00:00:09
hi and welcome to the neuroethics
00:00:11
learning collaborative I'm Martha Farah
00:00:14
I'm a cognitive neuroscientist here at
00:00:16
Penn I direct the center for
00:00:18
neuroscience and Society here and I've
00:00:20
also put together um the lectures in the
00:00:24
neuroethics learning collaborative some
00:00:26
of which I give like this one so today
00:00:29
we're going to talking about uh brain
00:00:32
computer
00:00:33
interfaces um and I'm going to begin by
00:00:36
telling you a little bit about the
00:00:37
Science and Technology of BCI and then
00:00:40
we'll move into the neuroethics the what
00:00:43
I'm calling here the societal
00:00:46
challenges so what is a brain computer
00:00:49
interface well it's pretty much what it
00:00:52
sounds like it is um a system for either
00:00:57
reading out information from the brain
00:01:00
um to be translated by a computer into
00:01:03
uh motor commands or other forms of
00:01:07
communication interacting with the
00:01:09
outside world
00:01:11
or rather than sending the information
00:01:14
out from the brain uh it's a it's an
00:01:17
interface that brings information in so
00:01:20
it's a system for bringing information
00:01:22
into the brain in a way that enables the
00:01:25
equivalent of sensation or perception
00:01:29
but what's key key is you know rather
00:01:31
than the usual ways that we either reach
00:01:34
out and act upon the world or um receive
00:01:37
information from the world with uh BCI
00:01:40
you have a a machine a computer
00:01:42
interfacing directly with the brain so
00:01:44
you're not using your peripheral nervous
00:01:47
system you're not uh um launching the
00:01:50
actions with your own muscles or getting
00:01:53
the information in through your own
00:01:55
retina or cocka or um
00:01:58
whatever so so the ways in which that uh
00:02:03
interface is implemented um can vary
00:02:06
greatly the um the kind of lowest Tech
00:02:10
easiest way to um rig up a brain
00:02:14
computer interface is using scalp
00:02:17
recorded EEG brain waves um and here you
00:02:21
can see a couple of uh systems that do
00:02:24
that one very simple system that
00:02:27
actually um can connect to your iPhone
00:02:29
phone and um Believe It or Not There are
00:02:33
apps that will um either record your EEG
00:02:37
to help you um to learn to meditate and
00:02:41
relax you know as sort of a biof
00:02:43
feedback uh mechanism um there are also
00:02:46
some uh game electronic games that use
00:02:51
um EEG uh picked up from from outside
00:02:54
the head uh to you know move players and
00:02:58
so forth in a game
00:03:00
um this is a headset uh created by a
00:03:03
company called neuros Sky um that uh you
00:03:07
know is is used for a kind of brain
00:03:09
machine interface that any of you could
00:03:12
order online I I don't know how much it
00:03:14
costs but uh um it's uh you know it's
00:03:17
readily
00:03:18
available um what you see below that is
00:03:22
um a system with many more electrodes um
00:03:26
that is able to direct a a um wheelchair
00:03:32
based on brain waves based on EEG and um
00:03:37
this is obviously you know a of a less
00:03:41
uh um frivolous use of um scalp recorded
00:03:46
EEG for brain machine interface um and
00:03:49
the idea is that someone who is
00:03:51
paralyzed could um could direct a
00:03:54
wheelchair like this experimental system
00:03:57
um demonstrates using brain activity
00:04:01
alone now the problem with these kinds
00:04:04
of systems that um severely limits their
00:04:08
usefulness is as as we know um from our
00:04:12
Neuroscience courses um EEG is a very
00:04:16
sort of uh very limited and kind of coar
00:04:20
grained measure of brain activity um and
00:04:23
really only picks up a subset of brain
00:04:26
activity um it has very poor localizing
00:04:29
ability because of um all the you know
00:04:34
the distance of the electrode from the
00:04:37
um generators of uh the activity and all
00:04:42
the volume conduction um that uh that
00:04:45
signal under goes going from the neurons
00:04:47
to the the electrode um and also um EEG
00:04:52
is only tapping
00:04:55
into the activity of um large numbers of
00:04:59
synchron IED neurons um that also happen
00:05:02
to be orientated in the right direction
00:05:04
and so forth so um so it's very limited
00:05:10
um you can get a more localized kind of
00:05:15
better differentiated system from um
00:05:21
electrocorticography uh abbreviated
00:05:24
EOG where they place the electrodes
00:05:28
right on the surface of the brain um
00:05:31
which obviously requires opening up the
00:05:33
head so it's you know more invasive um
00:05:37
but the the closer location um uh
00:05:43
enables um much more differentiated
00:05:46
signals um so you can learn um you know
00:05:50
more about uh sort of localized patterns
00:05:54
of brain activity certainly more
00:05:57
localized relative to what you get out
00:05:59
outside of the head like this um and
00:06:02
this kind of uh process is already done
00:06:05
um routinely uh for some cases of um uh
00:06:10
surgery for epilepsy before they um go
00:06:13
in and do the surgery uh they want to
00:06:15
map very precisely um the uh the
00:06:18
seizures and the brain activity
00:06:20
associated with important cognitive
00:06:22
functions that they don't want to
00:06:23
disrupt um so uh this is something that
00:06:27
is um while invasive um it is already
00:06:31
done routinely for other
00:06:33
reasons finally if you want to really
00:06:36
get in there and um get the most fine
00:06:40
grained um uh local and well
00:06:44
differentiated signals from neurons in
00:06:47
the cortex you need to put the
00:06:49
electrodes in the cortex and that is
00:06:52
what has been done in various kinds of
00:06:54
pre-clinical research with BCI so um
00:06:58
what you see up here is um a monkey who
00:07:03
has um electrodes in his motor cortex
00:07:06
and he's using them to direct a robotic
00:07:10
arm uh and um we'll talk more about uh
00:07:14
that monkey and you'll see a video of
00:07:16
him uh
00:07:17
momentarily um and finally uh there have
00:07:21
been a small number of humans who've
00:07:22
been implanted with
00:07:24
intracortical uh recording electrodes um
00:07:27
to uh to operate great
00:07:31
um robotic uh arms and computer cursors
00:07:36
and so forth um just by thought
00:07:42
alone
00:07:43
so continuing with just the basic
00:07:46
question of you know what what is the
00:07:47
brain computer interface what what do
00:07:49
you do with them well um there are two
00:07:52
main applications um the the first class
00:07:55
of applications is to bring information
00:07:59
into about the world um and that has
00:08:01
been done with um clear implants which
00:08:05
uh substitute for um uh um the sort of
00:08:11
uh
00:08:13
normal uh mechanism by which sound is
00:08:16
transduced um into um neural impulses uh
00:08:21
and artificial retinas which are still
00:08:24
experimental but do a similar thing with
00:08:27
uh Vision they're also used in um uh
00:08:31
enabling movement um and communication
00:08:34
for patients who can't move because of
00:08:38
paralysis uh you know spinal cord injury
00:08:41
or because of locked in
00:08:43
syndrome and what we'll see in the next
00:08:46
few slides is just examples of recent
00:08:49
work using BCI and these in these kinds
00:08:51
of um
00:08:53
applications uh that um focus on work
00:08:57
with the implanted electrodes um
00:09:00
focusing on the implanted electrodes
00:09:01
partly because um it's where you can do
00:09:04
the most you can get the best signals in
00:09:07
and out um and uh uh so um you know
00:09:12
ultimately you can solve more problems
00:09:15
and uh you know achieve better outcomes
00:09:18
for the patients um and also because
00:09:21
obviously that's a bigger deal ethically
00:09:23
and so the neuroethical issues you know
00:09:25
really come up when you're talking about
00:09:27
this very invasive procedure of sticking
00:09:29
Hardware in somebody's
00:09:30
head the first type of uh BCI that uh I
00:09:34
want to um me mention to you um is clear
00:09:39
implants um these are actually in use
00:09:43
now in fact um as of spring of 2009
00:09:48
there were
00:09:50
188,000 people in the world um with
00:09:53
clear implants um a clear implant uh
00:09:58
consists of a microphone and a signal
00:10:01
processor that's essentially the the
00:10:04
machine the the computer um that uh
00:10:09
interfaces via implanted electrodes with
00:10:13
the auditory nerve so this is a way to
00:10:17
um you know essentially restore hearing
00:10:20
not not to normal levels um but good
00:10:23
enough that you can um understand speech
00:10:27
even listen to music um this person here
00:10:30
is Michael chist he wrote um a wonderful
00:10:34
book about his own experience um as the
00:10:37
recipient of bilateral clear implants
00:10:40
called rebuilt how becoming part
00:10:43
computer made me more
00:10:45
human and um his hearing is good enough
00:10:49
that the very first time I ever
00:10:50
interacted with him it was via a phone
00:10:53
call and I really had no idea what to
00:10:56
expect if he was going to be able to
00:10:58
understand me or not and
00:11:00
I never would have known that uh I was
00:11:02
talking to somebody with a hearing
00:11:03
problem so um clear implants have been
00:11:06
wonderfully successful um in fact
00:11:10
successful enough that some within the
00:11:13
deaf Community fear that it is going to
00:11:16
spell you know the end of Deaf culture
00:11:19
um that uh there will be no more uh
00:11:22
people who are deaf and therefore kind
00:11:26
of uh invested in ASL um and all of the
00:11:31
the art and literature and culture that
00:11:33
goes with
00:11:36
it um there are also bcis for vision um
00:11:41
several different companies are
00:11:43
developing uh artificial retinas um some
00:11:46
of them uh most of them connect to the
00:11:49
optic nerve in the eye but some actually
00:11:52
directly stimulate visual cortex and
00:11:56
although these are um still experimental
00:11:59
they have certainly enabled people to
00:12:01
have some degree of form Vision um to
00:12:05
actually see which way the Big E on the
00:12:07
vision chart is pointing um and so
00:12:12
forth for movement um the research is uh
00:12:17
you know mostly what's called
00:12:20
preclinical um that is you know work
00:12:23
with animals to try to understand um the
00:12:26
general approach that might work and be
00:12:28
useful
00:12:29
um but
00:12:32
um uh
00:12:34
the research has been um advancing
00:12:38
quickly um in the last um 10 years um I
00:12:42
can tell you some of the achievements
00:12:45
they've been quite impressive um so uh
00:12:48
one one of the really um exciting
00:12:51
developments that I think you know made
00:12:53
the daily newspapers not not just
00:12:55
science journals was um out of U Miguel
00:13:01
ncis's Lab at Duke where um he had a
00:13:06
setup like this where he taught a monkey
00:13:09
to um move a cursor around on a screen
00:13:13
using a joystick meanwhile um recording
00:13:17
from electrodes implanted in the
00:13:19
monkey's motor
00:13:21
cortex and um unbeknownst to the monkey
00:13:25
um his uh actions were also being um or
00:13:30
the sorry the brain activity
00:13:32
corresponding to his actions were also
00:13:33
being relayed to a robotic arm um out of
00:13:38
sight of the monkey um and after a
00:13:43
while they the researchers disconnected
00:13:47
the
00:13:48
joystick from the computer okay so that
00:13:52
basically what was making the uh cursor
00:13:56
move around was the monkeys brain
00:14:00
activity very interestingly
00:14:04
um the at least two monkeys that they
00:14:08
tested this on um after a while figured
00:14:11
out that they did not need to be using
00:14:14
the joystick and just stopped stopped
00:14:18
touching it all together and sat there
00:14:21
just spontaneously um moving moving the
00:14:25
cursor using their brain waves Alone um
00:14:29
and of course in addition to moving the
00:14:31
cursor they were also manipulating the
00:14:33
robotic arm in a remote um
00:14:38
location um another oops there's another
00:14:43
um study that I'm going to show you a
00:14:44
video of where um uh Andrew Schwarz and
00:14:48
his colleagues at Pittsburgh um
00:14:51
implanted a monkey in a very similar way
00:14:53
motor cortex um this time they let the
00:14:56
monkey see the robotic arm and um get
00:15:00
experience um using it driving it with
00:15:04
their brain waves and found that the
00:15:06
monkey began to treat the arm very much
00:15:10
as if it was you know one of his own
00:15:11
appendages and could use it to um to
00:15:15
feed itself treats so let
00:15:18
me go to
00:15:23
this video you'll see the monkey um
00:15:26
feeding himself with a robotic arm that
00:15:29
he's controlling by his brain
00:15:46
waves so the experimentor is moving
00:15:48
around where the marshmallow is and uh
00:15:51
oh didn't quite get it there we
00:15:53
go there's one point on this video where
00:15:56
there's a little marshmallow goo left on
00:15:59
the robotic hand and the monkey licks it
00:16:03
I mean it really uh gives you the sense
00:16:04
that this monkey has you know
00:16:07
Incorporated this robotic arm into his
00:16:09
own uh body schema his own sense of
00:16:19
self okay well I think that's probably
00:16:23
enough of this monkey um there is also
00:16:27
research there also so whoops clinical
00:16:31
trials
00:16:33
um with
00:16:34
humans using the same kind of system so
00:16:39
in in this case um you're going to see
00:16:43
work done by a uh company called brain
00:16:46
gate that um has
00:16:49
implanted uh you know a chip with many
00:16:53
electrodes um into again motor cortex
00:16:57
and um
00:17:00
allowed uh paralyzed patients to learn
00:17:04
to um control
00:17:07
their brain activity and motor cortex um
00:17:11
essentially by thinking about making the
00:17:13
movements um to operate uh um external
00:17:18
devices and let me show you a
00:17:21
uh a video of that Kathy Hutchinson is
00:17:25
among the first humans to have her brain
00:17:28
directly wired to a
00:17:30
computer years ago Kathy suffered a
00:17:33
stroke that left her mentally sharp but
00:17:35
trapped inside a paralyzed body and
00:17:37
unable to speak locked in like Scott
00:17:40
mackler 3 years ago Kathy volunteered to
00:17:44
have the same kind of sensors we saw in
00:17:46
the monkeys implanted in her motor
00:17:48
cortex which controls movement and is
00:17:51
located right on the surface of the
00:17:53
brain the sensors connect to the
00:17:55
computer through this plug on her head
00:17:58
the system is called brain G and it was
00:18:00
created by a team led by Brown
00:18:02
University neuroscientist John Don if
00:18:05
you look at this Square each one of
00:18:08
these little black boxes is the
00:18:10
electrical signal coming from one
00:18:12
electrode in the brain and each one of
00:18:13
those is a neuron fire right it's its
00:18:16
electrical potential it lets out a
00:18:17
1,000th of a second pulse how well do we
00:18:21
understand this language we have a
00:18:24
somewhat of an understanding we know
00:18:25
that there's a general pattern of for
00:18:27
example left right up down even fast or
00:18:29
slow scottt Kathy now has neural control
00:18:32
over that person Dr Lee hawber of
00:18:35
Massachusetts General Hospital is
00:18:37
leading the clinical trial we watched
00:18:39
together as Kathy showed us what she can
00:18:42
do there's a handful of icons that have
00:18:44
been placed on the screen here's Google
00:18:46
The New York Times Boston Globe and
00:18:48
here's Mass General Hospital struggle
00:18:49
terms
00:18:51
y we're seeing Kathy moving this cursor
00:18:56
with nothing but her mind that's right
00:18:58
she's thinking about the movement of her
00:19:00
hand and uh she's moving the cursor much
00:19:02
much as if she had her hand on a mouse
00:19:05
so if a patient who's paralyzed thinks
00:19:09
move my left arm the brain fires those
00:19:12
neurons yes even though the arm does not
00:19:14
move yes it's very surprising it fires
00:19:16
even though you're not moving curs is
00:19:18
still a little bit wavier some moving
00:19:20
the cursor with her mind is not as fluid
00:19:23
or direct as using a mouse while we were
00:19:25
there the cursor meandered a bit
00:19:27
sometimes overshock
00:19:29
but Kathy always hit her Target in the
00:19:31
end you want to uh play some
00:19:34
music all right she'll click on it
00:19:37
imagine squeezing her hand which just
00:19:39
the or doing something else with the
00:19:41
click and she just clicked play yep she
00:19:49
did that's pretty
00:19:52
amazing and so if Kathy can control a
00:19:56
cursor she canol control anything a
00:20:00
computer is connected to that's the goal
00:20:03
the lights the temperature in the room
00:20:05
even even a wheelchair at some point
00:20:08
ready to try it for real in fact Kathy
00:20:10
has already driven a wheelchair see if
00:20:12
we can drive it right over to the door
00:20:14
they haven't let her write in it yet for
00:20:16
her own safety but with monkeys adopting
00:20:19
robot arms and a completely paralyzed
00:20:22
person driving a chair imagine where
00:20:25
this could be headed fantastic very good
00:20:33
okay so clearly in the last uh 10 years
00:20:36
there's been tremendous progress um in
00:20:40
uh interfacing brain activity to outside
00:20:46
Machinery um what are some of the
00:20:50
current um remaining barriers to be
00:20:53
overcome scientifically and
00:20:55
technically well you know I should say I
00:20:58
do not work in this area but um I
00:21:01
attended a meeting last fall um put on
00:21:04
by the New York Academy of Sciences and
00:21:06
the Aspen brain Institute there was kind
00:21:08
of a summit meeting of all the leading
00:21:10
researchers including the people that
00:21:12
have been mentioned so far um in brain
00:21:15
machine interfaces and one of the
00:21:17
interesting things about this meeting is
00:21:19
people were talking about you know what
00:21:20
are the outstanding problems that need
00:21:22
to be solved and what are some promising
00:21:25
Solutions um that people are working on
00:21:28
and I would say that um the two uh of
00:21:32
the two biggies the first is just
00:21:35
maintaining the connection between the
00:21:38
machinery and the Brain um the brain
00:21:42
like every other part of the body reacts
00:21:45
to um having a foreign body uh lodged in
00:21:48
it um there are all kinds of um you know
00:21:52
immune responses inflammation scarring
00:21:55
and so forth that um take what starts
00:21:59
out as a a good useful contact and um
00:22:04
and makes it over time uh convey much
00:22:08
less useful information so um people are
00:22:12
trying all kinds of uh you know end runs
00:22:16
around this all kinds of ways of solving
00:22:18
the problem you know the end runs
00:22:21
including um uh methods that don't
00:22:24
involve um electrodes at all but um
00:22:28
involve of uh Optical methods um and uh
00:22:32
and chemical
00:22:34
interfaces um there are also attempts to
00:22:36
design electrodes that um will sort of
00:22:39
encourage the the neurop pill to just
00:22:43
grow right into the electrode um and uh
00:22:46
you know rather than um sort of
00:22:49
encapsulate it um so much um varied and
00:22:55
you know interesting and very promising
00:22:57
work um along those lines the other big
00:23:01
challenge is um just figuring out how to
00:23:06
translate brain activity into um you
00:23:10
know commands to the outside world and
00:23:13
or um uh design signals that can be put
00:23:17
into the brain in a way that um uh is is
00:23:21
useful that the rest of the brain can
00:23:23
make use out of um and there's a lot of
00:23:26
work going on here this is you know
00:23:28
really the stuff of much basic science
00:23:31
research as well as you know research
00:23:33
aimed at brain machine interfaces um in
00:23:36
fact
00:23:38
um one of the big uh you know impetuses
00:23:42
to to start doing these um motor systems
00:23:47
um was work of
00:23:49
georgopoulos just you know basic science
00:23:52
research um looking at population coding
00:23:55
in uh motor cortex looking at the way um
00:23:58
information available from you know
00:24:01
neurons sampled across you know
00:24:04
electrode arrays um uh corresponded with
00:24:09
the monkey's intention to move his eyes
00:24:11
somewhere in space or reach somewhere in
00:24:13
space and um so that you know that work
00:24:17
goes on it's um uh it's of interest in
00:24:20
its own right scientifically and it's
00:24:23
one of the things that has to be better
00:24:25
understood um to create um
00:24:29
uh you know interfaces that um you know
00:24:33
enable people to do sort of more complex
00:24:35
and highly differentiated things rather
00:24:37
than just you know get a cursor to
00:24:40
slowly drift to one side or the other um
00:24:43
so sort of bandwidth being able to you
00:24:45
know get that information out quickly
00:24:47
and
00:24:50
precision so having given you a little
00:24:54
um summary of uh state-of-the-art in
00:24:57
brain machine interfaces um brain
00:25:00
computer interfaces now I want to go to
00:25:04
the neuroethics of the
00:25:07
area there are a tremendous variety of
00:25:10
ethical legal and social issues what we
00:25:14
call the LC's El
00:25:15
LSI um and uh you know very very
00:25:19
different in kind you know run running
00:25:21
the gamut from you know transhumanist
00:25:24
cyborgs to um uh you know in
00:25:29
intellectual
00:25:32
property problems you know how do you
00:25:34
encourage Innovation um encourage uh
00:25:38
collaboration and so forth um in terms
00:25:40
of the financial Stakes um so what I
00:25:44
want to do is try to subdivide these
00:25:46
different issues um to um just kind of
00:25:51
uh lay them out a little more
00:25:53
systematically and make them less like
00:25:54
just this mountain of um very
00:25:59
heterogeneous
00:26:00
issues um I'm going to subdivide them in
00:26:04
terms of time frame and uh
00:26:08
the the sort of boundary points between
00:26:13
immediate medium-term and long-term are
00:26:17
you know just my probably not terribly
00:26:20
educated uh guesses um but I think um
00:26:25
it's it's worth distinguishing between
00:26:27
the immediate the medium and the
00:26:28
long-term challenges and this is my you
00:26:32
know best guess at uh how that maps on
00:26:36
to a calendar um but you will see that
00:26:40
um the salt shakers are a reminder that
00:26:44
this you know is pure conjecture um so
00:26:47
take the actual numbers there with a
00:26:50
grain of salt um but whether the numbers
00:26:53
are um you know optimistic it's going to
00:26:57
actually take one longer pessimistic
00:26:59
it'll be here sooner um however the
00:27:02
numbers are off I do think that these
00:27:05
issues um do apply um for each of these
00:27:08
three
00:27:09
categories so let's start with the
00:27:12
long-term issues you know these are the
00:27:15
sort of big you know scary sexy
00:27:18
depending on your you know perspective
00:27:20
um issues of how Humanity itself may be
00:27:24
changed by merging with machines
00:27:28
um some people find this idea very scary
00:27:32
um and uh think it should be stopped um
00:27:36
Francis fukiyama um author of our
00:27:40
posthuman future um wrote a journal
00:27:43
article um a few years after posthuman
00:27:46
future was
00:27:47
published um called you know
00:27:50
transhumanism the world's most dangerous
00:27:52
idea um the idea is the idea of
00:27:55
transhumanism that fukiyama finds so
00:27:58
dangerous um is that as we
00:28:02
begin um adding to our capabilities and
00:28:07
changing the way our bodies and minds
00:28:10
work um using
00:28:12
technology um we will
00:28:15
eventually change ourselves so much that
00:28:20
we won't be human beings anymore that we
00:28:23
just will not be recognizable as members
00:28:26
of the species that we all are now
00:28:28
um and this is the idea of transhumanism
00:28:31
that we transcend um the limitations of
00:28:34
being a human being um we could be you
00:28:37
know smarter and have amazing memories
00:28:39
and be able to perceive you know
00:28:42
infrared and you know uh on and on it's
00:28:46
kind of Science Fiction like but I want
00:28:49
to say that there's
00:28:52
really no
00:28:55
reason that I have heard that suggests
00:28:59
that we couldn't do it or even that
00:29:02
suggests that we wouldn't do it um so
00:29:07
you know it's it's a long way off uh in
00:29:10
my
00:29:11
little uh timeline I'm saying you know
00:29:14
30 or more years away um but uh it's
00:29:19
it's definitely already on people's
00:29:20
minds um and some people think it's a
00:29:23
terrible thing now some people think
00:29:25
it's a Dandy thing they think being a
00:29:29
human being is all about
00:29:32
self-improvement um uh making over the
00:29:35
world and even making over yourself uh
00:29:37
to better suit your aims and goals and
00:29:41
desires and um what used to be called
00:29:44
the world transhumanist Association and
00:29:47
is now goes by the name Humanity plus um
00:29:51
is a whole International network of
00:29:54
people um who are interested in the idea
00:29:57
of transhumanism and who look forward to
00:30:01
it so what are some of the um ethical
00:30:04
issues that we will encounter um in the
00:30:08
long term uh when we begin to you know
00:30:11
link up with computers well one person
00:30:15
who has discussed this is a computer
00:30:18
science Professor um in England uh at
00:30:22
the University of reading named Kevin
00:30:25
Warwick um he wrote a an article
00:30:28
um on sort of ethical issues in uh you
00:30:32
know brain computer interfaces in 2003
00:30:35
in the journal ethics and information
00:30:37
technology and here's here's some of his
00:30:40
thoughts this is what we might be able
00:30:43
to do and he's saying like this is a
00:30:45
good thing um if uh um people could link
00:30:50
their brains directly to computers Oh
00:30:53
and before I before we go through this
00:30:55
list of um things that we could do um
00:30:58
once we're hooked up like that I should
00:31:00
tell you that he is such an Enthusiast
00:31:04
for brain machine interfaces and for you
00:31:07
know sort of transhumanism that he has
00:31:10
already um been implanted with various
00:31:14
um gadgets um including an electrode
00:31:17
array in his arm that you see here which
00:31:20
picks up um impulses from his uh from
00:31:25
from the nerves in his arm and um um
00:31:29
enable him to you know open doors and
00:31:32
turn lights on and off and so forth um
00:31:35
uh using just uh you know um peripheral
00:31:40
nervous activity so um so this man um
00:31:44
has put his money where his mouth is he
00:31:47
um has already gone cyborg um and uh you
00:31:51
know eagerly awaits the time
00:31:54
when the the chip will be in the brain
00:31:57
not the
00:31:58
and we can do things such as um use the
00:32:01
computer for Rapid maths I guess he's
00:32:04
English so he puts an s on the end of
00:32:06
that word um but you know the idea is
00:32:09
that um you don't need to like get out
00:32:11
the calculator and use your finger to
00:32:13
punch numbers in um you can just sort of
00:32:16
think you know the uh whatever long
00:32:19
complicated expression you want to get
00:32:22
the uh value of and the computer will um
00:32:28
read that thought and then send you the
00:32:31
answer um You can call on you know the
00:32:35
entire base of knowledge in the uh
00:32:38
worldwide web um you know rapidly and
00:32:41
effortlessly by thinking um you could
00:32:45
have
00:32:46
memories of experiences that you didn't
00:32:49
experience yourself so you could upload
00:32:51
somebody else's memory into your own
00:32:53
brain um you could sense the world in a
00:32:57
plethora of way ways um you know
00:33:00
ultrasound uh you know
00:33:03
infrared um you know what whatever
00:33:07
physical signals there are out there you
00:33:09
know polarized light whatever you could
00:33:11
um you could have peripheral devices
00:33:13
that pick those up and feed them to your
00:33:15
brain um you could understand
00:33:19
multi-dimensionality perhaps in a way
00:33:21
that we can't do without um you know
00:33:24
with our limited working memory
00:33:26
capacities and spatial ability um and
00:33:30
you could communicate by thought signals
00:33:32
alone brain to brain and I I do believe
00:33:35
his wife um also has a chip and you know
00:33:39
the idea is that someday they'll be able
00:33:40
to you know send um what do they call it
00:33:45
you know sweet little nothing to each
00:33:47
other um not whisper sweet little
00:33:49
nothings but um uh twitch
00:33:53
them by the arm
00:33:56
Okay so
00:33:59
um what are the um ethical implications
00:34:05
of this
00:34:08
um well he says that
00:34:13
um basically cyborgs may become so
00:34:17
capable
00:34:18
so uh intellectually capable that they
00:34:22
would find anything that an
00:34:26
unimplanted unenhanced ened human has to
00:34:29
say sort of
00:34:30
trivial um they would care about their
00:34:34
own kind you know much as we care about
00:34:36
our kind and um you
00:34:39
know uh humans may not figure too
00:34:42
highly uh in their View and
00:34:48
um then in terms of our treatment of
00:34:51
them uh could we could we expect them or
00:34:55
could we expect ourselves if if if we
00:34:58
upgrade um to uh to sort of you know
00:35:02
merge in effect and just become nodes in
00:35:04
some super Computing
00:35:08
Network I have to say that you know the
00:35:13
um by and large the the uses that
00:35:17
Warwick envisions in this article um
00:35:21
don't seem all that compelling I mean I
00:35:24
for one would not want you know brain
00:35:28
surgery um just to not have to put my
00:35:32
fingers on a
00:35:34
calculator um and you know likewise
00:35:38
searching the internet with go Google is
00:35:40
fine um I don't I don't mind having to
00:35:43
um uh do it outside my head um but I do
00:35:48
think that as we approach the the point
00:35:52
of you know 30 years from now or
00:35:54
whatever um we'll have more more
00:35:58
experience with the things that we can
00:36:00
do um and the things that work well and
00:36:03
the things that people like having tried
00:36:05
them um and I think we will discover
00:36:09
other kinds of attractive options um
00:36:12
that uh you know brain computer
00:36:15
interfaces um might offer us and I do
00:36:19
think that you know as with any kind of
00:36:21
permanent modification of a person
00:36:23
whether it's by genetic engineering or
00:36:26
by um
00:36:29
gadgetry um if the um you know if the
00:36:34
result
00:36:35
is uh individuals who are just you know
00:36:40
far um far smarter far more
00:36:44
capable um far more you know who knows
00:36:47
maybe far more ethical and caring I mean
00:36:50
you know there's a lots of different
00:36:52
ways a brain could be upgraded um I I
00:36:56
think um um I do somewhat resonate with
00:37:00
um you know with this concern like what
00:37:02
would happen to the people who don't
00:37:04
enhance um you know He suggests that um
00:37:08
these cyborgs of the future may regard
00:37:11
human beings kind of the way human
00:37:13
beings regard chimps right now and you
00:37:16
know some people are nice to chimps but
00:37:20
some chimps you
00:37:23
know suffer at the hands of humans and
00:37:27
um you know it's it's very hard to know
00:37:29
um what how that would go um if some of
00:37:33
our uh species decide to upgrade and and
00:37:37
others don't so this is I think you
00:37:40
know really long range reasoning about
00:37:43
what the ethical issues are um you know
00:37:46
it's very hard to not sound a little
00:37:48
silly and a little fanciful um and a
00:37:51
little like you're borrowing trouble
00:37:53
like you know you're afraid of being
00:37:54
oppressed by you know people with
00:37:56
computers in their heads you know aren't
00:37:57
there better things to worry about in
00:37:59
this day and age um and I think that's
00:38:01
all very true um and I think we just
00:38:03
have to recognize that we are talking
00:38:05
very long range here and maybe we can
00:38:08
more fruitfully revisit these issues uh
00:38:12
when some of the nearer term issues have
00:38:14
been um lived through you know
00:38:17
experienced um and learned
00:38:19
from so what about the medium term this
00:38:22
is again on my you know rough uh
00:38:26
guesstimate timeline
00:38:28
between about 10 years from now and 30
00:38:30
years from now this what I mean by
00:38:33
medium term is um when the Technologies
00:38:37
are routinely used okay so there are
00:38:40
some that
00:38:42
are used now in clinical practice copar
00:38:46
implants there are others that are um
00:38:48
getting there but what about um when you
00:38:53
know sizable numbers of our Brethren um
00:38:57
have brain implants that are either
00:38:59
bringing information in or taking
00:39:01
information out what will be the um what
00:39:04
will be the ethical challenges um at
00:39:07
that
00:39:09
point
00:39:13
and the answer that I have
00:39:23
is
00:39:26
okay when issue is going to be you know
00:39:28
more of a traditional bioethical issue
00:39:33
who's going to have access to brain
00:39:35
computer interfaces to therapeutic brain
00:39:37
computer interfaces okay we're again
00:39:39
we're not talking about you know making
00:39:41
super brains um we're talking about
00:39:44
helping paralyzed people to move through
00:39:46
the world we're talking about helping
00:39:48
blind people to see um these are
00:39:52
expensive Technologies and you know of
00:39:55
course price comes down as volume goes
00:39:57
up and all that but you know they um
00:40:00
they're expensive to implant and they
00:40:04
require maintenance it isn't just sort
00:40:06
of pop it in and you're good to go um so
00:40:11
you know compared to the cost of other
00:40:12
kinds of Health Care you know is this um
00:40:15
you know is this going to be viewed as
00:40:17
something that you know your health plan
00:40:20
should cover um uh you know does does
00:40:26
does the um
00:40:27
does is the expense justifiable um for
00:40:30
the help that it gives the individuals
00:40:31
who are helped um and uh you know how
00:40:36
how will how will we manage that as a
00:40:39
society um getting uh getting them to
00:40:42
the people who need
00:40:44
them uh
00:40:47
okay another ethical issue that I think
00:40:51
will arise once you know some of these
00:40:54
systems are actually out there in
00:40:56
clinical use is control control of the
00:40:59
inputs and the outputs um how much
00:41:02
control should patients have um you know
00:41:07
now um you can turn off your hearing aid
00:41:10
if you're you know sick of hearing your
00:41:14
old spouse Yammer on you know or um uh
00:41:19
you know you can um I don't know you
00:41:21
could you know
00:41:25
but the the question is um um should
00:41:28
patients be able to you know vary
00:41:32
um uh you know
00:41:35
the the range of um uh you know sensory
00:41:40
signals that they're getting um the
00:41:42
kinds of uh the ways they want the
00:41:46
output of their motor system to be used
00:41:48
um to what extent uh will um the medical
00:41:53
establishment sort of say yeah this is
00:41:55
something we're giving you it's to
00:41:57
enable you and you know you're in the
00:41:58
driver's seat you you make these
00:42:01
decisions um what about involuntary
00:42:05
treatment um you know certainly for some
00:42:09
psychiatric conditions um that might uh
00:42:12
benefit from um uh you know some sort of
00:42:16
a stimulation protocol that might be
00:42:19
initiated by a computer when certain
00:42:22
patterns of brain activity are detected
00:42:26
um um uh you could imagine um
00:42:31
uh you know involuntary treatment being
00:42:34
imposed with bcis the same way it's
00:42:38
imposed with drugs for the safety of the
00:42:41
person or Society um but unlike drugs
00:42:46
that you know basically wash out of your
00:42:48
system as soon as you stop taking them
00:42:51
you know this is this is Major brain
00:42:53
surgery that leaves you know a device in
00:42:55
your head um um so you know in the
00:42:59
medium term again a fairly
00:43:01
straightforward bioethical issue but one
00:43:03
that will I think arise once um once we
00:43:05
have uh um bcis in common use um finally
00:43:11
concerning control of inputs and outputs
00:43:14
um what to do about hackers so you know
00:43:17
people
00:43:19
already for for motivations that are you
00:43:23
know unclear to me you know write
00:43:26
computer viruses and send them around
00:43:27
the world um infecting people's
00:43:31
computers their cell phones um we have
00:43:35
people who you know for you know to to
00:43:38
to steal money or to just just out of
00:43:41
curiosity you know hack into secure
00:43:43
systems of banks or the Pentagon or
00:43:46
whatever um what if one of these hackers
00:43:49
wants to hack into your brain right um I
00:43:52
mean there are there are issues um not
00:43:54
you know not insurmountable issues like
00:43:56
all of these issues is just you know
00:43:58
there are precedents We have dealt with
00:44:02
um analogous kinds of problems before
00:44:05
but those problems I think will have to
00:44:07
be addressed here um because uh you
00:44:10
really don't want you know hackers
00:44:13
hacking into your
00:44:16
brain um another application uh sorry
00:44:20
another um medium-term ethical issue
00:44:24
once we have these uh being developed is
00:44:27
um which applications will be developed
00:44:31
so um you know we've already seen with
00:44:34
the iPhone you know on the outside of
00:44:36
the head um interface uh you know people
00:44:40
are designing gaming systems because you
00:44:44
can sell them and you don't have to
00:44:46
worry about lawsuits with you know
00:44:48
somebody being harmed um uh will um you
00:44:54
know games games might be uh an
00:44:57
economically attractive area um uh
00:45:00
certainly um treatments for very common
00:45:04
conditions are attractive for the device
00:45:07
manufacturers because those are
00:45:09
conditions that you know um they can
00:45:13
sell a lot of the product to but what
00:45:16
about you know orphan conditions you
00:45:19
know what about diseases that affect
00:45:22
very few people what about um diseases
00:45:25
that affect you know mainly poor people
00:45:27
third world people um where they aren't
00:45:30
going to be able to sell much are we
00:45:32
going to um uh are we going to ignore
00:45:36
those needs um again you know nothing
00:45:40
stunningly novel but I think um these
00:45:43
are the things that will arise with BCI
00:45:46
once um once it becomes established as a
00:45:50
treatment
00:45:52
modality um and finally um you know what
00:45:57
I'm referring to is the yuck factor um
00:46:01
if if
00:46:02
BCI um can be uh used um safely uh you
00:46:08
know without a lot of complications then
00:46:10
it will be considered for um
00:46:14
increasingly nonserious non- dire
00:46:18
conditions um will people will people uh
00:46:22
Embrace that or will they um will they
00:46:25
just find find it Unthinkable to um have
00:46:29
something you know put in their brain
00:46:31
for something that isn't you know I'm
00:46:34
locked in I need some way to communicate
00:46:36
with the outside world
00:46:38
um okay and then finally Frank
00:46:41
enhancement um will
00:46:45
um you know will
00:46:49
people contrary to the last point will
00:46:52
people actually seek out um uh enhan
00:46:57
ments based on brain machine interfaces
00:47:01
um maybe you're a patient who already
00:47:03
has to be implanted with an interface
00:47:06
device um let's say you know for vision
00:47:10
um but you say well gee as long as
00:47:12
you're you know rigging up this
00:47:14
artificial retina for me can you um let
00:47:17
me see you know infrared too um uh could
00:47:24
be useful for various uh
00:47:27
reasons I mean I'm not sure why somebody
00:47:29
would want to see infrared but but the
00:47:31
point is once
00:47:35
um once you have to be implanted for a
00:47:37
therapeutic reason anyway there may be
00:47:39
ways that your implant can extend your
00:47:43
capabilities beyond what um you know it
00:47:46
would what what it's intended to restore
00:47:49
um so maybe like motor uh motor function
00:47:52
you know could you um you know if we get
00:47:55
to the point where there are you know
00:47:56
sort of robotic exoskeletons that can um
00:48:00
you know not just let people move
00:48:02
cursors on screens with um with brain
00:48:05
machine interfaces but actually get up
00:48:07
and move around um you know should the
00:48:10
should the exoskeleton be you know
00:48:12
designed in a way that the person can
00:48:14
you know leap 10 foot high barriers and
00:48:18
be like a superhero
00:48:21
um uh that would be that would be an
00:48:23
enhancement that um seems like once
00:48:28
you're doing the therapy the enhancement
00:48:30
is um really just a short hop from there
00:48:33
but I think it has ethical
00:48:37
implications so enhancement is an LC
00:48:40
topic unto itself um and we're not going
00:48:42
to go into that um
00:48:46
tonight short-term issues um there's not
00:48:50
a lot to say here except that um in the
00:48:52
short term in the next 10 years I think
00:48:55
the way this field is going to develop
00:48:57
will depend a lot on various intertwined
00:49:01
um issues that come down to money so um
00:49:06
the issue of funding who's who's paying
00:49:08
for these systems to be developed um you
00:49:11
know what's the balance of private and
00:49:15
public
00:49:16
funding conflict of interest um you know
00:49:20
uh who who should
00:49:22
be uh you know who should be involved in
00:49:25
doing the research
00:49:27
um what what should their relationship
00:49:29
be with the private companies um how
00:49:32
about regulation um should The
00:49:34
Regulators be people who are in the
00:49:36
business and so forth um intellectual
00:49:39
property law um is another biggie that
00:49:42
has been um uh you know a contentious
00:49:47
issue in far the pharmaceutical industry
00:49:50
with development of new drugs um
00:49:53
depending on how the laws are written or
00:49:55
changed um it can have the effect of
00:49:58
encouraging Innovation or um sort of
00:50:01
taking the uh um incentive away or
00:50:06
taking the incentive away to collaborate
00:50:09
with um people who might otherwise be
00:50:12
considered
00:50:14
Rivals um regulation both of the
00:50:18
practice of using brain computer
00:50:20
interfaces once they're you know up and
00:50:22
available but also of clinical
00:50:25
trials
00:50:28
um the people at the conference that I
00:50:31
went to last fall um were complaining
00:50:35
they were all complaining about the
00:50:38
difficulty of um getting uh
00:50:42
investigative device um permits to do
00:50:46
clinical trials in this country um I
00:50:50
felt kind of sorry for the speaker from
00:50:51
the FDA she was she was um hammered um
00:50:57
so apparently uh clinical trial
00:50:59
regulation is at least perceived by the
00:51:01
people doing the research as um a real
00:51:04
barrier right
00:51:05
now so going back to the sort of scheme
00:51:10
of like three uh phases and not you know
00:51:13
not emphasizing the number of years uh
00:51:16
that I hung on them really but just sort
00:51:19
of the idea that we have these Here and
00:51:21
Now problems um and how we resolve these
00:51:25
issues of funding and regulation is
00:51:28
going to very much
00:51:31
determine um what the what applications
00:51:36
get developed um and uh what kinds of
00:51:40
conditions are you know these therapies
00:51:43
being aimed at and who owns them and you
00:51:46
know um what will the financial
00:51:48
incentives be to expand beyond the
00:51:51
initial um uh set of conditions um
00:51:58
addressed um once we get to this phase
00:52:02
and begin to deal with these ethical
00:52:03
challenges we will be you know in effect
00:52:06
living among people
00:52:08
who have brain chips have have uh you
00:52:12
know computers interface to their brains
00:52:14
and we will know so much more than we do
00:52:17
now about how people think about these
00:52:21
things how how it plays out economically
00:52:25
in terms of people people's individual
00:52:27
lives um and I think from that platform
00:52:32
um we will have a much better sense of
00:52:34
you know long-term um you know are
00:52:37
people going to be connecting to the
00:52:39
internet are they going to be as Warwick
00:52:41
said you know sort of merging into one
00:52:42
big mind melded you know web of humanity
00:52:47
um will they want to stay individual and
00:52:49
so forth um right now it's nothing but
00:52:52
you know pure speculation that we can
00:52:55
apply to thinking about these things but
00:52:58
once we've spent some time living you
00:53:00
know in this kind of a world um I think
00:53:04
we'll have a much better sense of of
00:53:06
where the where the concerns are and um
00:53:09
where the you know Wonderful
00:53:11
opportunities
00:53:13
are so that will from from a sort of
00:53:16
platform of living with various kinds of
00:53:21
brain machine interfaces for perception
00:53:23
and motor control um will have a much
00:53:26
better sense of you know uh how how that
00:53:31
might be parlayed into the longer term
00:53:34
future and speaking of sort of step by-
00:53:37
step going into the longer term future I
00:53:39
I just want to leave you with this last
00:53:42
um slide cartoon gay and Wilson um and
00:53:47
I'll read it aloud in case um it's not
00:53:50
uh visible here on the
00:53:52
screen the uh the caveman is saying to
00:53:58
the modern Homo sapiens um I was
00:54:01
wondering when you'd notice there's lots
00:54:03
more
00:54:06
steps so
00:54:11
thanks
