sponsored links
TEDxZurich 2013

Nicolas Perony: Puppies! Now that I’ve got your attention, complexity theory

October 3, 2013

Animal behavior isn't complicated, but it is complex. Nicolas Perony studies how individual animals -- be they Scottish Terriers, bats or meerkats -- follow simple rules that, collectively, create larger patterns of behavior. And how this complexity born of simplicity can help them adapt to new circumstances, as they arise.

Nicolas Perony - Animal scientist
Nicolas Perony models the movement of animal groups to understand: what is the individual behavior that guides the behavior of the larger society? Full bio

sponsored links
Double-click the English subtitles below to play the video.
Science,
00:15
science has allowed us to know so much
00:16
about the far reaches of the universe,
00:19
which is at the same time tremendously important
00:22
and extremely remote,
00:25
and yet much, much closer,
00:28
much more directly related to us,
00:30
there are many things we don't really understand.
00:32
And one of them is the extraordinary
00:35
social complexity of the animals around us,
00:37
and today I want to tell you a few stories
00:40
of animal complexity.
00:42
But first, what do we call complexity?
00:44
What is complex?
00:47
Well, complex is not complicated.
00:49
Something complicated comprises many small parts,
00:52
all different, and each of them
00:56
has its own precise role in the machinery.
00:58
On the opposite, a complex system
01:01
is made of many, many similar parts,
01:04
and it is their interaction
01:07
that produces a globally coherent behavior.
01:09
Complex systems have many interacting parts
01:12
which behave according to simple, individual rules,
01:16
and this results in emergent properties.
01:19
The behavior of the system as a whole
01:23
cannot be predicted
01:25
from the individual rules only.
01:26
As Aristotle wrote,
01:28
the whole is greater than the sum of its parts.
01:30
But from Aristotle, let's move onto
01:33
a more concrete example of complex systems.
01:36
These are Scottish terriers.
01:39
In the beginning, the system is disorganized.
01:41
Then comes a perturbation: milk.
01:45
Every individual starts pushing in one direction
01:49
and this is what happens.
01:53
The pinwheel is an emergent property
01:56
of the interactions between puppies
01:59
whose only rule is to try to keep access to the milk
02:01
and therefore to push in a random direction.
02:05
So it's all about finding the simple rules
02:08
from which complexity emerges.
02:12
I call this simplifying complexity,
02:15
and it's what we do at the chair of systems design
02:18
at ETH Zurich.
02:20
We collect data on animal populations,
02:22
analyze complex patterns, try to explain them.
02:26
It requires physicists who work with biologists,
02:30
with mathematicians and computer scientists,
02:32
and it is their interaction that produces
02:35
cross-boundary competence
02:38
to solve these problems.
02:40
So again, the whole is greater
02:41
than the sum of the parts.
02:43
In a way, collaboration
02:45
is another example of a complex system.
02:47
And you may be asking yourself
02:50
which side I'm on, biology or physics?
02:52
In fact, it's a little different,
02:55
and to explain, I need to tell you
02:57
a short story about myself.
02:59
When I was a child,
03:01
I loved to build stuff, to
create complicated machines.
03:03
So I set out to study electrical engineering
03:07
and robotics,
03:10
and my end-of-studies project
03:11
was about building a robot called ER-1 --
03:13
it looked like this—
03:16
that would collect information from its environment
03:18
and proceed to follow a white line on the ground.
03:21
It was very, very complicated,
03:24
but it worked beautifully in our test room,
03:26
and on demo day, professors had
assembled to grade the project.
03:29
So we took ER-1 to the evaluation room.
03:33
It turned out, the light in that room
03:36
was slightly different.
03:38
The robot's vision system got confused.
03:40
At the first bend in the line,
03:42
it left its course, and crashed into a wall.
03:44
We had spent weeks building it,
03:48
and all it took to destroy it
03:50
was a subtle change in the color of the light
03:52
in the room.
03:54
That's when I realized that
03:56
the more complicated you make a machine,
03:57
the more likely that it will fail
04:00
due to something absolutely unexpected.
04:02
And I decided that, in fact,
04:04
I didn't really want to create complicated stuff.
04:06
I wanted to understand complexity,
04:09
the complexity of the world around us
04:12
and especially in the animal kingdom.
04:14
Which brings us to bats.
04:16
Bechstein's bats are a common
species of European bats.
04:20
They are very social animals.
04:23
Mostly they roost, or sleep, together.
04:24
And they live in maternity colonies,
04:27
which means that every spring,
04:29
the females meet after the winter hibernation,
04:31
and they stay together for about six months
04:34
to rear their young,
04:36
and they all carry a very small chip,
04:39
which means that every time one of them
04:41
enters one of these specially equipped bat boxes,
04:43
we know where she is,
04:46
and more importantly,
04:48
we know with whom she is.
04:49
So I study roosting associations in bats,
04:52
and this is what it looks like.
04:55
During the day, the bats roost
04:58
in a number of sub-groups in different boxes.
05:00
It could be that on one day,
05:03
the colony is split between two boxes,
05:04
but on another day,
05:07
it could be together in a single box,
05:08
or split between three or more boxes,
05:10
and that all seems rather erratic, really.
05:13
It's called fission-fusion dynamics,
05:15
the property for an animal group
05:19
of regularly splitting and merging
05:20
into different subgroups.
05:23
So what we do is take all these data
05:24
from all these different days
05:27
and pool them together
05:28
to extract a long-term association pattern
05:30
by applying techniques with network analysis
05:33
to get a complete picture
05:35
of the social structure of the colony.
05:37
Okay? So that's what this picture looks like.
05:39
In this network, all the circles
05:44
are nodes, individual bats,
05:46
and the lines between them
05:49
are social bonds, associations between individuals.
05:50
It turns out this is a very interesting picture.
05:54
This bat colony is organized
05:57
in two different communities
05:59
which cannot be predicted
06:00
from the daily fission-fusion dynamics.
06:02
We call them cryptic social units.
06:05
Even more interesting, in fact:
06:08
Every year, around October,
06:10
the colony splits up,
06:12
and all bats hibernate separately,
06:14
but year after year,
06:16
when the bats come together again in the spring,
06:18
the communities stay the same.
06:21
So these bats remember their friends
06:23
for a really long time.
06:26
With a brain the size of a peanut,
06:28
they maintain individualized,
06:30
long-term social bonds,
06:33
We didn't know that was possible.
06:35
We knew that primates
06:36
and elephants and dolphins could do that,
06:38
but compared to bats, they have huge brains.
06:41
So how could it be
06:43
that the bats maintain this complex,
06:46
stable social structure
06:48
with such limited cognitive abilities?
06:49
And this is where complexity brings an answer.
06:53
To understand this system,
06:56
we built a computer model of roosting,
06:58
based on simple, individual rules,
07:01
and simulated thousands and thousands of days
07:03
in the virtual bat colony.
07:05
It's a mathematical model,
07:07
but it's not complicated.
07:09
What the model told us is that, in a nutshell,
07:11
each bat knows a few other colony members
07:14
as her friends, and is just slightly more likely
07:18
to roost in a box with them.
07:20
Simple, individual rules.
07:23
This is all it takes to explain
07:25
the social complexity of these bats.
07:27
But it gets better.
07:29
Between 2010 and 2011,
07:31
the colony lost more than two thirds of its members,
07:34
probably due to the very cold winter.
07:37
The next spring, it didn't form two communities
07:40
like every year,
07:43
which may have led the whole colony to die
07:45
because it had become too small.
07:47
Instead, it formed a single, cohesive social unit,
07:49
which allowed the colony to survive that season
07:54
and thrive again in the next two years.
07:57
What we know is that the bats
08:00
are not aware that their colony is doing this.
08:02
All they do is follow simple association rules,
08:05
and from this simplicity
08:08
emerges social complexity
08:10
which allows the colony to be resilient
08:12
against dramatic changes
in the population structure.
08:15
And I find this incredible.
08:18
Now I want to tell you another story,
08:21
but for this we have to travel from Europe
08:23
to the Kalahari Desert in South Africa.
08:24
This is where meerkats live.
08:27
I'm sure you know meerkats.
08:29
They're fascinating creatures.
08:31
They live in groups with a
very strict social hierarchy.
08:33
There is one dominant pair,
08:36
and many subordinates,
08:37
some acting as sentinels,
08:39
some acting as babysitters,
08:41
some teaching pups, and so on.
08:42
What we do is put very small GPS collars
08:44
on these animals
08:47
to study how they move together,
08:49
and what this has to do with their social structure.
08:50
And there's a very interesting example
08:54
of collective movement in meerkats.
08:56
In the middle of the reserve which they live in
08:58
lies a road.
09:01
On this road there are cars, so it's dangerous.
09:02
But the meerkats have to cross it
09:05
to get from one feeding place to another.
09:07
So we asked, how exactly do they do this?
09:10
We found that the dominant female
09:15
is mostly the one who leads the group to the road,
09:17
but when it comes to crossing it, crossing the road,
09:19
she gives way to the subordinates,
09:23
a manner of saying,
09:25
"Go ahead, tell me if it's safe."
09:27
What I didn't know, in fact,
09:29
was what rules in their behavior the meerkats follow
09:31
for this change at the edge of the group to happen
09:34
and if simple rules were sufficient to explain it.
09:37
So I built a model, a model of simulated meerkats
09:41
crossing a simulated road.
09:45
It's a simplistic model.
09:47
Moving meerkats are like random particles
09:49
whose unique rule is one of alignment.
09:52
They simply move together.
09:54
When these particles get to the road,
09:56
they sense some kind of obstacle,
09:59
and they bounce against it.
10:01
The only difference
10:03
between the dominant female, here in red,
10:05
and the other individuals,
10:07
is that for her, the height of the obstacle,
10:08
which is in fact the risk perceived from the road,
10:11
is just slightly higher,
10:13
and this tiny difference
10:15
in the individual's rule of movement
10:17
is sufficient to explain what we observe,
10:19
that the dominant female
10:21
leads her group to the road
10:24
and then gives way to the others
10:25
for them to cross first.
10:27
George Box, who was an English statistician,
10:30
once wrote, "All models are false,
10:33
but some models are useful."
10:36
And in fact, this model is obviously false,
10:38
because in reality, meerkats are
anything but random particles.
10:41
But it's also useful,
10:45
because it tells us that extreme simplicity
10:47
in movement rules at the individual level
10:50
can result in a great deal of complexity
10:53
at the level of the group.
10:55
So again, that's simplifying complexity.
10:57
I would like to conclude
11:01
on what this means for the whole species.
11:03
When the dominant female
11:06
gives way to a subordinate,
11:07
it's not out of courtesy.
11:09
In fact, the dominant female
11:11
is extremely important for the cohesion of the group.
11:13
If she dies on the road, the whole group is at risk.
11:15
So this behavior of risk avoidance
11:19
is a very old evolutionary response.
11:21
These meerkats are replicating an evolved tactic
11:24
that is thousands of generations old,
11:28
and they're adapting it to a modern risk,
11:30
in this case a road built by humans.
11:32
They adapt very simple rules,
11:36
and the resulting complex behavior
11:38
allows them to resist human encroachment
11:40
into their natural habitat.
11:43
In the end,
11:46
it may be bats which change their social structure
11:47
in response to a population crash,
11:50
or it may be meerkats
11:52
who show a novel adaptation to a human road,
11:54
or it may be another species.
11:57
My message here -- and it's not a complicated one,
12:00
but a simple one of wonder and hope --
12:03
my message here is that animals
12:05
show extraordinary social complexity,
12:08
and this allows them to adapt
12:11
and respond to changes in their environment.
12:13
In three words, in the animal kingdom,
12:17
simplicity leads to complexity
12:20
which leads to resilience.
12:22
Thank you.
12:24
(Applause)
12:26
Dania Gerhardt: Thank you very much, Nicolas,
12:42
for this great start. Little bit nervous?
12:44
Nicolas Perony: I'm okay, thanks.
12:47
DG: Okay, great. I'm sure a
lot of people in the audience
12:49
somehow tried to make associations
12:51
between the animals you were talking about --
12:53
the bats, meerkats -- and humans.
12:55
You brought some examples:
12:57
The females are the social ones,
12:58
the females are the dominant ones,
13:00
I'm not sure who thinks how.
13:02
But is it okay to do these associations?
13:03
Are there stereotypes you can confirm in this regard
13:06
that can be valid across all species?
13:09
NP: Well, I would say there are also
13:12
counter-examples to these stereotypes.
13:14
For examples, in sea horses or in koalas, in fact,
13:16
it is the males who take care of the young always.
13:19
And the lesson is that it's often difficult,
13:23
and sometimes even a bit dangerous,
13:28
to draw parallels between humans and animals.
13:30
So that's it.
13:32
DG: Okay. Thank you very much for this great start.
13:34
Thank you, Nicolas Perony.
13:37

sponsored links

Nicolas Perony - Animal scientist
Nicolas Perony models the movement of animal groups to understand: what is the individual behavior that guides the behavior of the larger society?

Why you should listen

Nicolas Perony started his career as a roboticist. But after one of his robots -- which was designed to follow a white line -- destroyed itself because of a lighting snafu on demo day, he realized that he was less interested in creating complicated robots and more interested in studying the complexity that already exists out there in the animal kingdom. He quickly changed course and is now a quantitative scientist at the Chair of Systems Design at ETH Zurich, where he studies the structure and dynamics of animal societies.

Perony conducts his research by placing GPS collars on animals like Bechstein's bats and meerkats, and studying the spacial data of the group. He creates models of the movement over time to see patterns. He then tries to ascertain at the simple rules that individuals in the animal group seem to be following that, when done en masse, result in the larger flow. In other words, he looks at the underlying mechanics that lead to the collective movement of animal groups.

sponsored links

If you need translations, you can install "Google Translate" extension into your Chrome Browser.
Furthermore, you can change playback rate by installing "Video Speed Controller" extension.

Data provided by TED.

This website is owned and operated by Tokyo English Network.
The developer's blog is here.