►
From YouTube: Jonathan Michaels - Interview With a Neuroscientist
Description
Matt and Jonathan discuss how objects are represented in the brain, especially with respect to behavior and actions.
A
A
Systems
laboratory
in
Stanford
and
what
was
the
name
of
that
group
should
know
my
group
anyway.
I
wanted
to
talk
to
you
about
representations
of
things
in
the
brain
and
about
motor
generation
and
that
sort
of
thing,
because
you
have
a
specialty
and
motor
the
motor
cortex
so
we'll
get
into
that.
But
first,
what
is
it
about?
The
motor
system
then
treat
your
interest.
B
A
B
B
A
B
A
And
then,
finally,
strange
attractors
we'll
talk
a
little
bit
about
dynamical
systems,
but
I'd
like
to
talk
first
about
the
grasping
monkeys-
and
maybe
you
can
some
context
to
this,
because
you've
done
a
lot
of
work
with
monkeys
and
grasping
hand
over
your
career.
So
we-
maybe
you
just
talk
about-
let's
say
I'm
a
monkey
and
I'm
about
to
grasp
something.
I
see
something
I'm
going
to
grasp
it.
What
happens
in
your
brain?
What
are
the
sort
of
this
stages
of
planning
and
execution?
B
Think
it's
always
important
that
you
know
so.
Obviously,
without
you
know,
minimizing
the
visual
system
too
much
there's
a
lot
that
goes
on
just
just
in
seeing
the
scenario
in
front
of
you
sure,
seeing
the
objects
in
front
of
you
just
being
able
to
process
where
the
objects
are,
what
they
are,
whether
they
correspond
to
something
we
you
know
we've
seen
before
right,
that
is
already
gonna.
Take
up
a
little
large
parts.
I
would
argue
you
know.
Potentially
up
to
a
third
of
your
brain
is
gonna,
be
somehow
involved
in
trying
to
figure
out.
B
B
B
Yes
and
that
example
is
used
very
often
as
an
example
of
something
like
a
cup
hot
cup
of
coffee.
Let's
say
right,
you
see
steam
coming
up
from
the
coffee
and-
and
you
think,
all
right-
well
maybe
I
want
this-
doesn't
have
a
handle,
but
maybe
I
want
to
grab
the
handle
instead
of
grabbing
the
side.
Yeah.
A
A
A
B
A
A
B
And
we
know
what
it
is,
then
we
have
to
also
go
off
of
our
experience
over
time
of
how
we
know
to
interact
those
objects,
which
of
course
depends
on
the
context
right.
So
I
know
that
that
these
things
can
be
lifted
by
like
putting
my
fingers
around
the
side
and
applying
some
amount
of
pressure
right.
A
B
A
A
But
you've
sort
of
prepped
the
space
because
you've
identified
the
object.
You
have
a
representation
of
the
thing
that
you're
going
to
grasp
or
move
to.
So
whatever
movement
happens
is
already
tied
specifically
to
that
object
in
place
right,
which
is
fascinating,
honestly
I
mean
even
if
you
think
about
the
object,
is
like
a
jar
with
a
lid
on
it
and
you're,
not
just
grasping,
but.
A
A
A
B
B
B
A
A
B
A
B
A
Totally
makes
sense,
and
even
from
the
standpoint
of
HTM
theory,
because
we
think
of
objects
and
object,
representations
and
we're
starting
to
think
of
objects
and
behaviors
of
those
objects
also
being
coded
in
the
same
space.
So,
for
example,
throwing
a
ball.
You
know
a
human
throwing
a
ball,
that's
a
pattern:
a
sequence
of
patterns,
I
think
that
anybody
can
bring
up
in
their
mind.
They
can
think
of
a
human
throwing.
A
ball
I
can
think
of
me
throwing
a
ball
Youth
or
any
ball.
Anybody
throwing
a
ball.
A
B
A
Your
brain,
you
know,
you
know
what
it
feels
like
to
throw
a
ball
just
by
thinking
about
yeah.
It
reminds
me
of
something
like
Dad
I
was
just
told
me
was
when
we
were
practicing.
Baseballs
practice
makes
permanent,
not
perfect.
You
would
always
say
that,
because
you
know
because
every
time
you
repeat
something
you're
you're
instilling
in
that
neural
pathway,
you
know
that
sequence
of
motor
commands
that
makes
that
perfect,
throw
you
know,
you're
honing
it
down.
So
let's,
let's
go
to
the
next
topic,
we're
just
talking
about
reading
the
brain
a
little
bit.
A
B
A
A
B
B
A
A
A
B
Is
really
a
completely
completely
different
scale?
The
only
way
you
can
go
farther
than
that
is
by
actually
having
an
electrode
directly
on
the
membrane
of
the
cell,
which
sometimes
people
also
do
because
then
you
can
measure
the
potentials
inside
itself.
Oh
the
actual
voltage.
Yes,
you
know
outside
the
cell.
You
see
that
that
the
spikes
yes,
but
you
don't
see
the
fluctuation.
B
A
Let
me
just
try
and
break
this
out
into
something
tangible,
so
let's
say
you've
got
one
neuron
that
you're
monitoring
out
of.
However
many
you're
monitoring
and
you
have
an
animal
and
experiment
doing
some
behavior,
so
you
could
maybe
map
that
one
neuron
to
how
fast
the
animals
moving
and
see
if
this
does.
This
indicate
give
any
indication
of
how
fast
the
animals
moving.
How
fast
that
neurons
firing.
B
B
A
It's
not
like
you
can,
so
you
can
say:
okay,
there
is
some
so
there's
some
correlation
between
this
neurons
firing
rate
and
this
particular
attribute
of
a
movement.
But
you
still
don't
necessarily
know
what
that
neurons
representing
or
what
how
what
it's
responding
to
in
the
environment
is.
A
part
of
the
movement
is
a
part
of
the
sensory
effect
of
grasping
something
you
don't
really
know.
B
A
B
A
B
B
A
whole
population
of
neurons,
but
if
you
actually
have
you
know
a
whole
set
of
neurons
recorded
at
the
same
time,
what
you
actually
see
is
that
if
you
want
to
predict,
let's
say
we
talk
in
terms
of
prediction.
If
you
want
to
predict
what
the
firing
rate
of
a
neuron
is
going
to
be
at
a
particular
time
or
the
next
time
point.