Wednesday, April 25, 2012

Can a Horde of Idiots be a Genius?

Let’s face it: The typical individual is not that bright. Just check out these human specimens:

Yet somehow, if you get enough numbskulls together, the group can make some pretty intelligent decisions. We’ve seen this in a wide variety of organisms facing a number of different challenges.

In a brilliant series of studies, Jean-Louis Deneubourg, a professor at the Free University of Brussels, and his colleagues tested the abilities of Argentine ants (a common dark-brown ant species) to collectively solve foraging problems. In one of these studies, the ants were provided with a bridge that connected the nest to a food source. This bridge split and fused in two places (like eyeglass frames), but at each split one branch was shorter than the other, resulting in a single shortest-path and multiple longer paths. After a few minutes, explorers crossed the bridge (by a meandering path) and discovered the food. This recruited foragers, each of which chose randomly between the short and the long branch at each split. Then suddenly, the foragers all started to prefer the shortest route. How did they do that?

This figure from the Goss et al 1989 paper in Naturwissemschaften shows (a) the design of a single module, (b) ants scattered on the bridge after 4 minutes (I promise they’re there), and (c) ants mostly on the shortest path after 8 minutes

You can think of it this way: a single individual often tries to make decisions based on the uncertain information available to it. But if you have a group of individuals, they will likely each have information that differs somewhat from the information of others in the group. If they each make a decision based on their own information alone, they will likely result in a number of poor decisions and a few good ones. But if they can each base their decisions on the accumulation of all of the information of the group, they stand a much better chance of making a good decision. The more information accumulated, the more likely they are to make the best possible decision.

In the case of the Argentine ant, the accumulated information takes the form of pheromone trails. Argentine ants lay pheromone trails both when leaving the nest and when returning to the nest. Ants that are lucky enough to take a shorter foraging route return to the nest sooner, increasing the pheromone concentration of the route each way. In this way, shorter routes develop more concentrated pheromone trails faster, which attract more ants, which further increase pheromone concentration of the shortest routes. In this way, an ant colony can make an intelligent decision (take the shortest foraging route) without any individual doing anything more intelligent than following a simple rule (follow the strongest pheromone signal).


Home is where the heart is. Photo of a bee swarm by Tom Seeley
Honeybee colonies also solve complicated tasks with the use of communication. Tom Seeley at Cornell University and his colleagues have investigated the honeybee group decision-making process of finding a new home. When a colony outgrows their hive, hundreds of scouts will go in search of a suitable new home, preferably one that is high off the ground with a south-facing entrance and room to grow. If a scout finds such a place, she returns to the colony and performs a waggle dance, a dance in which her body position and movements encode the directions to her site and her dancing vigor relates to how awesome she thinks the site is. 


Some scouts that see her dance may be persuaded to follow her directions and check out the site for themselves, and if impressed, may return to the hive and perform waggle dances too. Or they may follow another scout’s directions to a different site or even strike out on their own. Eventually, the majority of the scouts are all dancing the same vigorous dance. But interestingly, few scouts ever visit more than one site. Better sites simply receive more vigorous “dance-votes” and then attract more scouts to do the same. Like ants in search of a foraging path, the intensity of the collective signal drives the group towards the best decision. Once a quorum is reached, the honeybees fly off together to their new home.

But groups can develop better solutions than individuals even without communication. Gaia Dell’Ariccia at the University of Zurich in Switzerland and her colleagues explored homing pigeon navigation by placing GPS trackers on the backs of pigeons and releasing them from a familiar location either alone or in a group of six. Because they were all trained to fly home from this site, they all found their way home regardless of whether they were alone or in a group. But as a flock, the pigeons left sooner, rested less, flew faster, and took a more direct route than did the same birds when making the trip alone. By averaging the directional tendencies of everyone in the group, they were able to mutually correct the errors of each individual and follow the straightest path.

In each of these examples, each individual has limited and uncertain information, but each individual has information that may be slightly different than their neighbors’. By combining this diverse information and making a collective decision, hordes of idiots can make genius decisions.



Want to know more? Check these out:

1. Couzin, I. (2009). Collective cognition in animal groups Trends in Cognitive Sciences, 13 (1), 36-43 DOI: 10.1016/j.tics.2008.10.002

2. Goss, S., Aron, S., Deneubourg, J., & Pasteels, J. (1989). Self-organized shortcuts in the Argentine ant Naturwissenschaften, 76 (12), 579-581 DOI: 10.1007/BF00462870

3. Dussutour, A., Nicolis, S., Deneubourg, J., & Fourcassié, V. (2006). Collective decisions in ants when foraging under crowded conditions Behavioral Ecology and Sociobiology, 61 (1), 17-30 DOI: 10.1007/s00265-006-0233-x

4. List C, Elsholtz C, & Seeley TD (2009). Independence and interdependence in collective decision making: an agent-based model of nest-site choice by honeybee swarms. Philosophical transactions of the Royal Society of London. Series B, Biological sciences, 364 (1518), 755-62 PMID: 19073474

5. Dell'Ariccia, G., Dell'Omo, G., Wolfer, D., & Lipp, H. (2008). Flock flying improves pigeons' homing: GPS track analysis of individual flyers versus small groups Animal Behaviour, 76 (4), 1165-1172 DOI: 10.1016/j.anbehav.2008.05.022

6. Honeybee Democracy by Thomas Seeley

7. The Smart Swarm by Peter Miller

Wednesday, April 18, 2012

It Doesn’t Always Pay to Kill Your Siblings

A mother reed warbler feeding her "adoptive"
murderous cuckoo chick. Does she really think
this is her child? Photo by Per Harald Olsen
on Wikimedia Commons.
A woman, driven to not raise her own child, leaves her baby in another woman’s nursery, killing another baby that is there and replacing it with her own. As soon as the transplanted baby is strong enough, it slowly, methodically kills all the other children in the nursery, hording all of the adoptive mother’s attention for itself. With time, it grows needier and more demanding, running the poor adoptive mother ragged trying to care for the monstrous child that had murdered all her own children.

It may sound like the plotline of a horror flick, but it happens every year among several egg-laying species (granted, maybe without quite so much anthropomorphism and drama). Animals that lay their eggs among the egg clutches of other species and benefit from the “hosts” parenting their young are called obligate brood parasites. Like the creepy narrative above, the still-naked, recently-hatched offspring of many obligate brood parasites attack and kill the offspring of their hosts to ensure more parental attention.

Cuckoo chicks are well known to kill their “step-siblings” (the nestmates that are the chicks of their host parents) by pushing them out of the nest. Caution: Watching this BBC video can have emotional consequences:
 


The researchers observed the nests of two common shiny cowbird host species: the house wren, a much smaller host than the cowbird, and the chalk-browed mockingbird, a larger host than the cowbird. For each nest, they either placed a combination of one cowbird egg and a few host eggs, or one cowbird egg and a few artificial eggs. When the eggs hatched, this resulted in some cowbirds having host “step-siblings” and other cowbirds being “only children”. When the chicks were 4 days old and 8 days old, the researchers videotaped each nest, and measured how much food each chick was getting fed and each chick’s weight and tarsus length (that’s the length of one of the leg bones and gives a good method to compare body sizes in birds).


This is a picture of begging chicks in a parasitized nest of a
chalk-browed mockingbird, taken from a video. The chick with
the smaller, redder gape at the top of the image is the cowbird.
The other larger gapes belong to the mockingbird's own chicks.
Photo by Ros Gloag.
Overall, cowbirds grew to similar sizes at similar rates regardless of whether they were raised by mockingbird or house wren parents. But the presence or absence of nestmates had an interesting effect. Among cowbirds raised in mockingbird nests, cowbirds raised with nestmates faired far worse than cowbirds raised alone. Although mockingbird parents worked much harder and brought back more food to nests with more chicks (probably because the begging coming from the nest as a whole was much more rambunctious), cowbirds with mockingbird step-siblings had to share the provisions, whereas lone cowbird chicks got to eat everything the parents returned with. To make things worse, the mockingbird chicks grew faster than the cowbird chicks and they quickly outcompeted them. By the eighth day, cowbirds raised alone were larger and had higher survival rates than cowbirds raised with mockingbird step-siblings.

This graph shows how much food the parents gave to the shiny cowbird chick, depending on whether the chick had mockingbird or wren parents and on whether it had nestmates (mixed) or was alone. The dashed bars show the total amount of food brought back to the nest for nests with nestmates. Notice that although nests with nestmates had the most food brought to the nest (dashed bars) in all cases, the shiny cowbird chicks with mockingbird parents got more food on day 8 if they were alone than if they had nestmates (solid bars) and the shiny cowbird chicks with wren parents got more food on day 8 if they had nestmates than if they were alone (solid bars). Graph from Gloag, et. al 2011 Behavioral Ecology paper.

However, among cowbirds raised in house wren nests, cowbirds raised with wren step-siblings faired far better than cowbirds raised alone. As did the mockingbird parents, wren parents worked much harder and brought back more food to nests with more chicks. The cowbird chicks grew faster than the wren chicks, and so the cowbirds managed to eat more than their fair share of provisions. Additionally, sharing a nest with nestmates can have other benefits as well, such as helping to keep warm. As a result, cowbirds raised with wren step-siblings grew larger than cowbirds raised without wren step-siblings.

We don’t yet know whether obligate brood parasite chicks with multiple host species adjust their strategy (kill or tolerate) depending on the size of the host chicks, but this study suggests that they might. If you are interested in science, this may be a topic you could explore.

At any rate, killing your siblings or step-siblings probably isn’t a wise thing to do. They can help encourage our parents to do more for us, they can keep us warm, and when they’re not looking we can “borrow” their stuff. Besides, birds don’t have the same morals, ethics and court systems that we do.


Want to know more? Check this out:

Gloag, R., Tuero, D., Fiorini, V., Reboreda, J., & Kacelnik, A. (2011). The economics of nestmate killing in avian brood parasites: a provisions trade-off Behavioral Ecology, 23 (1), 132-140 DOI: 10.1093/beheco/arr166

Wednesday, April 11, 2012

The Social Punishment of Samantha Brick

An interesting thing happened this week in the world of collective human behavior. But before we go into that, let me ask you two questions: Have you heard of Samantha Brick? On a scale of 1 to 10, how attractive do you think she is?


Samantha Brick, a journalist, wrote an article for the Daily Mail called “'There are downsides to looking this pretty': Why women hate me for being beautiful”. Naturally, the response to hearing a story like this is, “Well, what does she look like?” Luckily for us, she graciously included many photos of herself for us to assess… and the collective assessment was apparently not what Samantha Brick had expected.

The internet and Twitter practically exploded with snarky Samantha Brick comments. The Daily Mail had to shut down the comments section on the article after collecting almost 6000 comments, mostly emphasizing how unattractive she is. A new Twitter hashtag, #samanthabrickfacts, appeared with notable one-liners such as, “James Blunt wrote 'You’re beautiful' after he briefly caught sight of Samantha Brick in a crowded place. #samanthabrickfacts” and “What was in the briefcase in Pulp Fiction? A photo of Samantha Brick. #samanthabrickfacts”. Even major media outlets in the UK (where Brick is from) and the United States picked up on the internet storm and added their own two cents (Tamron Hall on the Today Show thought it funny that “some people think she looks like Chucky the Doll”).

On a television interview on ITV about the backlash, Brick explained, “Women do not like attractive women, and that has been proved to me by the thousands of vile messages I’ve had on Twitter, the thousands of vile emails I’ve had to my personal account, the messages I’ve had on my own answer phones.” But Samantha Brick has a very different explanation for the backlash than almost everybody else. Barbara Walters on The View summed up the more popular explanation, “At the risk of being really not so nice, she’s got a problem. She’s not that beautiful. OK?”

In other words, Samantha Brick advertised herself as having a high attractiveness (with all the benefits that would include), when in fact her true attractiveness is moderate. In the animal world, this phenomenon is called dishonest signals, and it is believed to elicit social punishment in many animal species. Elizabeth Tibbetts and Amanda Izzo at the University of Michigan in Ann Arbor elegantly demonstrated this effect in Polistes dominulus wasps.

In Polistes dominulus wasps, aggressive contests are used to determine dominance among nest-founding queens. As with any aggressive contest, animals can benefit if they can assess their opponent’s abilities prior to risking getting their butt kicked. In this species, the “brokenness” of facial patterns is used as a signal to provide information about the wasp’s fighting abilities: Wasps with more broken facial patterns (high signal) are usually better fighters (high ability). Additionally, wasps with more juvenile hormone (that’s actually what it’s called) tend to be better fighters.

Portraits of five different Polistes dominulus females arranged from
low signal and low fighting ability (left) to high signal and high fighting ability (right).
Images from Tibbetts' and Izzo's 2010 Current Biology paper.

The researchers created a mismatch of facial markings and fighting ability in wasps by painting their faces to alter their signals and by giving them a compound similar to juvenile hormone to alter their fighting abilities. This created four groups of wasps: wasps that had a low fighting ability and low signal, wasps that had a high fighting ability and high signal, wasps that had a low fighting ability and high signal, and wasps that had a high fighting ability and low signal.

Individuals that had facial patterns inaccurately signaling high fighting ability received more aggression from their fellow wasps than any other group. These individuals also had difficulty establishing stable dominance hierarchies, which led to more friction with the group. So there are social costs imposed on wasps that have dishonest signals that suggest they are better fighters than they really are. Interestingly, there were also social costs to individuals which showed dishonest signals that suggested they were worse fighters than they really were: Although they could behave dominantly, the other wasps were less likely to submit to them and the dominance conflicts were prolonged. In both cases, the mismatch of signal and ability led to social costs in the form of increased aggression.

In Samantha Brick’s case, her news story was her “signal”, which indicated her attractiveness to be a 10 (on a scale of 1 to 10). I took her pictures (the ones above) to a Milwaukee Brewers game and later to class to ask baseball fans and college students the two questions above: “Have you heard of Samantha Brick?” and “On a scale of 1 to 10, how attractive do you think she is?” On average, the people that had never heard of her gave her about a 7.5…not a 10, but not bad. But the people that had heard of her story gave her an average score of 5.5… a solid 2-point deduction. One young woman at the Brewers game explained her own version of the phenomenon, saying, “I think she’s an 8 or a 9. But I could never give someone a score lower than a 5 or 6 – That would just be mean”. When I told her about the article Brick wrote, the woman said, “Oh. She’s a 4.”

Interestingly, this effect was not just in women. Men also rated her attractiveness as significantly less when they knew of her article. And men wrote many of the negative comments found online and on Twitter. This backlash wasn’t because “women do not like attractive women” (although there is some complicated truth in that statement). This backlash was because Brick sent out a public dishonest signal that she is a 10, when in fact she is a 7.5, and the public responded to the mismatch with social aggression.


Want to know more? Check these out:

Tibbetts, E., & Izzo, A. (2010). Social Punishment of Dishonest Signalers Caused by Mismatch between Signal and Behavior Current Biology, 20 (18), 1637-1640 DOI: 10.1016/j.cub.2010.07.042

Etcoff, N. (1999). Survival of the Prettiest: The Science of Beauty. Anchor Books, New York, NY.

Wednesday, April 4, 2012

Animal Mass Suicide and the Lemming Conspiracy

Ticked off Norway lemming doesn't like gossip!
Photo from Wikimedia Commons by Frode Inge Helland 
We all know the story: Every few years, millions of lemmings, driven by a deep-seated urge, run and leap off a cliff only to be dashed on the rocks below and eventually drowned in the raging sea. Stupid lemmings. It’s a story with staying power: short, not-so-sweet, and to the rocky point.

But it is a LIE.

And who, you may ask, would tell us such a horrendous fabrication? Walt Disney! Well, technically not Walt Disney himself… Let me explain:

The Disney Studio first took interest in the lemming mass suicide story when, in 1955, they published an Uncle Scrooge adventure comic called “The Lemming with the Locket” illustrated by Carl Barks. In this story, Uncle Scrooge takes Huey, Dewey and Louie in search of a lemming that stole a locket containing the combination to his vault … but they have to catch the lemming before it leaps with all his buddies into the sea forever. Three years later, Disney further popularized this idea in the 1958 documentary White Wilderness, which won that year’s Academy Award for Best Documentary Feature. A scene in White Wilderness supposedly depicts a mass lemming migration in which the lemmings leap en masse into the Canadian Arctic Ocean in a futile attempt to cross it.


In 1982, the fifth estate, a television news magazine by the CBC (that’s the Canadian Broadcasting Corporation), broadcast a documentary about animal cruelty in Hollywood. They revealed that the now infamous White Wilderness lemming scene was filmed on a constructed set at the Bow River in Canmore, Alberta, nowhere near the Arctic Ocean. Lemmings are not native to the area where they filmed, so they imported them from Churchill after being purchased from Inuit children for 25 cents each. To give the illusion of a mass migration, they installed a rotating turntable and filmed the few lemmings they had from multiple angles over and over again. As it turns out, the lemming species filmed (collared lemmings) are not even known to migrate (unlike some Norwegian lemmings). Worst of all, the lemmings did not voluntarily leap into the water, but were pushed by the turntable and the film crew. Oh, Uncle Walt! How could you?!

Norway lemmings really do migrate en masse, but they don't commit mass suicide.
Drawing titled Lemmings in Migration, in Popular Science Monthly Volume 11, 1877.
As far as we know, there are no species that purposely hurl themselves off cliffs to die en masse for migration. But, strangely enough, North Pacific salmon do purposely hurl themselves up cliffs to die en masse for migration. And what, you may ask, is worth such a sacrifice? Sex, of course!

Migrating sockeye salmon thinking about sex.
Photo from Wikimedia Commons by Joe Mabel.

The six common North Pacific salmon species are all anadromous (meaning that they are born in fresh water, spend most of their lives in the sea and return to fresh water to breed) and semelparous (meaning they only have a single reproductive event before they die). After years at sea, salmon swim sometimes thousands of miles to get to the mouth of the very same stream in which they were born. Exactly how they do this is still a mystery. Once they enter their stream, they stop eating and their stomach even begins to disintegrate to leave room for the developing eggs or sperm. Their bodies change in other ways as well, both for reproduction and to help them adapt to fresh water. They then swim upstream, sometimes thousands of miles more, and sometimes having to leap over multiple waterfalls, using up their precious energy reserves. Only the most athletic individuals even survive the journey. Once they reach the breeding grounds, the males immediately start to fight each other over breeding territories. The females arrive and begin to dig a shallow nest (called a redd) in which she releases a few thousand eggs, which are then fertilized by the male. They then move on, and if they have energy and gametes left, repeat the process with other mates, until they are completely spent. If the females have any energy left after laying all their eggs, they spend it guarding their nests. Having spent the last of their energy, they die and are washed up onto the banks of the stream.

Now that’s parental commitment! So the next time your parents start laying on the guilt about everything they’ve given up for you, share this nugget with them and remind them it could be worse…


Want to know more? Check these out:

1. Learn more about semelparity here

2. Learn more about salmon reproduction at Marine Science

3. And learn even more about salmon reproduction with this awesome post by science blogger and Aquatic and Fishery Sciences graduate student, Iris. Her current blog posts can be found here.

4. Ramsden E, & Wilson D (2010). The nature of suicide: science and the self-destructive animal. Endeavour, 34 (1), 21-4 PMID: 20144484