Category: Science
What falsehood would you most likely think is true?
Is that the best way to sum up the question? S.L., a loyal MR reader, asked:
What is the thing that would with highest likelihood be true if you didn't know it was probably not? This is extremely interesting but also too broad to get a handle on—easier to think about in certain areas. E.g. what is the economic non-fact that satisfies this, or the physical ("god does not play dice" is up there).
I will try "that the universe as we know has not been there forever." When I was a kid it seemed obvious to me that the sky and stars had always been there, albeit with some recycling of content, much like they changed the TV shows every few years or so. I was surprised to learn about the Big Bang, at first held out hope for steady state matter theories, and now again see that some form of constancy may survive at the level of the metaverse.
How about in economics? I am surprised how little power real interest rates have to predict investment. I am surprised how few investors can consistently violate/beat the weak efficient markets hypothesis (aren't lots of people simply smarter than the marginal investor?). And I am surprised that the Industrial Revolution ever took place, after not having taken place for so long.
Your picks?
Catch Shares
Catch Shares are expanding in California, Oregon and Washington starting in January.
Under the catch-share system, fishery managers set an overall catch limit at the beginning of the season. Each fisherman will own a percentage of that catch. Just like shares in the stock market, the quotas can be traded or sold. The idea is that a market-based system will give fishermen more flexibility.
Not everyone is happy. Larry Collins, president of the Crab Boat Owners Association in San Francisco, is doubtful.
Collins is concerned that a market-based system will bring market manipulation. Under the rules, you don't have to be a fisherman to buy fish quotas, so it's possible that speculators or even environmental groups could buy into the market.
"You want hedge fund managers deciding when the people catch fish? Is that who you want to own your fish, or do you want to own them?" Collins asks.
Collins is also concerned about fishermen who make smaller catches. In Alaskan fisheries that use catch shares, some smaller boats opted to sell their fish quotas.
"That concentrated the resource in fewer and fewer hands. Now, I tend to think that public trust resources should be used to employ as many people as possible," he says.
Both of these features are benefits not costs. It's true that speculation can create bubbles and other problems but speculators also make the market more future-oriented and this will help to avoid the collapse of fishing stocks by making prices a better early warning system. Moreover, if environmentalists want to buy catch shares to increase the fishing stock then I am all for it. In Modern Principles, Tyler and I discuss how we bought and retired some SO2 making the air cleaner for everyone (you are welcome! :)).
Fewer, larger boats is also a benefit not a cost. Under the current system the influx of small boats is simply a form of rent-seeking which raises net social costs–too many fisherman chasing too few fish. Catch shares reduce over-capitilization in the industry raising productivity (see also Modern Principles on this point).
The main problem with catch-shares is setting the size of the catch, which inevitably is a politicized process. Massachussetts congressmen, for example, are trying to withold funds from NOAA until catch shares are increased. Nevertheless with so many fishing stocks nearing collapse it is clear that some limits are needed. Moreover, before catch-shares are put in place few people in the industry appreciate that temporary limits can lead to long-term increases in catch as the fishing stocks recover to sustainable levels. After catch-shares are put in place, however, it is sometimes the fishermen who lobby for greater limits not for monopoly reasons but as they come to recognize that a smaller limit leads to a larger stock and larger profits.
Here is my previous post, Reversing the Decline in Fish Stocks, on catch shares with more links.
Hat tip: Daniel Lippman.
Papua New Guinea fact of the day
An analysis of this ancient DNA, published on Wednesday in Nature, reveals that the genomes of people from New Guinea contain 4.8 percent Denisovan DNA.
And who are they?
An international team of scientists has identified a previously shadowy human group known as the Denisovans as cousins to Neanderthals who lived in Asia from roughly 400,000 to 50,000 years ago…
Here is the article. It is suggested that the Denisovans are quite distant from both humans and Neanderthals. Here is the first cut take from Gene Expression. Here is more, from John Hawks. And more here. Ongoing updates here.
Ahead of their tenure clock
This is the conclusion of a new paper published in Biology Letters, a high-powered journal from the UK’s prestigious Royal Society. If its tone seems unusual, that’s because its authors are children from Blackawton Primary School in Devon, England. Aged between 8 and 10, the 25 children have just become the youngest scientists to ever be published in a Royal Society journal.
Their paper, based on fieldwork carried out in a local churchyard, describes how bumblebees can learn which flowers to forage from with more flexibility than anyone had thought. It’s the culmination of a project called ‘i, scientist‘, designed to get students to actually carry out scientific research themselves. The kids received some support from Beau Lotto, a neuroscientist at UCL, and David Strudwick, Blackawton’s head teacher. But the work is all their own.
The class (including Lotto’s son, Misha) came up with their own questions, devised hypotheses, designed experiments, and analysed data. They wrote the paper themselves (except for the abstract), and they drew all the figures with colouring pencils.
One version of the story is here, which offers an excellent account and lots of background detail. The experiment had not been done before. The abstract was the one part of the paper they could not write on their own.
The paper is here. There are no statistics and no references to previous literature. The first paragraph of the introduction is this:
People think that humans are the smartest of animals, and most people do not think about other animals as being smart, or at least think that they are not as smart as humans. Knowing that other animals are as smart as us means we can appreciate them more, which could also help us to help them.
What economics project could you imagine eight-year-olds doing and publishing?
For the pointer I thank numerous sources on Twitter.
Is there an ancestor effect?
An initial study involved 80 undergrads spending five minutes thinking about either their fifteenth century ancestors, their great-grandparents or a recent shopping trip. Afterwards, those students in the two ancestor conditions were more confident about their likely performance in future exams, an effect that seemed to be mediated by their feeling more in control of their lives.
Three further studies showed that thinking or writing about their recent or distant ancestors led students to actually perform better on a range of intelligence tests, including verbal and spatial tasks (in one test, students who thought about their distant ancestors scored an average of 14 out of 16, compared with an average of 10 out of 16 among controls). The ancestor benefit was mediated partly by students attempting more answers – what the researchers called having a 'promotion orientation'.
The full account is here. I would like to see this replicated, and subject to more variation, but in the meantime it's an interesting idea.
Paragraphs about prosopagnosia
Face space also explains why the favorite trick of editorial cartoonists works so well. By exaggerating features on a politician's face — Bush's eyebrows, Obama's ears — cartoonists push it farther away from the center of face space, to places where it has less competition from other faces we have stored in our memory. As a result, we recognize people from hand-drawn caricatures as quickly as from photographs — and sometimes even more quickly.
That is Carl Zimmer, from the January/February 2011 issue of Discover, not yet on-line. Here is a short piece on how to draw caricatures.
*The Year in Ideas*: Informal audit of past issues
The New York Times Sunday Magazine asked me to reread through all the previous issues of their "The Year in Ideas" feature and write down my impressions. The piece starts like this:
The editors asked Tyler Cowen, the economist who helps run the blog Marginal Revolution, to read the previous nine Ideas issues and send us his thoughts on which entries, with the benefit of hindsight, struck him as noteworthy. Do any ideas from this year’s issue look promising? “I recall reading the 2001 issue when it came out,” he says. “And I was hardly bowled over with excitement by thoughts of ‘Populist Editing.’ Now I use Wikipedia almost every day. The 2001 issue noted that, in its selection of items, ‘frivolous ideas are given the same prominence as weighty ones’; that is easiest to do when we still don’t know which are which.”
In the piece I select the best ideas, the most prescient picks, the most oversold, and so on. The most "off" picks were:
2001: “The ‘X-Files’ Conspiracy Trope is Dead“, and 2001: “American Imperialism, Embraced”
This project was fun. It was striking to me how many of the items in the series concerned information technology, how few concerned formal education, and how few of the non-internet items involved actual improvements in our living standards.
How to Make Friends Without Influencing People
Bryan Caplan had a great post last week combining statistics, biology, and parenting to lead to the conclusion that weird people should have more kids. First, the statistics. If there is a zero correlation between parental and child traits then your child is as likely to be as similar to you as is a stranger. If the correlation between parent and child traits is greater than zero then you are more likely to be like your child than a stranger but only if you yourself are not normal. Here is Bryan:
Take a look:
Parent-Child Correlation
r= 0
r=.5
You
Stranger
Child
Stranger
Child
Percentile/
Expected
Percentile
50th
50th
50th
50th
50th
95th
50th
50th
50th
80th
99.99th
50th
50th
50th
95th
Notice that regardless of the value of r, normal people can expect to be like their kids. But that's not saying much, because normal people can expect to be like any random person they meet! The story's very different for weirdos. By definition, weirdos never have much in common with random strangers. With a zero parent-child correlation, weirdos will feel equally alienated from their children. As the parent-child correlation rises, however, weirdos' incompatibility with strangers stays the same, but their expected compatibility with their children gets stronger and stronger.
Now let's look at these facts like a mad economist. There are two ways to surround yourself with people like you. One is to meet them; the other is to make them. If you're average, meeting people like yourself is easy; people like you are everywhere. If you're weird, though, meeting people like yourself is hard; people like you are few and far between. But fortunately, as the parent-child correlation rises, weirdos' odds of making people like themselves get better and better.
…The lesson: As your weirdness increases, so does your incentive to have kids. If you like football and American Idol, you're never really alone. You don't need to build a Xanadu for yourself. But if you're a lonely misfit, oddball, freak, or weirdo, then find a like-minded spouse and make new life together. Let the normals laugh at you. You'll have each other.
Markets in Everything: Name a Theorem
You can name your very own mathematical theorem, newly generated by one of the world's most advanced computerised theorem provers (a kind of robot mathematician), and you can immortalise your loved ones, teachers, friends and even yourself and your favourite pets.
I would be afraid that I would not understand my own theorem (see here for an example).
I will stick with Tabarrok's Wager (original paper here).
Hat tip: Boing Boing.
Rebounds per game, or rebounds per minute? (not a post about basketball)
When someone wins the Cy Young award with a 13-12 record, you reconsider the reliability of particular statistics and also the meaning of the award. In economics we are taught, correctly, that Ronald Coase is a world-class economist, despite his relatively small number of publications. Virtually each piece is a gem.
In the NBA, which is a better or more important stat? Rebounds per game, rebounds per minute, or how about "total rebound percentage"? Should not some measure of rebounding rate win out here?
Nonetheless, I still look first to total rebounds, whether in a game, in a year, or even in a career. How much time you are on the court is endogenous. If you are a superb rebounder but cannot play more than ten minutes a game — because of injury, uncooperativeness, or other missing skills — you will have a low number of total rebounds and that will reflect your broader deficits.
Greg Oden has a high rebound rate but he hardly plays, due to recurring injury. No one calls him the Ronald Coase of rebounding.
Similarly, Yao Ming has high success rates, but cannot stay on the court for very long, due to his bad feet. His team has plenty of talent but has not won much and it probably needs to be dismantled at this point.
In other words, it is often "brute total" statistics which are underrated (think about evaluating a potential spouse). And brute total statistics are most important when you must cooperate with others in complementary fashion and maintain their productivity as well as your own. They are least important when, like Wittgenstein, Coase, or Sraffa, you occasionally issue a missive of brilliance and then retreat for years. Coase did make his Chicago colleagues much more productive, but that effect would be weaker today in this age of specialization and co-authorship.
Both experimental economics and field experiments involve a lot of researcher cooperation and both are fields on the rise. Does this mean that total output statistics will/should become more important for assessing economists?
Circa 2010, should we be looking more for economists who are more like Nolan Ryan and less like Ronald Coase?
Addendum: Angus comments.
Are bees more Bayesian?
It appears, therefore, that a swarm's scout bees do something sharply different from what humans do to reach a full agreement in a debate. Both bees and humans need a group's members to avoid stubbornly supporting their first view, but whereas we humans will usually (and sensibly) ive up on a position only after we have learned of a better one, the bees will stop supporting a position automatically. As is shown…after a shorter or longer time, each scout bee becomes silent and leaves the rest of the debate to a new set of bees. Figure 6.7 shows how this regular turnover in which scouts are dancing can help a swarm's scouts quickly reach an agreement…
In other words, the bee algorithms allow attrition (a time-honored process of improving the scientific community) to operate at an especially rapid pace.
That is from the fascinating book Honeybee Democracy, by Thomas D. Seeley. Here is the book's home page. Here is a good review of the book:
In the final chapter, Seeley suggests five lessons we could learn from bees.
†¢ Compose a decision-making group of individuals with shared interests. Here bees have a higher stake than us: all members of a colony are related (sisters) and nobody can survive without the group.
†¢ Minimise the leader's influence on the group. Here we humans have much to learn.
†¢ Seek diverse solutions to the problem. Humans realised only recently that diversity is good for a group.
†¢ Update the group's knowledge through debate. Here again, bees are superior to us, as each scout's "dances" become less effective with time, no matter how good a new site is, while stubbornness can lead humans to argue forever.
†¢ Use quorums to gain cohesion, accuracy and speed. Impressively, bees came up with this concept long before the Greeks.
As a departmental chair at Cornell University, Seeley says, he applies these principles at faculty meetings with great success.
Definitely recommended.
Arsenic-based bacteria
In Mono Lake, created by scientists, working together with evolution:
Hours before their special news conference today, the cat is out of the bag: NASA has discovered a completely new life form that doesn’t share the biological building blocks of anything currently living in planet Earth. This changes everything.
(That is a bad summary it turns out, try the paper itself.) And here and here. (This piece covers the skeptics.) This is important, no? It implies a lot more life out there in the universe than many people had thought.
Ideas Behind Their Time
We are all familiar with ideas said to be ahead of their time, Babbage’s analytical engine and da Vinci’s helicopter are classic examples. We are also familiar with ideas “of their time,” ideas that were “in the air” and thus were often simultaneously discovered such as the telephone, calculus, evolution, and color photography. What is less commented on is the third possibility, ideas that could have been discovered much earlier but which were not, ideas behind their time.
Experimental economics was an idea behind its time. Experimental economics could have been invented by Adam Smith, it could have been invented by Ricardo or Marshall or Samuelson but it wasn’t. Experimental economics didn’t takeoff until the 1960s when Vernon Smith picked it up and ran with it (Vernon was not the first experimental economist but he was early).
(Economics, and perhaps social science in general, seems behind its time compared say with political science.)
A lot of the papers in say experimental social psychology published today could have been written a thousand years ago so psychology is behind its time. More generally, random clinical trials are way behind their time. An alternative history in which Aristotle or one of his students extolled the virtue of randomization and testing does not seem impossible and yet it would have changed the world.
Technology can also be behind its time. View morphing (“bullet time”) could have been used much more frequently well before The Matrix in 1999 (you simply need multiple cameras from different angles triggered at the same time and then inserted into a film) but despite some historical precedents the innovation didn’t happen.
Ideas behind their time may be harder to discover than other ideas–“if this is so great why hasn’t it been done before”? is an attack on ideas behind their time that other innovations do not have to meet. Is this why social innovations are often behind their time?
What other ideas were behind their time? Are some types of ideas more likely to be behind their time than others? Why?
Addendum: See Jason Crawford on Why did it take so long to invent X?
Richard Thaler’s question
I am doing research for a new book and would like to hope to elicit informed responses to the following question:
The flat earth and geocentric world are examples of wrong scientific beliefs that were held for long periods. Can you name your favorite example and for extra credit why it was believed to be true?
Please note that I am interested in things we once thought were true and took forever to unlearn. I am looking for wrong scientific beliefs that we've already learned were wrong, rather than those the respondent is predicting will be wrong which makes it different from the usual Edge prediction sort of question.
There are many answers here (scroll down). Hat tip goes to Edge on Twitter.
Observations on computer chess spectatorship
1. People enjoy watching a live internet human vs. human game more, when they can watch a computer judging the human moves and evaluating the position.
2. Few people enjoy watching live computer vs. computer games, even though the quality of play is much higher and the likelihood of a complex, wild position is much higher. Even if you care at all, there is little in-progress suspense; you might as well look back at the moves once they are over. How many other activities would we enjoy watching or experiencing less if they were done by computers?
3. The quality of play in a computer vs. computer game is so high it is often difficult for humans to tell where the losing computer went wrong, even if the spectator human has the help of a chess-playing computer.
4. I find only the very best computer (Rybka) of interest, although I do not feel the same way about the human players. Furthermore the fifth best computer is still much better than the best human players.
5. The notion of a computer chess tournament taking place "in time" is an odd one. You can play all the games back-to-back or simply use multiple copies of the programs and finish the entire tournament in a few hours; see #2.
6. Watching a computer play chess is a window onto a world where, once the opening is past (often, computers are simply told what to do by a pre-programmed "openings book"), there are many fewer presuppositions than what a human mind will bring to bear on the problem. It's a very good way of learning, in convincing form (the computer will beat you), how much your intuitions lead you astray. It's not just your "bad moves" which cause you to lose, it's also the moves which still seem pretty good to you.
7. There are nonetheless many computer moves which I simply cannot believe are any good. It does seem that every now and then computers get stuck in a "dogmatic trap," usually because of their limited time horizons for evaluation. Playing against a computer, you will do best in the early middle game and then progressively fall apart as its combinatorial powers destroy you.
8. You can watch chess computers play against each other at www.chessbomb.com. Click on "enter" and then TCEC5.