Puppies! Now that I’ve got your attention, complexity theory
-
0:03 - 0:05Science,
-
0:05 - 0:08science has allowed us to know so much
-
0:08 - 0:11about the far reaches of the universe,
-
0:11 - 0:14which is at the same time tremendously important
-
0:14 - 0:16and extremely remote,
-
0:16 - 0:19and yet much, much closer,
-
0:19 - 0:21much more directly related to us,
-
0:21 - 0:23there are many things we don't really understand.
-
0:23 - 0:25And one of them is the extraordinary
-
0:25 - 0:29social complexity of the animals around us,
-
0:29 - 0:31and today I want to tell you a few stories
-
0:31 - 0:33of animal complexity.
-
0:33 - 0:36But first, what do we call complexity?
-
0:36 - 0:38What is complex?
-
0:38 - 0:41Well, complex is not complicated.
-
0:41 - 0:44Something complicated comprises many small parts,
-
0:44 - 0:47all different, and each of them
-
0:47 - 0:50has its own precise role in the machinery.
-
0:50 - 0:53On the opposite, a complex system
-
0:53 - 0:55is made of many, many similar parts,
-
0:55 - 0:57and it is their interaction
-
0:57 - 1:01that produces a globally coherent behavior.
-
1:01 - 1:05Complex systems have many interacting parts
-
1:05 - 1:08which behave according to simple, individual rules,
-
1:08 - 1:11and this results in emergent properties.
-
1:11 - 1:13The behavior of the system as a whole
-
1:13 - 1:15cannot be predicted
-
1:15 - 1:17from the individual rules only.
-
1:17 - 1:19As Aristotle wrote,
-
1:19 - 1:22the whole is greater than the sum of its parts.
-
1:22 - 1:24But from Aristotle, let's move onto
-
1:24 - 1:28a more concrete example of complex systems.
-
1:28 - 1:30These are Scottish terriers.
-
1:30 - 1:34In the beginning, the system is disorganized.
-
1:34 - 1:38Then comes a perturbation: milk.
-
1:38 - 1:41Every individual starts pushing in one direction
-
1:41 - 1:45and this is what happens.
-
1:45 - 1:48The pinwheel is an emergent property
-
1:48 - 1:50of the interactions between puppies
-
1:50 - 1:53whose only rule is to try to keep access to the milk
-
1:53 - 1:57and therefore to push in a random direction.
-
1:57 - 2:01So it's all about finding the simple rules
-
2:01 - 2:04from which complexity emerges.
-
2:04 - 2:07I call this simplifying complexity,
-
2:07 - 2:09and it's what we do at the chair of systems design
-
2:09 - 2:11at ETH Zurich.
-
2:11 - 2:15We collect data on animal populations,
-
2:15 - 2:18analyze complex patterns, try to explain them.
-
2:18 - 2:21It requires physicists who work with biologists,
-
2:21 - 2:24with mathematicians and computer scientists,
-
2:24 - 2:26and it is their interaction that produces
-
2:26 - 2:28cross-boundary competence
-
2:28 - 2:30to solve these problems.
-
2:30 - 2:32So again, the whole is greater
-
2:32 - 2:33than the sum of the parts.
-
2:33 - 2:36In a way, collaboration
-
2:36 - 2:39is another example of a complex system.
-
2:39 - 2:41And you may be asking yourself
-
2:41 - 2:44which side I'm on, biology or physics?
-
2:44 - 2:46In fact, it's a little different,
-
2:46 - 2:47and to explain, I need to tell you
-
2:47 - 2:50a short story about myself.
-
2:50 - 2:52When I was a child,
-
2:52 - 2:56I loved to build stuff, to
create complicated machines. -
2:56 - 2:58So I set out to study electrical engineering
-
2:58 - 3:00and robotics,
-
3:00 - 3:02and my end-of-studies project
-
3:02 - 3:05was about building a robot called ER-1 --
-
3:05 - 3:07it looked like this—
-
3:07 - 3:09that would collect information from its environment
-
3:09 - 3:13and proceed to follow a white line on the ground.
-
3:13 - 3:15It was very, very complicated,
-
3:15 - 3:18but it worked beautifully in our test room,
-
3:18 - 3:22and on demo day, professors had
assembled to grade the project. -
3:22 - 3:25So we took ER-1 to the evaluation room.
-
3:25 - 3:27It turned out, the light in that room
-
3:27 - 3:29was slightly different.
-
3:29 - 3:31The robot's vision system got confused.
-
3:31 - 3:33At the first bend in the line,
-
3:33 - 3:36it left its course, and crashed into a wall.
-
3:36 - 3:39We had spent weeks building it,
-
3:39 - 3:40and all it took to destroy it
-
3:40 - 3:43was a subtle change in the color of the light
-
3:43 - 3:44in the room.
-
3:44 - 3:46That's when I realized that
-
3:46 - 3:48the more complicated you make a machine,
-
3:48 - 3:50the more likely that it will fail
-
3:50 - 3:53due to something absolutely unexpected.
-
3:53 - 3:55And I decided that, in fact,
-
3:55 - 3:58I didn't really want to create complicated stuff.
-
3:58 - 4:01I wanted to understand complexity,
-
4:01 - 4:03the complexity of the world around us
-
4:03 - 4:05and especially in the animal kingdom.
-
4:05 - 4:08Which brings us to bats.
-
4:08 - 4:11Bechstein's bats are a common
species of European bats. -
4:11 - 4:13They are very social animals.
-
4:13 - 4:16Mostly they roost, or sleep, together.
-
4:16 - 4:18And they live in maternity colonies,
-
4:18 - 4:19which means that every spring,
-
4:19 - 4:23the females meet after the winter hibernation,
-
4:23 - 4:25and they stay together for about six months
-
4:25 - 4:27to rear their young,
-
4:27 - 4:30and they all carry a very small chip,
-
4:30 - 4:32which means that every time one of them
-
4:32 - 4:35enters one of these specially equipped bat boxes,
-
4:35 - 4:37we know where she is,
-
4:37 - 4:38and more importantly,
-
4:38 - 4:40we know with whom she is.
-
4:40 - 4:44So I study roosting associations in bats,
-
4:44 - 4:46and this is what it looks like.
-
4:46 - 4:49During the day, the bats roost
-
4:49 - 4:51in a number of sub-groups in different boxes.
-
4:51 - 4:53It could be that on one day,
-
4:53 - 4:55the colony is split between two boxes,
-
4:55 - 4:57but on another day,
-
4:57 - 4:59it could be together in a single box,
-
4:59 - 5:01or split between three or more boxes,
-
5:01 - 5:04and that all seems rather erratic, really.
-
5:04 - 5:07It's called fission-fusion dynamics,
-
5:07 - 5:09the property for an animal group
-
5:09 - 5:11of regularly splitting and merging
-
5:11 - 5:13into different subgroups.
-
5:13 - 5:15So what we do is take all these data
-
5:15 - 5:17from all these different days
-
5:17 - 5:19and pool them together
-
5:19 - 5:21to extract a long-term association pattern
-
5:21 - 5:24by applying techniques with network analysis
-
5:24 - 5:25to get a complete picture
-
5:25 - 5:28of the social structure of the colony.
-
5:28 - 5:32Okay? So that's what this picture looks like.
-
5:32 - 5:35In this network, all the circles
-
5:35 - 5:37are nodes, individual bats,
-
5:37 - 5:39and the lines between them
-
5:39 - 5:43are social bonds, associations between individuals.
-
5:43 - 5:45It turns out this is a very interesting picture.
-
5:45 - 5:47This bat colony is organized
-
5:47 - 5:49in two different communities
-
5:49 - 5:51which cannot be predicted
-
5:51 - 5:53from the daily fission-fusion dynamics.
-
5:53 - 5:57We call them cryptic social units.
-
5:57 - 5:58Even more interesting, in fact:
-
5:58 - 6:01Every year, around October,
-
6:01 - 6:02the colony splits up,
-
6:02 - 6:05and all bats hibernate separately,
-
6:05 - 6:06but year after year,
-
6:06 - 6:10when the bats come together again in the spring,
-
6:10 - 6:12the communities stay the same.
-
6:12 - 6:15So these bats remember their friends
-
6:15 - 6:17for a really long time.
-
6:17 - 6:19With a brain the size of a peanut,
-
6:19 - 6:21they maintain individualized,
-
6:21 - 6:23long-term social bonds,
-
6:23 - 6:25We didn't know that was possible.
-
6:25 - 6:27We knew that primates
-
6:27 - 6:29and elephants and dolphins could do that,
-
6:29 - 6:32but compared to bats, they have huge brains.
-
6:32 - 6:34So how could it be
-
6:34 - 6:36that the bats maintain this complex,
-
6:36 - 6:38stable social structure
-
6:38 - 6:42with such limited cognitive abilities?
-
6:42 - 6:45And this is where complexity brings an answer.
-
6:45 - 6:47To understand this system,
-
6:47 - 6:49we built a computer model of roosting,
-
6:49 - 6:52based on simple, individual rules,
-
6:52 - 6:54and simulated thousands and thousands of days
-
6:54 - 6:56in the virtual bat colony.
-
6:56 - 6:58It's a mathematical model,
-
6:58 - 7:00but it's not complicated.
-
7:00 - 7:03What the model told us is that, in a nutshell,
-
7:03 - 7:06each bat knows a few other colony members
-
7:06 - 7:09as her friends, and is just slightly more likely
-
7:09 - 7:11to roost in a box with them.
-
7:11 - 7:14Simple, individual rules.
-
7:14 - 7:15This is all it takes to explain
-
7:15 - 7:18the social complexity of these bats.
-
7:18 - 7:20But it gets better.
-
7:20 - 7:22Between 2010 and 2011,
-
7:22 - 7:26the colony lost more than two thirds of its members,
-
7:26 - 7:29probably due to the very cold winter.
-
7:29 - 7:32The next spring, it didn't form two communities
-
7:32 - 7:33like every year,
-
7:33 - 7:35which may have led the whole colony to die
-
7:35 - 7:38because it had become too small.
-
7:38 - 7:43Instead, it formed a single, cohesive social unit,
-
7:43 - 7:46which allowed the colony to survive that season
-
7:46 - 7:49and thrive again in the next two years.
-
7:49 - 7:51What we know is that the bats
-
7:51 - 7:53are not aware that their colony is doing this.
-
7:53 - 7:57All they do is follow simple association rules,
-
7:57 - 7:58and from this simplicity
-
7:58 - 8:01emerges social complexity
-
8:01 - 8:04which allows the colony to be resilient
-
8:04 - 8:07against dramatic changes
in the population structure. -
8:07 - 8:09And I find this incredible.
-
8:09 - 8:11Now I want to tell you another story,
-
8:11 - 8:13but for this we have to travel from Europe
-
8:13 - 8:16to the Kalahari Desert in South Africa.
-
8:16 - 8:18This is where meerkats live.
-
8:18 - 8:20I'm sure you know meerkats.
-
8:20 - 8:22They're fascinating creatures.
-
8:22 - 8:25They live in groups with a
very strict social hierarchy. -
8:25 - 8:26There is one dominant pair,
-
8:26 - 8:27and many subordinates,
-
8:27 - 8:29some acting as sentinels,
-
8:29 - 8:31some acting as babysitters,
-
8:31 - 8:32some teaching pups, and so on.
-
8:32 - 8:36What we do is put very small GPS collars
-
8:36 - 8:37on these animals
-
8:37 - 8:39to study how they move together,
-
8:39 - 8:43and what this has to do with their social structure.
-
8:43 - 8:44And there's a very interesting example
-
8:44 - 8:47of collective movement in meerkats.
-
8:47 - 8:49In the middle of the reserve which they live in
-
8:49 - 8:51lies a road.
-
8:51 - 8:54On this road there are cars, so it's dangerous.
-
8:54 - 8:56But the meerkats have to cross it
-
8:56 - 8:59to get from one feeding place to another.
-
8:59 - 9:03So we asked, how exactly do they do this?
-
9:03 - 9:05We found that the dominant female
-
9:05 - 9:08is mostly the one who leads the group to the road,
-
9:08 - 9:11but when it comes to crossing it, crossing the road,
-
9:11 - 9:14she gives way to the subordinates,
-
9:14 - 9:15a manner of saying,
-
9:15 - 9:18"Go ahead, tell me if it's safe."
-
9:18 - 9:20What I didn't know, in fact,
-
9:20 - 9:23was what rules in their behavior the meerkats follow
-
9:23 - 9:26for this change at the edge of the group to happen
-
9:26 - 9:30and if simple rules were sufficient to explain it.
-
9:30 - 9:34So I built a model, a model of simulated meerkats
-
9:34 - 9:36crossing a simulated road.
-
9:36 - 9:37It's a simplistic model.
-
9:37 - 9:40Moving meerkats are like random particles
-
9:40 - 9:42whose unique rule is one of alignment.
-
9:42 - 9:45They simply move together.
-
9:45 - 9:48When these particles get to the road,
-
9:48 - 9:50they sense some kind of obstacle,
-
9:50 - 9:52and they bounce against it.
-
9:52 - 9:53The only difference
-
9:53 - 9:55between the dominant female, here in red,
-
9:55 - 9:57and the other individuals,
-
9:57 - 9:59is that for her, the height of the obstacle,
-
9:59 - 10:02which is in fact the risk perceived from the road,
-
10:02 - 10:04is just slightly higher,
-
10:04 - 10:05and this tiny difference
-
10:05 - 10:07in the individual's rule of movement
-
10:07 - 10:10is sufficient to explain what we observe,
-
10:10 - 10:12that the dominant female
-
10:12 - 10:14leads her group to the road
-
10:14 - 10:15and then gives way to the others
-
10:15 - 10:18for them to cross first.
-
10:18 - 10:22George Box, who was an English statistician,
-
10:22 - 10:25once wrote, "All models are false,
-
10:25 - 10:27but some models are useful."
-
10:27 - 10:30And in fact, this model is obviously false,
-
10:30 - 10:34because in reality, meerkats are
anything but random particles. -
10:34 - 10:36But it's also useful,
-
10:36 - 10:38because it tells us that extreme simplicity
-
10:38 - 10:42in movement rules at the individual level
-
10:42 - 10:44can result in a great deal of complexity
-
10:44 - 10:46at the level of the group.
-
10:46 - 10:50So again, that's simplifying complexity.
-
10:50 - 10:52I would like to conclude
-
10:52 - 10:54on what this means for the whole species.
-
10:54 - 10:56When the dominant female
-
10:56 - 10:58gives way to a subordinate,
-
10:58 - 11:00it's not out of courtesy.
-
11:00 - 11:01In fact, the dominant female
-
11:01 - 11:04is extremely important for the cohesion of the group.
-
11:04 - 11:07If she dies on the road, the whole group is at risk.
-
11:07 - 11:10So this behavior of risk avoidance
-
11:10 - 11:12is a very old evolutionary response.
-
11:12 - 11:16These meerkats are replicating an evolved tactic
-
11:16 - 11:18that is thousands of generations old,
-
11:18 - 11:21and they're adapting it to a modern risk,
-
11:21 - 11:24in this case a road built by humans.
-
11:24 - 11:27They adapt very simple rules,
-
11:27 - 11:29and the resulting complex behavior
-
11:29 - 11:32allows them to resist human encroachment
-
11:32 - 11:34into their natural habitat.
-
11:34 - 11:36In the end,
-
11:36 - 11:39it may be bats which change their social structure
-
11:39 - 11:41in response to a population crash,
-
11:41 - 11:43or it may be meerkats
-
11:43 - 11:46who show a novel adaptation to a human road,
-
11:46 - 11:48or it may be another species.
-
11:48 - 11:51My message here -- and it's not a complicated one,
-
11:51 - 11:54but a simple one of wonder and hope --
-
11:54 - 11:57my message here is that animals
-
11:57 - 12:00show extraordinary social complexity,
-
12:00 - 12:02and this allows them to adapt
-
12:02 - 12:05and respond to changes in their environment.
-
12:05 - 12:08In three words, in the animal kingdom,
-
12:08 - 12:11simplicity leads to complexity
-
12:11 - 12:12which leads to resilience.
-
12:12 - 12:15Thank you.
-
12:15 - 12:21(Applause)
-
12:31 - 12:33Dania Gerhardt: Thank you very much, Nicolas,
-
12:33 - 12:36for this great start. Little bit nervous?
-
12:36 - 12:38Nicolas Perony: I'm okay, thanks.
-
12:38 - 12:40DG: Okay, great. I'm sure a
lot of people in the audience -
12:40 - 12:42somehow tried to make associations
-
12:42 - 12:44between the animals you were talking about --
-
12:44 - 12:46the bats, meerkats -- and humans.
-
12:46 - 12:47You brought some examples:
-
12:47 - 12:49The females are the social ones,
-
12:49 - 12:50the females are the dominant ones,
-
12:50 - 12:52I'm not sure who thinks how.
-
12:52 - 12:55But is it okay to do these associations?
-
12:55 - 12:58Are there stereotypes you can confirm in this regard
-
12:58 - 13:01that can be valid across all species?
-
13:01 - 13:03NP: Well, I would say there are also
-
13:03 - 13:05counter-examples to these stereotypes.
-
13:05 - 13:08For examples, in sea horses or in koalas, in fact,
-
13:08 - 13:11it is the males who take care of the young always.
-
13:11 - 13:17And the lesson is that it's often difficult,
-
13:17 - 13:18and sometimes even a bit dangerous,
-
13:18 - 13:21to draw parallels between humans and animals.
-
13:21 - 13:23So that's it.
-
13:23 - 13:26DG: Okay. Thank you very much for this great start.
-
13:26 - 13:28Thank you, Nicolas Perony.
- Title:
- Puppies! Now that I’ve got your attention, complexity theory
- Speaker:
- Nicolas Perony
- Description:
-
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.
- Video Language:
- English
- Team:
- closed TED
- Project:
- TEDTalks
- Duration:
- 13:45
Morton Bast edited English subtitles for Puppies! Now that I’ve got your attention, complexity theory | ||
Morton Bast edited English subtitles for Puppies! Now that I’ve got your attention, complexity theory | ||
Morton Bast approved English subtitles for Puppies! Now that I’ve got your attention, complexity theory | ||
Morton Bast edited English subtitles for Puppies! Now that I’ve got your attention, complexity theory | ||
Morton Bast edited English subtitles for Puppies! Now that I’ve got your attention, complexity theory | ||
Madeleine Aronson accepted English subtitles for Puppies! Now that I’ve got your attention, complexity theory | ||
Madeleine Aronson edited English subtitles for Puppies! Now that I’ve got your attention, complexity theory | ||
Joseph Geni edited English subtitles for Puppies! Now that I’ve got your attention, complexity theory |