes.

(Hums, microphone hiss,
dampened taps, clothes rustling)

This is typical.

Most ears emit just a handful of tones,

but some can emit as many as 30.

Every ear is unique, so my right ear
is different from my left,

my ear is different from your ear,

but unless an ear is damaged,

it continues to emit
the same spectrum of frequencies

over a period of years or even decades.

So what’s going on?

It turns out that the ear
can control its own sensitivity,

its own amplification.

So if you’re in a very loud environment,
like a sporting event

or a musical concert,

you don’t need any amplification,

and the system is turned down all the way.

If you are in a room like this auditorium,

you might have a little bit
of amplification,

but of course the public address system
does most of the work for you.

And finally, if you go into
a really quiet room

where you can hear a pin drop,

the system is turned up
almost all the way.

But if you go into an ultraquiet room
such as a sound chamber,

the system turns itself up to 11,

it goes unstable

and it begins to emit sound.

And these emissions constitute
a really strong demonstration

of just how active the hair cell can be.

So in the last minute, I want to turn
to another question that might come up,

which is: Where do we go from here?

And I would say that there
are three issues

that I would really like
to address in the future.

The first is: What is the molecular motor

that’s responsible for
the hair cell’s amplification?

Somehow, nature has stumbled
across a system

that can oscillate or amplify
at 20,000 cycles per second,

or even more.

That’s much faster than any other
biological oscillation,

and we would like to understand
where it comes from.

The second issue is how the hair cell’s
amplification is adjusted

to deal with the acoustic circumstances.

Who turns the knob to increase
or decrease the amplification

in a quiet or in a loud environment?

And the third issue is one
that concerns all of us,

which is what we can do
about the deterioration of our hearing.

Thirty million Americans,

and more than 400 million
people worldwide,

have significant problems on a daily basis

with understanding speech
in a noisy environment

or over the telephone.

Many have even worse deficits.

Moreover, these deficits
tend to get worse with time,

because when human hair cells die,

they’re not replaced by cell division.

But we know that nonmammalian animals
can replace their cells,

and those creatures' cells are dying
and being replaced throughout life,

so the animals maintain normal hearing.

Here’s an example
from a little zebra fish.

The cell at the top
will undergo a division

to produce two new hair cells.

They dance for a little bit,

and then settle down and go to work.

So we believe that if we can decode
the molecular signals that are used

by these other animals
to regenerate their hair cells,

we’ll be able to do
the same thing for humans.

And our group and many other groups
are now engaged in research

trying to resurrect
these amazing hair cells.

Thank you for your attention.

(Applause)