The secret social lives of bats: Nicolas Perony at TEDxZurich

"Simplicity leads to complexity which leads to resilience," says Nicolas Perony, an engineer turned into a complex systems scientist who studies the social structure of animal groups. Perony shares his findings on the resilience of bat social networks and the adaptation of Kalahari meerkats to human encroachment, and gives us a peek into the wonderful hidden complexity of the animal kingdom.

Science.

Science has allowed us to know so much

about the far reaches of the universe,

which is at the same time

tremendously important
and extremely remote.

And yet, much closer,

much more directly related to us,

there are many things
we don't fully understand.

One of them is the extraordinary
social complexity

of the animals around us.

And today I want to tell you
a few stories of animal complexity.

But first, what do we call complexity?

What is complex?

Well, complex is not complicated.

Something complicated
comprises many small parts,

all different, and each of them

has its own precise role
in the machinery.

On the opposite, a complex system

is made of many, many similar parts

and it is their interaction

that produces the globally
coherent behavior.

Complex systems have
many interacting parts,

which behave according
to simple individual rules

and this results in emergent properties.

The behavior of the system as a whole

cannot be predicted
from the individual rules only.

As Aristotle wrote:

The whole is greater
than the sum of its parts.

But, from Aristotle, let's move onto

a more concrete example of complex system.

These are Scottish Terriers.

In the beginning,
the system is disorganised.

Then comes a perturbation: milk.

Every individual starts pushing

in one direction -- (Laughter)

-- and this is what happens.

The pinwheel is an emergent property

of the interactions between puppies,

whose only rule is to try
and keep access to the milk

and therefore to push
in a random direction.

So it's all about finding
the simple rules

from which complexity emerges.

I call this "simplifying complexity".

And that is what we do
at the Chair of Systems Design

at ETH Zurich.

We collect data on animal populations,

analyze complex patterns,
try to explain them.

It requires physicists
who work with biologists,

with mathematicians
and computer scientists

and it is their interaction

that produces cross-boundary competence

to solve these problems.

So again, the whole is greater
than the sum of the parts.

In a way, collaboration is another
example of a complex system.

And you may be asking yourself
which side I'm on.

Biology or Physics?

In fact, it's a little different.

To explain, I need to tell you
a short story about myself.

When I was a child,
I loved to build stuff,

to create complicated machines.

So I set out to study electrical
engineering and robotics.

And my end-of-studies project

was about building a robot called ER1 --

which looked like this --

that would collect information
from its environment

and proceed to follow
a white line on the ground.

It was very, very complicated,

but it worked beautifully
in our test room.

And on demo day, professors had assembled
to grade the projects,

so we took ER1 to the evaluation room.

It turned out that the light in that room

was slightly different.

The robot's vision system got confused.

At the first bend of the line,

it left its course
and crashed into a wall.

We had spent weeks building it
and all it took to destroy it

was a subtle change in the color
of the light in the room.

That's when I realized that

the more complicated
you make a machine

the more likely it will fail

due to something absolutely unexpected.

And I decided that in fact

I did not really want
to create complicated stuff.

I wanted to understand complexity,

the complexity of the world around us

and especially in the animal kingdom,

which brings us to bats.

Bechstein's bats are a common
species of European bats.

They are very social animals.

Mostly, they roost -- or sleep -- together.

And they live in maternity colonies,
which means that, every spring,

the females meet
after the winter hibernation

and they stay together
for about 6 months

to rear their young.

And they all carry a very small chip,

which means that every time
one of them

enters one of these
specially equipped bat boxes,

we know where she is.

And more importantly,
we know with whom she is.

So I study roosting associations in bats.

And this is what it looks like.

During the day, the bats roost

in a number of subgroups
in different boxes.

It could be that, on one day,
the colony is split between two boxes.

But, on another day, it could be
together in a single box

or split between 3 or more boxes.

And that all seems rather erratic, really,

it's called fission-fusion dynamics --

-- the property for an animal group
of regularly splitting

and merging into different subgroups.

So what we do is to take all these data

from all these different days

and pool them together

to extract a long-term
association pattern

by applying techniques
of network analysis

to get a complete picture

of the social structure of the colony.

OK? So that's what
this picture looks like.

In this network, all the circles
are nodes -- individual bats --

and the lines between them
are social bonds --

associations between individuals.

It turns out this is
a very interesting picture.

This bat colony is organized
in two different communities

which cannot be predicted

from the daily fission-fusion dynamics.

We call them "cryptic" social units.

Even more interesting, in fact:
every year around October,

the colony splits up and all bats
hibernate separately.

But year after year,

when the bats come together
again in the spring,

the communities stay the same.

So these bats remember their friends
for a really long time.

With a brain of the size of a peanut,

they maintain individualized
long-term social bonds.

We didn't know that was possible.

We knew that primates and elephants
and dolphins could do that

but compared to bats
they have huge brains.

So, how could it be that the bats

maintain this complex
stable social structure

with such limited cognitive abilities?

And this is where complexity
brings an answer.

To understand this system,

we built a computer model of roosting,

based on simple individual rules,

and simulated thousands
and thousands of days

in a virtual bat colony.

It's a mathematical model,

but it is not complicated.

What the model told us
is that, in a nutshell,

each bat knows a few other
colony members as her friends,

and is just slightly more likely
to roost in a box with them.

Simple, individual rules.

This is all it takes to explain

the social complexity of these bats.

But it gets better.

Between 2010 and 2011,

the colony lost more than
two thirds of its members,

probably due to the very cold winter.

The next spring, it didn't form
2 communities like every year,

which may have led
the whole colony to die

because it had become too small.

Instead, it formed a single
cohesive social unit,

which allowed the colony
to survive that season

and thrive again in the next two years.

What we know is that the bats
are not aware

that their colony is doing this.

All they do is follow
simple association rules

and from this simplicity
emerges social complexity,

which allows the colony to be resilient

against dramatic changes
in the population structure.

And I find this incredible.

Now I want to tell you another story.

But for this, we have to travel
from Europe

to the Kalahari Desert, in South Africa.

This is where meerkats live.

I am sure you know meerkats,
they are fascinating creatures.

They live in groups
with a very strict social hierarchy.

There is one dominant pair
and many subordinates,

some acting as sentinels,
some acting as babysitters,

some teaching pups and so on.

What we do is put very small GPS collars
on these animals

to study how they move together

and what this has to do
with their social structure.

And there is a very interesting example

of collective movement in meerkats.

In the middle of the reserve,

which they live in, lies a road.

On this road there are cars,
so it is dangerous.

But the meerkats have to cross it

to get from one feeding place to another.

So we asked, how exactly
do they do this?

We found out that
the dominant female

is mostly the one who leads
the group to the road,

but when it comes to crossing it,
crossing the road,

she gives way to the subordinates,

a manner of saying, you know,
"Go ahead, tell me if it's safe!".

(Laughter)

What I didn't know, in fact,

was what rules in their behavior
the meerkats follow

for this change at the edge
of the group to happen

and these simple rules were
sufficient to explain it.

So I built a model,

a model of simulated meerkats
crossing a simulated road.

It's a simplistic model.

Moving meerkats are like
random particles

whose unique rule is
one of alignment.

They simply move together.

When these particles get to the road,

they sense some kind of obstacle

and they bounce against it.

The only difference between
the dominant female -- here in red --

and the other individuals

is that, for her, the height
of the obstacle,

which is in fact the risk
perceived from the road,

is just slightly higher.

And this tiny difference
in the individual rule of movement

is sufficient to explain what we observe,

that the dominant female
leads her group to the road

and then gives way to the others
for them to cross first.

George Box, who was
an English statistician, once wrote:

"All models are false,
but some models are useful".

And in fact, this model
is obviously false,

because in reality meerkats
are anything but random particles.

But it's also useful, because it tells us

that extreme simplicity in movement rules
at the individual level

can result in a great deal of complexity

at the level of the group.

So again, that's simplifying complexity.

And I would like to conclude
on what this means

for the whole species.

When the dominant female
gives way to a subordinate,

it's not out of courtesy.

In fact, the dominant female
is extremely important

for the cohesion of the group.

If she dies on the road,
the whole group is at risk.

So this behavior of risk avoidance

is a very old evolutionary response.

These meerkats are replicating
an evolved tactic

that is thousands of generations old,

and they are adapting it
to a modern risk;

in this case, a road built by humans.

They adapt very simple rules

and the resulting complex behavior

allows them to resist
human encroachment

into their natural habitat.

In the end, it may be bats
who change their social structure

in response to a population crash.

Or it may be meerkats who show
a novel adaptation to a human road.

Or it may be another species.

My message here --
and it's not a complicated one --

but a simple one of wonder and hope.

My message here is that animals
show extraordinary social complexity

and this allows them to adapt
and respond to changes

in their environment.

In three words: in the animal kingdom,

simplicity leads to complexity,
which leads to resilience.

Thank you.

(Applause)