February 17, 2011 • 25 mins
A single ant might not be Mensa material, but a colony of ants can produce some amazingly clever decisions. Tune in as Robert and Julie explore the study of swarm intelligence -- and how it compares to human intelligence.
Learn more about your ad-choices at https://www.iheartpodcastnetwork.com
See omnystudio.com/listener for privacy information.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:03):
Welcome to stuff to Blow your Mind from how Stuff
Works dot com. Hey, welcome to Stuff to Blow your Mind.
My name is Robert Lamb, and we are part of
a machine. Um, I think pretty much most unless you
happened to be like that. That's sort of that iconic
(00:24):
like guy living out in the middle of the desert
in a shack or you know, or you know, Georgia
O'Keefe shutting herself away off the Yeah, living off the
grid and not having any contact with people. Unless you're
doing that kind of life, you're probably part of of
the machine or a machine. Yeah, Like I got up
this morning, got got into Marta. Suddenly I'm in the
(00:45):
I'm in the crowd with the with the with the
other commuters, and I'm instantly part of something that It's
kind of like a big animal, like a big super organism,
a big cranky, uh, kind of stinky superganism that crams
itself into trains and delivers itself to the front door
of its respective workplaces where it becomes part of another
(01:08):
super exactly. Yeah, I mean yeah, just like us How
Stuff Works, Discovery Communications. It's a giant thing and we're
all just like parts of it. We're part of the hive. Yeah,
and we're not I'm not. Aga'm not saying that as
a as any kind of a slam on on Demand
or on Corporate America or any anything like that, because
it's it's the more we look, it's like, no matter
(01:29):
where you are again, unless you're living in the desert
and that's in that shack or or you know, or
up in an isolated apartment on the upper east Side
with the tinfoil over all the windows, you're a part
of some sort of social animal. Right, you're part of
a community. We call it a community. Right now, bees
do it, we call it a swarm. If ants do it,
(01:49):
we call it. I guess that's a swarm too, or
an army at any rate. Yeah, and uh, when when
birds do it and they do it in mass, in
thousands and cousins of birds and they undulate over the horizon,
we just call that scary because we're like, what is that?
It looks like a funnel cloud of birds coming at us,
(02:10):
And then why is it doing that? And beady eyes? Right,
my wife finds the the the tongues terrifying on birds,
like their black tongues. She's anytime we go to Our Thomas,
which is a restaurant here in the Atlanta that birds
and cages outside. She's always little freaked out, like the
two Cans. Did you know that that guy started the
Popeye's chain? And yeah, and Our Thomas is sort of
(02:31):
his way of giving back to the community in terms
of healthy food and birds. Interesting, Yeah, there you go,
but no, but yes, birds, super swarms, super organisms happen
every single day. And one of the reasons they're they're
kind of that they can often be unnerving is that
it's he gets down to the basic idea that an aunt,
(02:52):
a single aunt. There's nothing I can out smart in aunt.
Most people can't. Everybody can can out smart And then
you can take a magnifying glass too and burn it too.
If you were a three year old who a single bird,
you can take that on with a tennis racket. Um,
if it's bees, one b is not an issue. But
when they swarm, hey, you're dealing with a large mass
of things, but also you're dealing with something you can
(03:14):
You're suddenly dealing with something that has um an emergent intelligence,
that that has a it has it's it basically is
it is a macro intelligence that arises from local interactions, right,
and not merely simple interactions, uh you know that are
necessary to coordinate movement, but a group intelligence that learns
(03:38):
that achieves goals, and it engages in an overall self
preservation Okay. And that's what I think is so cool
about super organisms, like you said, on their own as
individuals and they're a little worthless right in the in
the insect community, but together they can accomplish incredible things.
And it's still somewhat of a mystery to scientists. I mean,
(03:58):
they've begun to figure out why birds and swarms of
thousands can uh turn themselves into these crazy vortex um
circular shapes and and not run into each other or
drop dead in collision, you know. Yeah, Or or like
fish that that sort that get into a swarm and
it kind of looks like a single thing, like a
(04:20):
single larger sea creature. Yeah, which is so cool. If
you've never seen footage on that before, it's really worth
checking out. Um. But but again it's pointing to that
sort of intelligence which is far superior than any individual one. Yeah.
In in Steve Johnson's book Emergence, which deals with this
and so generally regard is a really good uh introduction
(04:40):
to to all this. Uh he uh. He lays out
like a few uh necessary items on the checklist for
emergence intelligence. One of them is a critical massive participation.
So you need a like a minimum number of participants
in a social group for like, like like if I
was in Marta and it was like three dudes, Uh,
there's not really gon be much of a It's it's
(05:01):
kind of like a mob, you know, like like if
three guys are standing outside of a of a business
protesting that people aren't gonna say there's a mob out there.
There's no telling what it'll do. No, it's just like Joe,
Charlie and Mary and you can sort of deal with
them on an individual basis and there's there's not going
to be that that thing that takes over um well,
and even in riots, right, there seems to be some
(05:22):
sort of self organizing aspect to it. Right. Yeah, there's
been a there's been some really interesting stuff coming out
recently talking about the at the time of this recording,
current protests in Egypt. You know, because it gets down
to like you can't with a with a group like this,
you can't say oh, there's the leader. No, there are
money leaders there that are that have sort of organizing
roles in it, But you can't. That's I mean, that's
(05:45):
why the mob is such a scary prospect to any
um seat of power, because what does it want? You can't.
It's it's you're dealing with it with an intelligence that
can't be boiled down to a single individual, right right,
that's right. You can't appease that one single individual. And
yet there's an intelligence that comes from that, and there
are demands that come from that. Yeah. Like like I
(06:06):
would like to just to think back to like a corporation,
like like any kind of corporation or big business, Like
what motivates everyone? Like the lowly creative type may just
want to make something they can feel proud of. The
the sales rep might be motivated by, you know, inner
department competition, who can get the biggest sale, who's gonna
be the top seller? Um, who's gonna get yelled out
(06:26):
by Alec Baldwin? Uh. The account it may just have
eyes on you know the bonus uh, and the CEO
is you know, it might just be chasing the American
dream of filling his solid gold submarine with Brazilian supermodels
and circus pandas you know, Yeah, that's it's it's it's
within their grass. So so yeah, but when you talk
about the corporation as a whole, like what does it want? What?
(06:47):
You know, you're dealing with a different set of of
of of goals. Yeah, that parallels are there, right, I mean,
because there's a certain kind of intelligence that that happens
when you put a group of people together. I mean
that come up with solutions that you might not have
come up with yourself, or that people collaborate on and
all of a sudden, you've you've created a thing, right now.
(07:08):
Other items on Steve Johnson's list a local focus, even
if the effects are more widespread. So like the local
focus of of a corporation might be just to you know,
you may just boil it down to making money or
you know, or getting uh you know, or if it's
like a nonprofit, it might be to you know, to
bring about a certain social change, or to get certain
legislature passed. In the in the event of a protest,
(07:30):
it may be to alsta a particular individual or to
or to oppose a particular unpopular um law or ruling. Um.
Then another item on this list is random interactions between
groups and individuals outside. Uh so this is necessary for
a learning process. You know. It's it's like the mob pushing.
(07:54):
It's like you know, any kind of scene where it's
like the man with his police and his and his
riot shield olds on one side, it's the protesters on
the other, and there's like a there's kind of a push.
There's a testing of the of the two sides. As
as the mob is is learning, it's seeing what it
can it can get away with, what it can do, Yeah,
learning what his boundaries are. And um, you know there's
(08:16):
some there's so many people on the lookout and they're
all gathering data and pulling it in one way or another. Right,
which is the same thing if you look at and
some with some insects, with some species that they're all
kind of like ants, for instance, they're communicating with pheromones, right,
and they're basically they're saying, Okay, hey, I'm going to
leave this chemical trail to this food source that I
(08:37):
just found. And then all of a sudden, you've got
one ant following that, five ants, ten ants, a thousand ants.
And that's how you're getting this sort of group movement
toward that food source or the bara metric pressure drops
for instance, and the ants start building these little sand
turrets to protect themselves from what they think maybe rain
(09:00):
in pretty soon. So again, it's that collective information and
sharing of it and this self directing movement that comes
out of it that's so fascinating. Yeah, with ants, it's
important to note and just to remind everybody, they're they're
like twelve thousand uh known ant species. They've been ruling
the earth for about a hundred and forty million years
(09:20):
and uh. And while an ant colony or or or
even like a bee hive will have a queen, these
are not rulers. Like the queen plays an important reproductive
role in a in a in a hive or colony,
but they're not calling the shots. Now. She's just a
cog in the wheel. Right. Yeah, we we label them
a queen and that comes with certain certain anthropomorphic baggage.
(09:43):
But just leave that baggage behind, because the queen is
just there to to pump out more more bees or ants. Right,
that were just that they she's just useful her for
her ovaries. Really to throw it down to that and
that's true of all these species, right, the insects species
that that show emergence. Yeah, there's no one leader, there's
no boss. Yeah. She's not saying, hey, I've got an
(10:04):
idea how to solve this problem about food foraging. No,
it's it's it's emerging from the the the interactions of
the group as it pushes against outside stimuli and what
the group needs. Yeah, there's an interesting study from US
Stanford University's Deborah Gordon. Uh and uh this dealt with
with the with particularly with forager ants and figuring the
(10:25):
ant colony, figuring out how many foragers it needs to
send out. Um. And they found that it came down
interactions between foragers and patrollers. So when a forager has
contact with patroller you know looking around, um that they
end up passing on these these pheromones. But a forager
needs several contacts no more than like ten seconds apart
(10:48):
before it'll go out. So there has to be like
it's it's like a very simple like you know, uh
here here the patrollers coming in, patrollers coming in, and
once it like counts off enough pheromones within a certain
amount of time, then it knows Oh I need to
go out and collect Okay. So it's sort of like, again,
is that mass movement? Right? So once it starts to
accumulate all that data from ten of those and it says, okay,
that's the direction I need to go in. Yeah. That
(11:10):
They concluded that forgers use the rate of their encounters
with patrollers to tell if it's safe to go out. Okay,
so you know, in case there's like an ant eater
out there just gobbling up everything, they won't just blindly
march out into the mall. You know, there's something altruistic
about that. It seems you know, again that's that's anthropomorphis
morphosizing this. But it does seem like, okay, well I've
(11:32):
got your back here, or you know, or I may
even die for the cause as I'm patrolling. Yeah, And it's, uh,
that was so romantic. I can't have much. It's hard
not to get romantic about it. This presentation is brought
to you by Intel Sponsors of Tomorrow. Another really interesting
(12:00):
part of all this is that but we've been able
to to look at some of these systems and learn
from them and apply them to, uh, to our own
human systems, our own um transportation or delivery UM systems
such as UH. There's an industrial medical gas guru by
the name of air Liquid and they worked with an
(12:21):
AI form firm to develop a computer model based on
algorithms that were inspired by the forging behavior of argentine ants.
And this is again the species that we're talking about,
the deposits chemical pheromones um and and uses that to dictate,
you know, how much how many forgers need to go
out to get the food. So air Liquid is able
(12:41):
to combine the and the ant approach with other artificial
intelligence techniques, throw in data about like the weather, customer
demand and all, and they use that to figure out
how much how much of their product they need to
actually ship out. Okay, and this is based on the
algorithms that they observed. Yeah, okay, based on algorithm on
the on the way the ants are handling their problem,
and they're able to up with a way to to
handle their uh, their distributing needs. UM. And you've heard
(13:06):
of void technology before too, which is sort of similar
in that UM BOYD boy. This is bo I D
B o I D, which I guess that's sort of
a void pronunciation of bird, right, And this is Craig
Reynolds who created UM, an artificial life program, and this
(13:26):
is a distributed behavioral model to simulate the motion of
a flock of birds on a computer. So same thing, right,
and nowther space goes back to the the really central
human need to make believable batswarms in the movie Batman. Yes, yeah,
actually you're right. I mean that that wasn't why he
actually created this UH technology, but that's one of the
(13:48):
uses for it. Was c G I s to to
simulate something that looked realistic in terms of a flock
of bats UM. But what he observed is that each
boid is implemented as an agent and moves according to
its own understanding of the changing environment. So think about
bird formations like V formations UM. There are four rules
(14:08):
with this. First, a boid must move away from other
boyds they're too close stuff because you don't want to
applyde Second, they have to fly in a general direction
that the flock is moving, okay. Third, they need to
minimize their exposure to flox exterior by moving towards the
perceived centers, which is interesting if you think about that
V formation. I like how your use of BOYD has
made you pronounced toward more toward. Of course, that's why
(14:32):
he came up with his name. He's like, this is
totally going to screw the podcasters in the future. I'm
just I'm taking the spirit of it, you know, I'm
running with it. UM. And then fourth, a boyd should
move laterally away from any boy that blocks its view.
So I mean that's very simple, right, taking the forcible
rules in creating UM. This this algorithm UM that has
also been used by Southwest Airlines. Uh, They've got a
(14:54):
routing system that in which the pilots try to find
the best airport gate based on the group information coming in.
So again you've got the group intelligence aspect of it. Yeah.
It's kind of like ants figuring out which entrigued is
the colony to take when returning home. Yeah. Yeah, And
it's a little bit scary with with you know, with
the plane, just because you think, okay, I'm relying on
(15:15):
the group intelligence for this, But over and over again,
there's so many different studies and counting jelly beans, you know,
guessing the weight of something. Yeah, like the just the
basic like second third grade activity of getting all the
kids to to guess the jelly bean number. And then
everybody comes up with these totals, and some kids maybe
way out, some kids don't know what numbers are, and
they're like, you know, drawing a cat on the ballot.
(15:37):
But but when you put them all together and you
you get you get an average. You can estimate estimate
more or less exactly how many there are in there.
The mean of right, the mean of that the group
estimate is always closer or the closest to what it
actually is, which is astounding. The main key here is
that you want a diverse members, you want independent minded members,
and you want a mechanism of voting or averaging to
(16:00):
figure out how it works. Right. Okay, you don't want
like fifty engineers all in the same room, right, I
mean you actually want it to be a diverse population
of people. Um. I mean, you could have fifty engineers
in a room. But the point is is that it
doesn't it doesn't really matter. You have people from all
social all sort of so so economic strata um giving
(16:21):
their information, and the mean will always be more accurate.
It's kind of like the whole how many blank does
it take to fill in a light bulb? You know?
Because I mean it's always a joke about like how
many you know, how many hipsters, how many mechanics, how
many you know blondes or whatever? Um, you know, does
it take? And it's kind of the message is kind
of if you get all, if you don't have a
(16:42):
diverse group, they're not going to be able to fulfill
even a simple task to the best of their ability.
But if it were a priest, a rabbi, and uh
whatever the third one is, then hey, they were an engineer,
an engineer, then they'd have that light screwed in no time.
That's right. Yeah, it takes up the logic. I guess
it's the point of all emergence here. Um, I mean
(17:04):
it really is. It's like it's such a simple construct,
but it really has changed the way that we think
about intelligence and the way that we operate in our
own society and uh, the sort of meaning that we
ascribe to CEOs or kings or presidents. Um, if we
all are essentially self organizing, uh like ants, then and
(17:28):
we don't necessarily need a leader, and of course we
need it like psychologically. Right, Um, we're still gonna be
okay because we're gonna still come out in theory, um
with something that makes sense for our community, right because
we're gonna be acting in our best interests. There's along
those those lines and kind of taking him to a
slightly more troublesome territory. Um, there's the guy, a guy
(17:52):
by the name of John Robb. He wrote a book
called The Brave New War, and he has a really
cool blog called Global Guerrillas. And he made an interesting
point where he pointed to the insurgency in a rack
in Iraq is a kind of emergence intelligence, uh, he said. Quote.
He talked about it as quote a complex series of
local interactions that leads to shifts in its behavior that
(18:15):
reflect a complex learning goal attainment and self preservation despite
a lack of a leadership hierarchy. Uh And and he
boils this down into three like key facts about the
insurgency in I rock that one the insurgency will continue
to improve over time to breakout as possible, and three
that it is impossible to discern the motives of this movement.
(18:38):
So uh, I found that interesting commentary, you know, because
it's it's the idea that you're not you're not fighting
a single individual that as much as is everyone like
to you know, believe it's like, all right, this this guy,
let's find him, uh kill him or or you know,
overthrow him, and it's then it's done. But no, you're
dealing with a I mean, ay, you're dealing with a
far more complicated social and political situation. But also you're
(19:02):
you're dealing with a movement that doesn't doesn't have a leader.
That's uh, it's you're dealing with the emergence intelligence. And
again it's how do you figure out what that wants?
And is the second point to break out is that
the aspect of almost like when with a bird flock,
where it all of a sudden it changes direction and
then you know, we maybe have the movement of several
birds and then that prompts um the other ones to
(19:23):
sort of follow domino like yeah, or like with you know,
sometimes you'll have like whales that will watch upon the
beach because one has kind of like had gone has
gone haywire, and the others just follow. UM. This technology
is also being used with robots of course, right right,
artificial intelligence. UM. It's based on the collective behavior again
(19:47):
of decentralized self organized systems. Just to throw that out
there again and again it's identifying best solutions and communicating them.
So they're they're hoping to have this same sort of
behavior in robots UM, and how it could be applied
is that you could use something called morphogen morphogenesis UH
to create a group of robots that can reconfigure and
(20:08):
undertake different functions. For instance, you could have a group
of ten robots that may communicate among each other by
admitting different color lights, and that light it's say if
it's green and blue, could mean that the other robots
need to attach to each other and create some sort
of formation. And the formation could be to traverse some
sort of gap in the landscape if that's what the
(20:29):
robots were doing, or could become up exactly, it becomes
a stack up on each other UM, or it could
be used to form a shovel shape to push a
heavy object. So they're trying to use again that swarm
technology there. But more interestingly I think that UM in Georgia,
at Georgia Tech of course, where they're doing tons of
(20:50):
crazy stuff with robots and creating robot armies. UM they've
developed palm sized robots to use in combat zones, and
they're smart enough to go up on their own and
alert you when they find something. So they're trying to
figure out how best to have these bots interact with
each other, and right now it's looking like the research
is favoring a decentralized approach in which the individual robots
(21:13):
would share information among their companions to form a more
complete picture. So what are we talking about here? Emergence intelligence? Right? Yeah,
so I mean here you go again, robots, emergence intelligence, intelligence,
emergence intelligence, military intelligence. I guess you could say that. Yeah. Yeah,
(21:33):
So you know, anytime we talk about robots, which I
know we've been doing a lot lately, we we sort
of get a little bit depressed because we think, oh god,
they're they're being programmed to destroy us again, or destroy
someone or do something. But I think that, um, the
really cool thing about emergence is that it could actually
lead to a better understanding of our own human nature.
(21:55):
Right if we apply that sort of intelligence to ourselves.
And I am not saying that we don't need presidents,
CEOs or power structures are hierarchy in place, Yeah, because
without the CEO, the people who make the Golden Submarines
are not going to make a living, right, It's very important, Yeah,
I mean, the economy will go down and it'll be
it'll be a bad thing. Um. But I think that
it's interesting to think that we've got the knowledge that
(22:17):
we are self directing and there's something comforting in that
and that. Um. You know, maybe again here I am
romanticizing things, but romantic in this one. Um. But given
the chance, maybe we could bring up the better angels
of our nature. Yeah, as we we figure out how
to make a machine culture work, we can see how
we should actually make a human culture work. Right, I
(22:39):
dig it. Okay, we'll go with it. Let's put it
into place. Hey, So let us know what you think
if you've seen some some cool examples of emergence intelligence
in the people around you and the animals around you,
or or the company that you're a part of, or
or or even in machines and robots, let us know
that's right. And if locust start converging, don't worry. They're
(23:01):
just they can't help it. That's what they do. Yeah.
I don't hate the individual locusts. It's uh, you know,
the mob been taught. So hey, I have a couple
of listener mail items here to read. The first comes
from Lisa and This was left on our Facebook page. Um.
She says, I like to listen to your podcast as
I fall asleep, which is great, it helps me sleep.
(23:23):
I listened to it again once I'm awake while I
learned the stuff while I'm asleep. The side effect to
all this really strange and scary dreams. So sorry about that,
which it'll happen. I, Um, I forget if I mentioned this,
but when we were researching our podcast on I was
really reading that and researching it right before bed and
(23:44):
I had a bunch of crazy dreams with like atomic
tests going on on a college campus, and it was
it was kind of unsettling. Yeah. Yeah, I have to
say I haven't been able to access my dreams lately.
That's probably a good thing. Yeah, because it's just loaded
with robots running away. Yeah, robo Nanny's and uh robot
locus cool. I have another uh uh. This one is
(24:05):
an email from our listener Chris, and he says, I
just want to tell you that there are several types
of deafness in your podcast. The werewolf principle. You sayd
that deaf people don't get motion sick. That would only
be if they had neural deafness deafness. There are two
basic types of deafness, neural and conductive. The neural deafness
is when the cochlea doesn't send the sounds to the
(24:29):
brain because the cochlea isn't working or the cochlea is
missing um. The conductive hearing loss is when the sound
doesn't get to the cochlea. This occurs when the little
bones in the ear uh malius incas and states don't
send the sound to the cochlea or the ear drum
is missing. If you are going to make people death,
(24:51):
you would have have to uh do something to the
cochlea or the malius incas or Stapus. Just wanted you
to know, so, all right, cool. I should have maybe
looked up the pronunciation and all those things through that email,
but I felt like you were going Latin there. It's
kind of exciting. Yeah, So hey, if you have anything
(25:14):
to uh to let us know about, you can drop
by our Twitter or our Facebook account. We are blow
the mind on both of those and we keep that
updated with all sorts of cool links to stuff going
on and how stuff works and also you know, neat
little news stories and mind blowing tidnets from throughout the web,
and you can always drop us a line that blow
the mind at how stuff works dot com for moral this,
(25:40):
and thousands of other topics. Is it how stuff works
dot com. To learn more about the podcast, click on
the podcast icon in the upper right corner of our homepage.
The How stuff Works iPhone app has a ride. Download
it today on iTunes.
Welcome to stuff to Blow your Mind from how Stuff
Works dot com. Hey, welcome to Stuff to Blow your Mind.
My name is Robert Lamb, and we are part of
a machine. Um, I think pretty much most unless you
happened to be like that. That's sort of that iconic
(00:24):
like guy living out in the middle of the desert
in a shack or you know, or you know, Georgia
O'Keefe shutting herself away off the Yeah, living off the
grid and not having any contact with people. Unless you're
doing that kind of life, you're probably part of of
the machine or a machine. Yeah, Like I got up
this morning, got got into Marta. Suddenly I'm in the
(00:45):
I'm in the crowd with the with the with the
other commuters, and I'm instantly part of something that It's
kind of like a big animal, like a big super organism,
a big cranky, uh, kind of stinky superganism that crams
itself into trains and delivers itself to the front door
of its respective workplaces where it becomes part of another
(01:08):
super exactly. Yeah, I mean yeah, just like us How
Stuff Works, Discovery Communications. It's a giant thing and we're
all just like parts of it. We're part of the hive. Yeah,
and we're not I'm not. Aga'm not saying that as
a as any kind of a slam on on Demand
or on Corporate America or any anything like that, because
it's it's the more we look, it's like, no matter
(01:29):
where you are again, unless you're living in the desert
and that's in that shack or or you know, or
up in an isolated apartment on the upper east Side
with the tinfoil over all the windows, you're a part
of some sort of social animal. Right, you're part of
a community. We call it a community. Right now, bees
do it, we call it a swarm. If ants do it,
(01:49):
we call it. I guess that's a swarm too, or
an army at any rate. Yeah, and uh, when when
birds do it and they do it in mass, in
thousands and cousins of birds and they undulate over the horizon,
we just call that scary because we're like, what is that?
It looks like a funnel cloud of birds coming at us,
(02:10):
And then why is it doing that? And beady eyes? Right,
my wife finds the the the tongues terrifying on birds,
like their black tongues. She's anytime we go to Our Thomas,
which is a restaurant here in the Atlanta that birds
and cages outside. She's always little freaked out, like the
two Cans. Did you know that that guy started the
Popeye's chain? And yeah, and Our Thomas is sort of
(02:31):
his way of giving back to the community in terms
of healthy food and birds. Interesting, Yeah, there you go,
but no, but yes, birds, super swarms, super organisms happen
every single day. And one of the reasons they're they're
kind of that they can often be unnerving is that
it's he gets down to the basic idea that an aunt,
(02:52):
a single aunt. There's nothing I can out smart in aunt.
Most people can't. Everybody can can out smart And then
you can take a magnifying glass too and burn it too.
If you were a three year old who a single bird,
you can take that on with a tennis racket. Um,
if it's bees, one b is not an issue. But
when they swarm, hey, you're dealing with a large mass
of things, but also you're dealing with something you can
(03:14):
You're suddenly dealing with something that has um an emergent intelligence,
that that has a it has it's it basically is
it is a macro intelligence that arises from local interactions, right,
and not merely simple interactions, uh you know that are
necessary to coordinate movement, but a group intelligence that learns
(03:38):
that achieves goals, and it engages in an overall self
preservation Okay. And that's what I think is so cool
about super organisms, like you said, on their own as
individuals and they're a little worthless right in the in
the insect community, but together they can accomplish incredible things.
And it's still somewhat of a mystery to scientists. I mean,
(03:58):
they've begun to figure out why birds and swarms of
thousands can uh turn themselves into these crazy vortex um
circular shapes and and not run into each other or
drop dead in collision, you know. Yeah, Or or like
fish that that sort that get into a swarm and
it kind of looks like a single thing, like a
(04:20):
single larger sea creature. Yeah, which is so cool. If
you've never seen footage on that before, it's really worth
checking out. Um. But but again it's pointing to that
sort of intelligence which is far superior than any individual one. Yeah.
In in Steve Johnson's book Emergence, which deals with this
and so generally regard is a really good uh introduction
(04:40):
to to all this. Uh he uh. He lays out
like a few uh necessary items on the checklist for
emergence intelligence. One of them is a critical massive participation.
So you need a like a minimum number of participants
in a social group for like, like like if I
was in Marta and it was like three dudes, Uh,
there's not really gon be much of a It's it's
(05:01):
kind of like a mob, you know, like like if
three guys are standing outside of a of a business
protesting that people aren't gonna say there's a mob out there.
There's no telling what it'll do. No, it's just like Joe,
Charlie and Mary and you can sort of deal with
them on an individual basis and there's there's not going
to be that that thing that takes over um well,
and even in riots, right, there seems to be some
(05:22):
sort of self organizing aspect to it. Right. Yeah, there's
been a there's been some really interesting stuff coming out
recently talking about the at the time of this recording,
current protests in Egypt. You know, because it gets down
to like you can't with a with a group like this,
you can't say oh, there's the leader. No, there are
money leaders there that are that have sort of organizing
roles in it, But you can't. That's I mean, that's
(05:45):
why the mob is such a scary prospect to any
um seat of power, because what does it want? You can't.
It's it's you're dealing with it with an intelligence that
can't be boiled down to a single individual, right right,
that's right. You can't appease that one single individual. And
yet there's an intelligence that comes from that, and there
are demands that come from that. Yeah. Like like I
(06:06):
would like to just to think back to like a corporation,
like like any kind of corporation or big business, Like
what motivates everyone? Like the lowly creative type may just
want to make something they can feel proud of. The
the sales rep might be motivated by, you know, inner
department competition, who can get the biggest sale, who's gonna
be the top seller? Um, who's gonna get yelled out
(06:26):
by Alec Baldwin? Uh. The account it may just have
eyes on you know the bonus uh, and the CEO
is you know, it might just be chasing the American
dream of filling his solid gold submarine with Brazilian supermodels
and circus pandas you know, Yeah, that's it's it's it's
within their grass. So so yeah, but when you talk
about the corporation as a whole, like what does it want? What?
(06:47):
You know, you're dealing with a different set of of
of of goals. Yeah, that parallels are there, right, I mean,
because there's a certain kind of intelligence that that happens
when you put a group of people together. I mean
that come up with solutions that you might not have
come up with yourself, or that people collaborate on and
all of a sudden, you've you've created a thing, right now.
(07:08):
Other items on Steve Johnson's list a local focus, even
if the effects are more widespread. So like the local
focus of of a corporation might be just to you know,
you may just boil it down to making money or
you know, or getting uh you know, or if it's
like a nonprofit, it might be to you know, to
bring about a certain social change, or to get certain
legislature passed. In the in the event of a protest,
(07:30):
it may be to alsta a particular individual or to
or to oppose a particular unpopular um law or ruling. Um.
Then another item on this list is random interactions between
groups and individuals outside. Uh so this is necessary for
a learning process. You know. It's it's like the mob pushing.
(07:54):
It's like you know, any kind of scene where it's
like the man with his police and his and his
riot shield olds on one side, it's the protesters on
the other, and there's like a there's kind of a push.
There's a testing of the of the two sides. As
as the mob is is learning, it's seeing what it
can it can get away with, what it can do, Yeah,
learning what his boundaries are. And um, you know there's
(08:16):
some there's so many people on the lookout and they're
all gathering data and pulling it in one way or another. Right,
which is the same thing if you look at and
some with some insects, with some species that they're all
kind of like ants, for instance, they're communicating with pheromones, right,
and they're basically they're saying, Okay, hey, I'm going to
leave this chemical trail to this food source that I
(08:37):
just found. And then all of a sudden, you've got
one ant following that, five ants, ten ants, a thousand ants.
And that's how you're getting this sort of group movement
toward that food source or the bara metric pressure drops
for instance, and the ants start building these little sand
turrets to protect themselves from what they think maybe rain
(09:00):
in pretty soon. So again, it's that collective information and
sharing of it and this self directing movement that comes
out of it that's so fascinating. Yeah, with ants, it's
important to note and just to remind everybody, they're they're
like twelve thousand uh known ant species. They've been ruling
the earth for about a hundred and forty million years
(09:20):
and uh. And while an ant colony or or or
even like a bee hive will have a queen, these
are not rulers. Like the queen plays an important reproductive
role in a in a in a hive or colony,
but they're not calling the shots. Now. She's just a
cog in the wheel. Right. Yeah, we we label them
a queen and that comes with certain certain anthropomorphic baggage.
(09:43):
But just leave that baggage behind, because the queen is
just there to to pump out more more bees or ants. Right,
that were just that they she's just useful her for
her ovaries. Really to throw it down to that and
that's true of all these species, right, the insects species
that that show emergence. Yeah, there's no one leader, there's
no boss. Yeah. She's not saying, hey, I've got an
(10:04):
idea how to solve this problem about food foraging. No,
it's it's it's emerging from the the the interactions of
the group as it pushes against outside stimuli and what
the group needs. Yeah, there's an interesting study from US
Stanford University's Deborah Gordon. Uh and uh this dealt with
with the with particularly with forager ants and figuring the
(10:25):
ant colony, figuring out how many foragers it needs to
send out. Um. And they found that it came down
interactions between foragers and patrollers. So when a forager has
contact with patroller you know looking around, um that they
end up passing on these these pheromones. But a forager
needs several contacts no more than like ten seconds apart
(10:48):
before it'll go out. So there has to be like
it's it's like a very simple like you know, uh
here here the patrollers coming in, patrollers coming in, and
once it like counts off enough pheromones within a certain
amount of time, then it knows Oh I need to
go out and collect Okay. So it's sort of like, again,
is that mass movement? Right? So once it starts to
accumulate all that data from ten of those and it says, okay,
that's the direction I need to go in. Yeah. That
(11:10):
They concluded that forgers use the rate of their encounters
with patrollers to tell if it's safe to go out. Okay,
so you know, in case there's like an ant eater
out there just gobbling up everything, they won't just blindly
march out into the mall. You know, there's something altruistic
about that. It seems you know, again that's that's anthropomorphis
morphosizing this. But it does seem like, okay, well I've
(11:32):
got your back here, or you know, or I may
even die for the cause as I'm patrolling. Yeah, And it's, uh,
that was so romantic. I can't have much. It's hard
not to get romantic about it. This presentation is brought
to you by Intel Sponsors of Tomorrow. Another really interesting
(12:00):
part of all this is that but we've been able
to to look at some of these systems and learn
from them and apply them to, uh, to our own
human systems, our own um transportation or delivery UM systems
such as UH. There's an industrial medical gas guru by
the name of air Liquid and they worked with an
(12:21):
AI form firm to develop a computer model based on
algorithms that were inspired by the forging behavior of argentine ants.
And this is again the species that we're talking about,
the deposits chemical pheromones um and and uses that to dictate,
you know, how much how many forgers need to go
out to get the food. So air Liquid is able
(12:41):
to combine the and the ant approach with other artificial
intelligence techniques, throw in data about like the weather, customer
demand and all, and they use that to figure out
how much how much of their product they need to
actually ship out. Okay, and this is based on the
algorithms that they observed. Yeah, okay, based on algorithm on
the on the way the ants are handling their problem,
and they're able to up with a way to to
handle their uh, their distributing needs. UM. And you've heard
(13:06):
of void technology before too, which is sort of similar
in that UM BOYD boy. This is bo I D
B o I D, which I guess that's sort of
a void pronunciation of bird, right, And this is Craig
Reynolds who created UM, an artificial life program, and this
(13:26):
is a distributed behavioral model to simulate the motion of
a flock of birds on a computer. So same thing, right,
and nowther space goes back to the the really central
human need to make believable batswarms in the movie Batman. Yes, yeah,
actually you're right. I mean that that wasn't why he
actually created this UH technology, but that's one of the
(13:48):
uses for it. Was c G I s to to
simulate something that looked realistic in terms of a flock
of bats UM. But what he observed is that each
boid is implemented as an agent and moves according to
its own understanding of the changing environment. So think about
bird formations like V formations UM. There are four rules
(14:08):
with this. First, a boid must move away from other
boyds they're too close stuff because you don't want to
applyde Second, they have to fly in a general direction
that the flock is moving, okay. Third, they need to
minimize their exposure to flox exterior by moving towards the
perceived centers, which is interesting if you think about that
V formation. I like how your use of BOYD has
made you pronounced toward more toward. Of course, that's why
(14:32):
he came up with his name. He's like, this is
totally going to screw the podcasters in the future. I'm
just I'm taking the spirit of it, you know, I'm
running with it. UM. And then fourth, a boyd should
move laterally away from any boy that blocks its view.
So I mean that's very simple, right, taking the forcible
rules in creating UM. This this algorithm UM that has
also been used by Southwest Airlines. Uh, They've got a
(14:54):
routing system that in which the pilots try to find
the best airport gate based on the group information coming in.
So again you've got the group intelligence aspect of it. Yeah.
It's kind of like ants figuring out which entrigued is
the colony to take when returning home. Yeah. Yeah, And
it's a little bit scary with with you know, with
the plane, just because you think, okay, I'm relying on
(15:15):
the group intelligence for this, But over and over again,
there's so many different studies and counting jelly beans, you know,
guessing the weight of something. Yeah, like the just the
basic like second third grade activity of getting all the
kids to to guess the jelly bean number. And then
everybody comes up with these totals, and some kids maybe
way out, some kids don't know what numbers are, and
they're like, you know, drawing a cat on the ballot.
(15:37):
But but when you put them all together and you
you get you get an average. You can estimate estimate
more or less exactly how many there are in there.
The mean of right, the mean of that the group
estimate is always closer or the closest to what it
actually is, which is astounding. The main key here is
that you want a diverse members, you want independent minded members,
and you want a mechanism of voting or averaging to
(16:00):
figure out how it works. Right. Okay, you don't want
like fifty engineers all in the same room, right, I
mean you actually want it to be a diverse population
of people. Um. I mean, you could have fifty engineers
in a room. But the point is is that it
doesn't it doesn't really matter. You have people from all
social all sort of so so economic strata um giving
(16:21):
their information, and the mean will always be more accurate.
It's kind of like the whole how many blank does
it take to fill in a light bulb? You know?
Because I mean it's always a joke about like how
many you know, how many hipsters, how many mechanics, how
many you know blondes or whatever? Um, you know, does
it take? And it's kind of the message is kind
of if you get all, if you don't have a
(16:42):
diverse group, they're not going to be able to fulfill
even a simple task to the best of their ability.
But if it were a priest, a rabbi, and uh
whatever the third one is, then hey, they were an engineer,
an engineer, then they'd have that light screwed in no time.
That's right. Yeah, it takes up the logic. I guess
it's the point of all emergence here. Um, I mean
(17:04):
it really is. It's like it's such a simple construct,
but it really has changed the way that we think
about intelligence and the way that we operate in our
own society and uh, the sort of meaning that we
ascribe to CEOs or kings or presidents. Um, if we
all are essentially self organizing, uh like ants, then and
(17:28):
we don't necessarily need a leader, and of course we
need it like psychologically. Right, Um, we're still gonna be
okay because we're gonna still come out in theory, um
with something that makes sense for our community, right because
we're gonna be acting in our best interests. There's along
those those lines and kind of taking him to a
slightly more troublesome territory. Um, there's the guy, a guy
(17:52):
by the name of John Robb. He wrote a book
called The Brave New War, and he has a really
cool blog called Global Guerrillas. And he made an interesting
point where he pointed to the insurgency in a rack
in Iraq is a kind of emergence intelligence, uh, he said. Quote.
He talked about it as quote a complex series of
local interactions that leads to shifts in its behavior that
(18:15):
reflect a complex learning goal attainment and self preservation despite
a lack of a leadership hierarchy. Uh And and he
boils this down into three like key facts about the
insurgency in I rock that one the insurgency will continue
to improve over time to breakout as possible, and three
that it is impossible to discern the motives of this movement.
(18:38):
So uh, I found that interesting commentary, you know, because
it's it's the idea that you're not you're not fighting
a single individual that as much as is everyone like
to you know, believe it's like, all right, this this guy,
let's find him, uh kill him or or you know,
overthrow him, and it's then it's done. But no, you're
dealing with a I mean, ay, you're dealing with a
far more complicated social and political situation. But also you're
(19:02):
you're dealing with a movement that doesn't doesn't have a leader.
That's uh, it's you're dealing with the emergence intelligence. And
again it's how do you figure out what that wants?
And is the second point to break out is that
the aspect of almost like when with a bird flock,
where it all of a sudden it changes direction and
then you know, we maybe have the movement of several
birds and then that prompts um the other ones to
(19:23):
sort of follow domino like yeah, or like with you know,
sometimes you'll have like whales that will watch upon the
beach because one has kind of like had gone has
gone haywire, and the others just follow. UM. This technology
is also being used with robots of course, right right,
artificial intelligence. UM. It's based on the collective behavior again
(19:47):
of decentralized self organized systems. Just to throw that out
there again and again it's identifying best solutions and communicating them.
So they're they're hoping to have this same sort of
behavior in robots UM, and how it could be applied
is that you could use something called morphogen morphogenesis UH
to create a group of robots that can reconfigure and
(20:08):
undertake different functions. For instance, you could have a group
of ten robots that may communicate among each other by
admitting different color lights, and that light it's say if
it's green and blue, could mean that the other robots
need to attach to each other and create some sort
of formation. And the formation could be to traverse some
sort of gap in the landscape if that's what the
(20:29):
robots were doing, or could become up exactly, it becomes
a stack up on each other UM, or it could
be used to form a shovel shape to push a
heavy object. So they're trying to use again that swarm
technology there. But more interestingly I think that UM in Georgia,
at Georgia Tech of course, where they're doing tons of
(20:50):
crazy stuff with robots and creating robot armies. UM they've
developed palm sized robots to use in combat zones, and
they're smart enough to go up on their own and
alert you when they find something. So they're trying to
figure out how best to have these bots interact with
each other, and right now it's looking like the research
is favoring a decentralized approach in which the individual robots
(21:13):
would share information among their companions to form a more
complete picture. So what are we talking about here? Emergence intelligence? Right? Yeah,
so I mean here you go again, robots, emergence intelligence, intelligence,
emergence intelligence, military intelligence. I guess you could say that. Yeah. Yeah,
(21:33):
So you know, anytime we talk about robots, which I
know we've been doing a lot lately, we we sort
of get a little bit depressed because we think, oh god,
they're they're being programmed to destroy us again, or destroy
someone or do something. But I think that, um, the
really cool thing about emergence is that it could actually
lead to a better understanding of our own human nature.
(21:55):
Right if we apply that sort of intelligence to ourselves.
And I am not saying that we don't need presidents,
CEOs or power structures are hierarchy in place, Yeah, because
without the CEO, the people who make the Golden Submarines
are not going to make a living, right, It's very important, Yeah,
I mean, the economy will go down and it'll be
it'll be a bad thing. Um. But I think that
it's interesting to think that we've got the knowledge that
(22:17):
we are self directing and there's something comforting in that
and that. Um. You know, maybe again here I am
romanticizing things, but romantic in this one. Um. But given
the chance, maybe we could bring up the better angels
of our nature. Yeah, as we we figure out how
to make a machine culture work, we can see how
we should actually make a human culture work. Right, I
(22:39):
dig it. Okay, we'll go with it. Let's put it
into place. Hey, So let us know what you think
if you've seen some some cool examples of emergence intelligence
in the people around you and the animals around you,
or or the company that you're a part of, or
or or even in machines and robots, let us know
that's right. And if locust start converging, don't worry. They're
(23:01):
just they can't help it. That's what they do. Yeah.
I don't hate the individual locusts. It's uh, you know,
the mob been taught. So hey, I have a couple
of listener mail items here to read. The first comes
from Lisa and This was left on our Facebook page. Um.
She says, I like to listen to your podcast as
I fall asleep, which is great, it helps me sleep.
(23:23):
I listened to it again once I'm awake while I
learned the stuff while I'm asleep. The side effect to
all this really strange and scary dreams. So sorry about that,
which it'll happen. I, Um, I forget if I mentioned this,
but when we were researching our podcast on I was
really reading that and researching it right before bed and
(23:44):
I had a bunch of crazy dreams with like atomic
tests going on on a college campus, and it was
it was kind of unsettling. Yeah. Yeah, I have to
say I haven't been able to access my dreams lately.
That's probably a good thing. Yeah, because it's just loaded
with robots running away. Yeah, robo Nanny's and uh robot
locus cool. I have another uh uh. This one is
(24:05):
an email from our listener Chris, and he says, I
just want to tell you that there are several types
of deafness in your podcast. The werewolf principle. You sayd
that deaf people don't get motion sick. That would only
be if they had neural deafness deafness. There are two
basic types of deafness, neural and conductive. The neural deafness
is when the cochlea doesn't send the sounds to the
(24:29):
brain because the cochlea isn't working or the cochlea is
missing um. The conductive hearing loss is when the sound
doesn't get to the cochlea. This occurs when the little
bones in the ear uh malius incas and states don't
send the sound to the cochlea or the ear drum
is missing. If you are going to make people death,
(24:51):
you would have have to uh do something to the
cochlea or the malius incas or Stapus. Just wanted you
to know, so, all right, cool. I should have maybe
looked up the pronunciation and all those things through that email,
but I felt like you were going Latin there. It's
kind of exciting. Yeah, So hey, if you have anything
(25:14):
to uh to let us know about, you can drop
by our Twitter or our Facebook account. We are blow
the mind on both of those and we keep that
updated with all sorts of cool links to stuff going
on and how stuff works and also you know, neat
little news stories and mind blowing tidnets from throughout the web,
and you can always drop us a line that blow
the mind at how stuff works dot com for moral this,
(25:40):
and thousands of other topics. Is it how stuff works
dot com. To learn more about the podcast, click on
the podcast icon in the upper right corner of our homepage.
The How stuff Works iPhone app has a ride. Download
it today on iTunes.
Stuff To Blow Your Mind News
Hosts And Creators
Robert Lamb
Joe McCormick
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
Las Culturistas with Matt Rogers and Bowen Yang
Ding dong! Join your culture consultants, Matt Rogers and Bowen Yang, on an unforgettable journey into the beating heart of CULTURE. Alongside sizzling special guests, they GET INTO the hottest pop-culture moments of the day and the formative cultural experiences that turned them into Culturistas. Produced by the Big Money Players Network and iHeartRadio.