I camped over at the Occupy LA protest a few weeks ago. It was fun– most of the people seemed thoughtful and genuine and I sympathize with a lot of their concerns. I have some friends who are “occupying” as I write this. People are mad, and I get it.

I also have friends and family who work in the greater investment community. Their opinions of OWS are all over the map: some welcome the protests if they can bring about more market transparency and accountability, others are quite frustrated by the protesters’ general lack understanding, sophistication or solutions. To put (my) words to their feelings: things aren’t perfect, but many of the protesters’ demands betray a striking lack of comprehension about how the market works.

I’m not here to pick a winner. I am here to say that this dispute hides a much bigger problem, one independent from any issue of corruption and one that will unravel the fabric of society if we sleepwalk into it. Settle in, get a cup of coffee, and I’ll explain why.

Why people are mad: it’s not winning, it’s cheating
Matt Taibbi at Rolling Stone succinctly explains why the Occupy Wall Street protestors are mad:

Success is the national religion, and almost everyone is a believer. Americans love winners.  But that’s just the problem. These guys on Wall Street are not winning – they’re cheating. And as much as we love the self-made success story, we hate the cheater that much more.

In this country, we cheer for people who hit their own home runs – not shortcut-chasing juicers like Bonds and McGwire, Blankfein and Dimon.

That’s why it’s so obnoxious when people say the protesters are just sore losers who are jealous of these smart guys in suits who beat them at the game of life. This isn’t disappointment at having lost. It’s anger because those other guys didn’t really win. And people now want the score overturned.

All weekend I was thinking about this “jealousy” question, and I just kept coming back to all the different ways the game is rigged. People aren’t jealous and they don’t want privileges. They just want a level playing field[.]

Wall Street’s recent antics would probably irk people a lot less if the public didn’t have the perception the bankers are playing a “heads I win, tails you lose” game backed by taxpayer money. Institutions that are “too big to fail” take great investment risks and are either rewarded handsomely, or get bailed out by the government (read: taxpayers).

Nassim Taleb (of Black Swan fame) offers a possible solution for this kind of moral hazard:

[I]t’s time for a fundamental reform: Any person who works for a company that, regardless of its current financial health, would require a taxpayer-financed bailout if it failed should not get a bonus, ever. In fact, all pay at systemically important financial institutions — big banks, but also some insurance companies and even huge hedge funds — should be strictly regulated.

Critics like the Occupy Wall Street demonstrators decry the bonus system for its lack of fairness and its contribution to widening inequality. But the greater problem is that it provides an incentive to take risks. The asymmetric nature of the bonus (an incentive for success without a corresponding disincentive for failure) causes hidden risks to accumulate in the financial system and become a catalyst for disaster. This violates the fundamental rules of capitalism; Adam Smith himself was wary of the effect of limiting liability, a bedrock principle of the modern corporation.

I don’t agree with many of the OWS complaints, but it’s clear that some parts of Wall Street are broken. Regulatory capture is common, banks seem to operate above the law, and the game is often subtly rigged in a thousand little ways. HFT is arguably a parasite twice-over, siphoning both money and bright young people away from other areas into a negative-sum pit of cheat-or-be-cheated. Goldman Sachs does resemble, in the words of Taibbi, “a great vampire squid wrapped around the face of humanity, relentlessly jamming its blood funnel into anything that smells like money.”[1]

The richest 1% (and arguably the financial sector) have seemingly diverted almost all the economic growth of the past 30 years into their pockets, and to a large extent that’s been opportunistic, not meritocratic. What’s more, this ‘short-term greedy’ way of doing business threatens to spread, like a sickening disease, across more parts of our economy and government. We need to fix this or it’ll drag us, kicking and screaming, into a true kleptocracy.[2]

Many in OWS have completely lost faith in the political process, believing that these Powers That Be have successfully co-opted the acceptable mechanisms for change so completely that political disobedience is the only way to send a message.[3] I want to think they’re wrong, but at some level I have to agree.

Image: retired Philadelphia Police Capt. Ray Lewis being arrested for ‘unlawfully blocking traffic’ during OWS. He had been loudly accusing the NYPD of serving as mercenaries for Wall Street. Given that there have been no high-profile arrests of the people who were responsible for defrauding billions from pensions and taxpayers and (proximately) causing the financial crisis, he may have a point.

Having said all that– and the more I think about this the more certain I am– OWS protesters are on the right side of this battle, but the wrong side of history. The story of why starts with monkeys and ends with robots.

Perception and complexity
Jonah Lehrer recounts a primate study centered around instinctual reactions to perceived injustice:

A similar lesson emerges from a classic experiment conducted by Franz de Waals and Sarah Brosnan. The primatologists trained brown capuchin monkeys to give them pebbles in exchange for cucumbers. Almost overnight, a capuchin economy developed, with hungry monkeys harvesting small stones. But the marketplace was disrupted when the scientists got mischievous: instead of giving every monkey a cucumber in exchange for pebbles, they started giving some monkeys a tasty grape instead. (Monkeys prefer grapes to cucumbers.) After witnessing this injustice, the monkeys earning cucumbers went on strike. Some started throwing their cucumbers at the scientists; the vast majority just stopped collecting pebbles. The capuchin economy ground to a halt. The monkeys were willing to forfeit cheap food simply to register their anger at the arbitrary pay scale.

This labor unrest among monkeys illuminates our innate sense of fairness. It’s not that the primates demanded equality — some capuchins collected many more pebbles than others, and that never created a problem — it’s that they couldn’t stand when the inequality was a result of injustice. Humans act the same way. When the rich do something to deserve their riches, nobody complains; that’s just the meritocracy at work. But when those at the bottom don’t understand the unequal distribution of wealth — when it seems as if the winners are getting rewarded for no reason — they get furious. They doubt the integrity of the system and become more sensitive to perceived inequities. They start camping out in parks. They reject the very premise of the game.

This is a fantastic experiment; I only wish the researchers had delved into the gray areas between ‘totally fair’ and ‘totally unfair’ — where monkeys were rewarded with grapes based on consistent but increasingly complex conditions. Where would their breaking point be?

Just like monkeys, our capacity for understanding economic complexity is not infinite.[4] A friend who works in finance suggested that past a certain threshold, it might be impossible to figure out if any specific part of our economy is fair. And some people are bound to interpret inability to determine fairness as unfairness.

I think this is true. On the other hand, perhaps past a certain complexity threshold, economies play host to greatly more parasitic activities. In fact, I think the complexity-driven decoupling between wealth creation and compensation almost guarantees it. Is the greater total wealth created in more complex societies worth the greater parasite load? And does a default assumption of fairness or unfairness serve us better when we see something we don’t understand? I can see it going both ways. The consequences of going too far in either direction are severe:
- with a too-permissive view of corruption, economically parasitic behavior and distrust can spread from one sector into others, undermining the contracts society is based upon;
- with a too-sensitive corruption radar,  the perception of unfairness (as with any negative emotion) can drive a vicious psychological feedback loop, and people (like monkeys) stop working. It might take a Hard Reset like the 1930s to cure this.

It’s a difficult balance, and the devil’s in the details. But the takeaway is that we’re not always in a position to judge what is fair, and, very importantly…

An increasing economic gap can be healthy!
Silicon Valley entrepreneur Paul Graham has a thoughtful article about attitudes toward wealth, arguing that a widening gap between rich and poor can be a sign of health if it reflects people getting paid for creating stuff people want. It’s the same form of argument that conservatives have been using for 30 years to justify the widening gap, but Graham’s perspective from the world of software startups, where success is tightly tied to the wealth one creates for others, is pretty powerful.

Where does wealth come from? People make it. This was easier to grasp when most people lived on farms, and made many of the things they wanted with their own hands. Then you could see in the house, the herds, and the granary the wealth that each family created. It was obvious then too that the wealth of the world was not a fixed quantity that had to be shared out, like slices of a pie. If you wanted more wealth, you could make it.

…

And since the ability and desire to create it vary from person to person, it’s not made equally.

You get paid by doing or making something people want, and those who make more money are often simply better at doing what people want. Top actors make a lot more money than B-list actors. The B-list actors might be almost as charismatic, but when people go to the theater and look at the list of movies playing, they want that extra oomph that the big stars have.

Doing what people want is not the only way to get money, of course. You could also rob banks, or solicit bribes, or establish a monopoly. Such tricks account for some variation in wealth, and indeed for some of the biggest individual fortunes, but they are not the root cause of variation in income. The root cause of variation in income, as Occam’s Razor implies, is the same as the root cause of variation in every other human skill.

In the United States, the CEO of a large public company makes about 100 times as much as the average person. [3] Basketball players make about 128 times as much, and baseball players 72 times as much. Editorials quote this kind of statistic with horror. But I have no trouble imagining that one person could be 100 times as productive as another. In ancient Rome the price of slaves varied by a factor of 50 depending on their skills. [4] And that’s without considering motivation, or the extra leverage in productivity that you can get from modern technology.

None of this excuses Wall Street corruption– far from it, a society where it’s easier to steal wealth than make it will demotivate talented people, and divert their efforts from creation towards theft– but the point is some people are naturally better at creating wealth than others. Moreover, technology leads to more winner-take-all scenarios, and Graham predicts this gap widening:

Will technology increase the gap between rich and poor? It will certainly increase the gap between the productive and the unproductive. That’s the whole point of technology. With a tractor an energetic farmer could plow six times as much land in a day as he could with a team of horses. But only if he mastered a new kind of farming.

I’ve seen the lever of technology grow visibly in my own time. In high school I made money by mowing lawns and scooping ice cream at Baskin-Robbins. This was the only kind of work available at the time. Now high school kids could write software or design web sites. But only some of them will; the rest will still be scooping ice cream.

The larger trend here is that:

Jobs are harder to find, easier to invent.
Thomas “the world is flat” Friedman has a great column about this theme that ‘jobs are getting harder to find, but easier to invent’:

I’VE done a lot of television book interviews lately, and I continue to be struck at what a difference there is in the technology in just a few years’ time.

Here is a typical evening at a major cable TV network: arrive at Washington studio and be asked to sign in by a contract security guard. Be met by either a young employee who appears to still be in college or an older person who seems to have hung on with tenure. Have your nose powdered by that person. Have your microphone attached by that person. Be positioned in the studio chair by that person, and then look directly into a robotic camera being manipulated by someone in a control room in New York and speak to whoever the host is wherever he or she is. That’s it: one employee, a robot and you.

Think of how many jobs — makeup artist, receptionist, camera person, producer-director — have been collapsed into one. I raise this point because there is no doubt that the main reason for our 9.1 percent unemployment rate is the steep drop in aggregate demand in the Great Recession. But it is not the only reason. “The Great Recession” is also coinciding with — and driving — “The Great Inflection.”

In the last decade, we have gone from a connected world (thanks to the end of the cold war, globalization and the Internet) to a hyperconnected world (thanks to those same forces expanding even faster). And it matters. The connected world was a challenge to blue-collar workers in the industrialized West. They had to compete with a bigger pool of cheap labor. The hyperconnected world is now a challenge to white-collar workers. They have to compete with a bigger pool of cheap geniuses — some of whom are people and some are now robots, microchips and software-guided machines.

I wrote about the connected world in 2004, arguing that the world had gotten “flat.” When I made that argument, though, Facebook barely existed — and Twitter, cloud computing, iPhones, LinkedIn, iPads, the “applications” industry and Skype had either not been invented or were in their infancy. Now they are exploding, taking us from connected to hyperconnected. It is a huge inflection point masked by the Great Recession.

It is also both a huge challenge and opportunity. It has never been harder to find a job and never been easier — for those prepared for this world — to invent a job or find a customer. Anyone with the spark of an idea can start a company overnight, using a credit card, while accessing brains, brawn and customers anywhere. [emphasis added]

This isn’t the only employment trend afoot, but it’s one of the most significant.[5] It’s hard to disagree with Friedman’s advice, but for one thing:

Half of all people are below average(!)
Friedman’s advice is not likely to be feasible for, or read by. the average person. And, save for exceptional (and fictional) havens like Lake Wobegone, half of all people are below average.

Fred Reed, a curmudgeon if there ever was one, grumbles about this lack of realism among the op-ed class:

Letting Them Eat Cake

When I read columnists or listen to talking heads on the lobotomy box, they strike me as delusional. What are these decapitated crania prattling about? From what morgue did they escape? What country are they from? Certainly not the America I grew up in.

I conclude that they suffer from Commentator’s Disease, which consists in the confluence of several disabilities, the first being high intelligence. Washington, being a center of power, politics, graft, and corruption, attracts the very bright. An acquaintance once said, “Inside the Beltway, you assume that everyone is in the ninety-ninth percentile.” She meant that in the circles in which she moved, this was true. The city is rife with the very bright, most of them being invisible: campaign planners, pollsters, lawyers, scientists from NIH. The class includes many of the talking heads, the Pat Buchanans and Charles Krathammrs. They may be liberal or conservative, depending on their individual defects of character, but they are way smart.

The exceedingly intelligent form a social class seldom mentioned but inordinately influential. They are not recognized as what they are because they do not append IQs to their by-lines. As a quite ordinary example, consider the magazine The American Conservative, with many of whose writers I have some familiarity. The publisher, Ron Unz, studied theoretical physics at Stanford after graduating from Harvard. Bill Lind, Pat Buchanan, Taki, Steve Sailer, Kara Hopkins, John Derbyshire—I doubt that there is an IQ below 140 in the bunch. The same could be said of many other political slicks, left or right.

These people are not intellectual snobs. In the crowd they run with, they are average. The problem with them is that they hang out together. People tend to associate with those with whom they have things in common. At a hole-in-the-wall in DC like the Zoo Bar on upper Connecticut you may find a table of eight people in jeans and running shoes—Washington is about power, not style—consisting of a biochemist, an editor of a technical newsletter, a talking head you’ve seen, and so on, all highly educated. This clustering together by intelligence is sometimes called “cognitive stratification.” It exists, big time. The clusterers are by and large decent people, not full of themselves, and mean well.

But.

But they don’t know what they are talking about in important respects. They think the Beltway contains America.

The second symptom of Commentator’s Disease is relative prosperity. The nature of Washington is that the very bright usually do well financially. I don’t mean that they are rich, though some are, but that they manage to find secure jobs in government or with law firms or they invest wisely or, in the case of commentators, angle for well-paid gigs with syndicates or networks. Usually there is nothing crooked in this. They are simply smart enough to work the system, and they live where the system is.

The aggregate effect of their brains, security, and isolation is that they are out of touch with the country as it really is. They do not know the bleak strip-development of Route 1 South toward Fredericksburg, red dirt and franchised cholesterol chutes and roaring traffic. Here the diabetic veteran lives in a decayed residential motel and makes his way on crutches to the down-scale diner where he drinks beer and waits to die because he hasn’t got anything else to wait for. (The example is not hypothetical.) Here the aging waitress gets to the diner somehow, aching with arthritis. “Too tired to work, too poor to stop.” I knew this woman. She is much of America. You don’t see her at the Zoo Bar. She has never been to such a place.

I often see victims of Commentator’s Disease arguing against the minimum wage on abstract grounds of economic theory. It is what commentators do—bandy abstractions, railing for or against Keynes, assaulting their ideological opponents with pointed phrases. They have never had to do the arithmetic of forty times the minimum wage minus taxes minus bus fare minus rent and gotta pay the cable because it is the only thing they have after work. They have never had to choose between the electric bill and a new coat as winter comes on.

The commentators don’t realize that not everybody is like them. Those with IQs of 140 and up (130 gets you into Mensa, I think) unconsciously believe that anything is possible. Denizens of this class know that if they decided to learn, say, classical Greek, they could. You get the book and go at it. It would take work, yes, and time, but the outcome would be certain.

They don’t understand that the waitress has an IQ of 85 and can’t learn much of anything.

Conservatives think in terms of merciless abstractions and liberals insist that everyone is equal. Not even close. Further, people with barely a high-school education and low-voltage minds regard any intellectual task with utter discouragement.

Some commentators urge letting people invest their Social Security taxes in the stock market. To them it is a question of abstract freedom and probably the Federalist papers. The commentators are smart enough to invest money. I’ll guess that at least half the population isn’t. Go into the tit bar (does it still exist) in Waldorf, Maryland, and ask the dump-truck drivers and nail-pounders what NASDAQ is.

Liberal commentators want everyone to go to college, when about a fifth of people have the brains. Conservatives think that people can rise by hard work and sacrifice as certainly many people have. Thing is, most people can’t. Commentators only see those who made it.

The tendency of the Beltway 99th to live in an imaginary world, of conservatives to think that everybody can be a Horatio Alger, of liberals to believe that inequality arises from discrimination, guarantees wretched policy. Those who can do almost anything need to recognize the existence of those who can do almost nothing. Few of the latter are parasites. The waitress has worked all her life, as has the truck driver. They ended up with nothing.

Which is easy to do. A girl marries her high-school sweetheart in Busted Hump, Tennessee and he goes to work for the local pickle-bottling plant, which switches to hiring people as independent contractors to avoid paying benefits. Neither of the pair is real bright, just ordinary Americans trying to make a living. They live paycheck to paycheck because they don’t know how not to. Neither is lazy. They just don’t know how to start the next Microsoft. He dies of a heart attack at 45, she can’t make the mortgage, and…she is well and truly screwed.

At the Zoo Bar, they have great wings and some really good walk-in blues bands, and what you have to understand about Keynes is….

Commentator’s Disease.

Ultimately, I think Reed is right– life is tough for those on the left half of the bell-curve, poverty imposes its own cognitive costs, and a lot of people are simply incapable of following advice like Friedman’s.[5] In an increasingly ultra-darwinistic economy, the ungifted and unmotivated are pretty much screwed. And that’s scary, since…

In the long run, we’re all unemployable
This is deep crystal ball territory, but it seems clear to me the way the wind is blowing: sooner or later we’re all going to be on the left half of the bell curve, driven from our jobs by smarter robots and software. The automation trends that started in agriculture and manufacturing have spread to such rarified fields as translation, chess, and legal and medical research. Whatever rationale people have for why an algorithm can’t replace them is, regardless of field, relentlessly eroding.

Let me be clear: algorithms are sneaking up on your job.

IBM’s Watson and Apple’s new Siri digital assistant are essentially early prototypes of AI, and as time goes on they’ll only get better and more generally-capable, able to replace and improve more sorts of traditionally-human tasks. Over the next ten years many such jobs are going to disappear, never to return. Some of the people displaced will read Friedman and figure out how to thrive in this new “winner-take-all, darwinian, creative destruction economy”, but let’s be honest. Most won’t.

Thresholds, the unemployable, and a new social contract
Vernor Vinge has a quote that I’ve always found striking: “The work that is truly productive is the domain of a steadily smaller and more elite fraction of humanity.” To wit, as time goes on, there will be more and more total wealth, but a smaller and smaller fraction are actually creating the wealth. It’s not necessarily that the poor are lazy: often it’s that they really have nothing to offer, economically speaking. There are jobs, but not for them. (Or their jobs are subsidized– as when the average Walmart employee receives ~$1,000 of Medicaid, food stamps, and cash assistance from the government.)

So ultimately, how do we help our fellow citizens deal with this trend? When 25%– then 75%– of humanity have absolutely no conventionally-marketable skills, are absolutely not economic to hire (vs robots and software), what do we do? We can offer platitudes about education, retraining, and inventing your own job, but that’s not going to hold back the tide very long. There’s a trend here: we’re nearing the end of the “living wage era,” where we could talk about ‘everybody having the right to a living wage’ because this wasn’t too far from the fundamental economic realities. It was nice and egalitarian while it lasted, but what’s next?

Should society keep expanding social programs to permanently support this growing pool of unemployables? Should we keep subsidizing them (and Walmart) through aid to low-income workers? The Romans bought domestic peace with bread-and-circuses; will we need to take that route also? (Aren’t we already?) Or should we e.g., just keep expanding the DMV and TSA to give this class of workers a place to harmlessly soak up government dollars while feeling productive? (I say this only half in jest, since it appears to be the road we’re currently on.)

Moreover, given the social unrest possible once the unemployed/unemployable grows past certain thresholds, can we afford not to?[7]

I’m not being a snob; on a long enough timeline, I think we’re all in this boat.

Will OWS radicalize?
However one slices it, I think it’s clear that job economics will tighten. The “Occupy” groups are the vanguard in the populist pushback on this, and are spot on with some of their complaints.

It’s not always easy to see, but the political pendulum swings back and forth with tremendous force. We’re in a part of the cycle where we can see very keenly the power and abuses of large multinational corporations. It won’t always be like this: the OWS movement may be the start of powerful modern populism, cycling into focus just when the long-term feasibility of a “living wage for everyone” is naturally and irreversibly eroding.[8] It’s an interesting contradiction, and one which will bring its own excesses. With time we may remember parts of the current economic era fondly.

One danger I see is that this increasing technological displacement and growing pool of unemployables could strengthen, poison, and radicalize the OWS movement, giving it a louder voice and a weaker argument (re: living wages) at the same time. I think it’s a real possibility. We’ll need to figure out some sort of a fair solution to all of this before things turn ugly. I just don’t know how yet.

Interesting times.

Notes:

[1] Is there anything good about Goldman? In short, yes. Aside from the usual dealmaking, capital allocation, and underwriting tasks investment banks perform, Goldman’s known for being willing to make new markets — being willing to serve as a middleman (and sometimes counterparty… and sometimes both(!)) on unusual sorts of wagers (such as CDOs). This isn’t always economically healthy, as it can increase market complexity faster than regulators (or participants) can understand what’s happening, but being able to effectively buy or ‘short’ certain behaviors can also bring substantial market efficiency.

[2] Here’s Taibbi again, on Goldman and financial regulations:

The way I look at it is that this isn’t really a financial story, it’s a political story. It’s about how power works in America. They’ve figured out a way to hide power in these little regulations and the minutiae and it’s like a gigantic bulwark that separates ordinary people from those of influence. So in order to be a journalist you have to go through that whole maze. You need space and you need time, and nobody has it anymore….

   Do you still follow Russian politics? They’re gearing up for an election. I tend to envision Putin placing Medvedev on a platter and eating him on live television, or something of the sort.

They’re just your basic third-world kleptocracy—which is where everybody is headed. Well, everybody who still has a functioning government.

   Elaborate.

I think people are going to realize what a blip on the radar American-style democracy in the 20th century was. A big middle class that had a huge powerbase, financial interests, bosses giving benefits… all those things. It’s just a little blip in history. For the most part, concentrated wealth will make all the decisions and everybody else is dictated to.  

When it’s not uncommon for large, profitable corporations to pay their CEOs more than they pay in federal income tax, I think we have to take this notion of a sea change seriously.

Will OWS turn to economic vigilantism as a force for change if political disobedience fails? And if they do, can we blame them?

[3] Many liberal groups would love to use OWS to advance their goals; while I was at Occupy LA there were activists (mostly racial and environmental) who were trying hard to get their pet causes into the mission statement. I think it’ll take a lot of energy to fend off opportunists and keep focused. The Democratic Party in particular is drooling over the possibility of co-opting OWS (just as the Tea Party has been co-opted), much to the movement’s irritation:

The corporate media is anointing a false leader of the Occupy Movement in Van Jones of Rebuild the Dream.

The former Obama administration official, who received a golden parachute at Princeton and the Democratic think tank Center for American Progress when he left the administration, is doing what Democrats always do—see the energy of an independent movement, race to the front, then lead it down a dead end and essentially destroy it. …

At Occupy Washington, DC, we recognize that putting our time, energy and resources into elections will not produce the change we want to see. What we need to do right now is build a dynamic movement supported by independent media that stands in stark contrast to both corporate-bought-and-paid-for parties.  

Democratic operatives want to steal the energy of the Occupy Movement because they do not have any of their own.

[4] I personally think humans have a really good ‘unfairness detector’ module. Give us a complex math problem, and most of us will wilt; give us a card game (of similar abstract complexity) where someone’s cheating, and many of us will figure it out. But again, our capacity here is not infinite.

[5] A large factor in wages is simple supply-and-demand. People get paid more for doing what the market needs. And employment supply-and-demand is notoriously dynamic. A friend relates that the boom in domestic oil exploration has led to people with degrees in petroleum geology landing jobs paying over a million dollars right out of college. People who are willing to move to North Dakota and drive trucks for the oil economy are making over $120k. Meanwhile, people with degrees in English Literature and Social Work are competing with hundreds of other applicants for starting salaries in the low 30ks, salaries that look even bleaker once today’s ridiculous student loan payments are factored in. An average of 40% of college grads resort to jobs that don’t require degrees, a number which is steadily rising.

Is this fair? Do people who choose to study e.g., art history deserve to make living wage working in their field? I think “deserve” is a loaded term here. The top 1%-.1% of a field will always be able to write their own ticket, but supply-and-demand dynamics will drive compensation for the rest, with little regard for concepts such as a “living wage”.

Of course, supply-and-demand doesn’t happen in a vacuum – policy plays a huge role in creating the employment context it emerges from. We could learn a lot from China in how to nurture emerging domestic industries (or at least not smothering them on the altar of the status quo). Part of this is willingness to experiment– China is now limiting college majors which are correlated with high post-graduation unemployment, an interesting move. Immigration policy matters. Education matters and we could be doing better (though perhaps we tend to be too hard on ourselves, mistaking demographics for quality of instruction).

[6] The unspoken assumption of the Friedman school is that once we provide good, rugged, and cheap internet devices to the third world, they’ll be able to lift themselves out of poverty by educating themselves, learning foreign languages, science, programming, and design via something like OpenCourseWare, and joining the global economy. It’s a great idea, certainly worth doing, and will work for some. However, regardless of where one stands on IQ as a primary component of economic potential or a lagging correlation which rises with prosperity, I think it’s betrays an overenthusiastic estimate of the quick malleability of human capital, and I also doubt the Vast majority will be able to bootstrap before algorithms eat their niches.

[7] There is precedent here: many people have argued that welfare got its start essentially as protection money paid to those in LA’s slums, to give them something to lose so they wouldn’t burn down the place.

[8] My father, on economic policy:

Natural selection is the way things work, but I should say that I’m not at all comfortable with the results – when it comes to people.  It is easy to see that the benefits of living in this country are unequally distributed, and, as you point out, that can be beneficial.  I think we need to ask some questions, however.  Who “owns” this country?  How much control should someone who wants to work hard and make lots of money have over someone who does not wish to devote his life to making money?  The two are not separate, as one will drive up prices of goods and that will affect the other – right?  Does one person have the right to produce products offshore and sell them freely here in the US, even though it makes the labor of the other person much less valuable?  And, does the laborer have the right to band together in a union to gain advantage over other workers – using the political process to mandate that advantage?  If you take all the laborers here in the US, do they have the right to demand that manufacturing jobs be returned to the US if the investors wish to sell the goods here?  (This would simply be a much larger labor union, wouldn’t it?)

Politics is the art of taking money from one group and giving it to another.

…

And, I think we need to have a discussion about not only the benefits of living here, but we need to talk about the responsibilities too – on both sides of the spectrum.

(1) physics is ultimately the foundation for basically everything,

(2) when we get down to details it’s pretty darn weird, and

(3) while most fields have moved away from metaphysical questions and toward inaccessible problems of complex emergence, there are still cool, unsolved, fundamental mysteries in physics.

People seem really engaged by the weirdness and mysteries in theoretical physics, even to the point of feeling an ownership interest in them, and I think that’s awesome.

And so with this year’s Nobel Prize in Physics announced, I wanted to give readers a quick rundown on current Big Mysteries in Physics. It’s not a comprehensive list[1], but I argue that most other questions will ultimately trace back to these three.

1. How do we combine General Relativity and Quantum Dynamics?

Right now Physics rests uneasily on two fundamental theories. General Relativity deals with relationships between spacetime, velocity, and gravity (generally speaking, properties associated with large objects) and is amazingly predictive at what it does. Quantum Dynamics deals with sub-atomic particles, the quantized nature of the strong, weak, and electromagnetic forces, and the weird statistical rules these things obey (generally speaking, properties associated with very small things), and is amazingly predictive on quantum scales. We have one theory for big things like planets and spaceships, and another for small things like electrons and quarks.

The trouble is, the math– and metaphysical assumptions about reality– of these two theories are very different, and we don’t have a good way to fuse them together to talk about things like black holes or the big bang, things which straddle both the quantum and relativistic. Most physicists find the situation very troubling, not to mention deeply ugly, since it feels like the universe must have a single set of rules, not two. Presumably, if we found a more general model which explained each theory as a special case of a more general system, all sorts of little mysteries in physics might solve themselves (just like the theory of DNA solved lots of mysteries in biology). String theory, quantum gravity, and other, even more esoteric field theories are attempts at unification, but to date no attempt at unification has made any successful prediction that departs from what each separate theory suggests.

2. What is Dark Matter?

There are two huge fudge-factors in physics. One is Dark Matter– a hypothetical sort of matter that interacts with other matter only via gravity (“dark” means “we can’t see it”). It was introduced in 1934 to explain why galaxies rotate so fast: according to our equations, without this fudge factor, many galaxies rotate fast enough that they should simply fly apart. However, instead of disappearing quietly like fudge factors often do, we still need it today to explain galactic dynamics and certain other observations. Most cosmologists agree that it’s very probably not just an artifact of some mistake in our calculations, but some very real and very mysterious type of matter.

What we know: based on our calculations ~83% of all matter is “dark”. We think this dark matter is found in most or all galaxies, and there’s a good chance some passed through you as you read this. There are conflicting theories about where it’s most heavily concentrated– some models have it primarily concentrated in the dense center of galaxies, some have it more spread out, some in a halo. We’re pretty sure, whatever it is, that dark matter is “cold” — i.e., not moving at a significant fraction of the speed of light. There are a lot of experiments trying to conclusively detect dark matter, either (1) from its gravitational effects or (2) directly, if dark matter happens to occasionally interact (‘weakly interact’, in the lingo) with normal matter. (A shoutout here to the Sanford Underground Laboratory, which is in the running, and within spitting distance from my folks’ house.)

3. What is Dark Energy?

The 2011 Nobel Prize in Physics was awarded for discovering a mystery: that our universe’s expansion hasn’t slowed down since the Big Bang. In fact, it’s actually sped up. And we have no idea why.

The standard assumption prior to 1998 was that our universe was either going to contract due to gravity (the “big crunch”), or was somehow exactly balanced (Einstein’s static universe hypothesis), or that the initial energy from the Big Bang would keep the universe expanding, albeit ever more slowly as gravity tried to pull everything together.

An examination of a specific type of star explosion– Type Ia supernovae, which due to various mass dynamics all explode with roughly equal energy and brightness– provided a basis for an historical record of the universe’s expansion. Since we know how much energy is released in these explosions, we can calculate how far (which is another way of saying ‘how old’) it is based on how bright it is for us. Likewise, if it’s moving toward us, the light will be “blueshifted”, or if it’s moving away from us, it’ll be “redshifted” (think of how a siren’s frequency changes depending on whether it’s moving toward or away from you).

What we found when we put these things together was that basically everything is moving away from us, but– here’s the kicker– the closer, newer stars are moving away from us proportionally faster than older stars. The universe is not only expanding, but the expansion is accelerating.

Cosmologists don’t know what’s causing this. The convention has been to refer to it as “dark energy” since the cause of the expansion is generally fudged-in as an energy term in our equations, but we don’t really know if it’s a hidden form of energy, an emergent property of space, or something even more esoteric. There are theories, but they tend to be mathematically inelegant – and given our lack of a high-resolution expansion timeline, remain little more than untested guesses. If it is actually energy, there’s a lot of it:

Pretty large for a fudge-factor. Image credit: NASA.

A shameless plug for a pet theory:

I don’t know what Dark Energy is, but I do actually have a guess. If you’re in the mood for some cosmological speculation, and particularly if you’re in a position to give feedback on such, I encourage you to check it out. Like any new theory, it’s probably wrong– but based on my reading of the field, I don’t think it’s more likely to be wrong than other theories on the topic, and throwing one’s hat in the ring is how science progresses.

A Suggested Model for Dark Energy.

[1] Another major mystery is why there’s way more matter than antimatter in the universe. “Antimatter” sounds so weird and esoteric, but it’s actually rather common– there’s probably lots of antimatter popping in and out of existence in the room you’re in now. We commonly create antimatter in labs, and it actually forms the basis for tech like PET scans. It’s just that matter is WAY more common, and there’s no a priori reason we can see that this should be the case. I talked a bit about this in my 2008 obituary of John Wheeler.

In practice, even though we have billions of nucleotides, our DNA cannot follow billions of genealogical lines. Recombination over 30 — 40 generations does not divide chromosomes down to individual nucleotides. In the medium term, most human DNA is separated by recombination hotspots into lengths of around 50 kilobases. Across very short spans of 30 generations, DNA is for the most part inherited in chunks of hundreds of kilobases or longer. So dividing six billion nucleotides by 50 kilobases yields a number of around 120,000 ancestral lines at most from which any individual inherits his or her DNA. Recombination will increase this number somewhat further and further back in time, but not nearly so fast as the doubling of possible ancestral lines in every generation. This means that the vast majority of your ancestral lines more than around 17 generations ago have left no DNA to you whatsoever.

Granted, this is relative to the massive redundancy in our family trees– humankind is one huge, partially-inbred extended family. I.e.– if you go back 40 generations, you have over a trillion great-great-great-(etc) grandparents. There weren’t a trillion people alive in 1000AD, so a lot of those slots were filled by the same people.

(This is really technical and hypothetical; if you don’t enjoy mathematics and speculative neuroscience and would prefer alternative entertainment, why not check out these captioned pictures of cats instead?)

Similarities between music and thought: keys and chord structures?

In short, I suspect that we’ll find the structure of frequency patterns in music to be rough reflections of the structure of frequency patterns within our heads. We’ll be able to look at brain frequency patterns via many of the same types of signal analysis we normally apply to music. That’s suggestive but a bit vague, so here’s a specific prediction: brain frequency patterns will have identifiable chord structures, and these chord structures are what inherently encode affect in the brain.

E.g., the direct, mechanistic cause of happy/sad affect is that the frequency of one’s neural firings is arranged in a strongly major/minor chord pattern. Being sad means having thoughts which are literally in a minor key– i.e., the same sorts of mathematical ratios between frequencies that are universally present[1] in minor chords will also be present in someone suffering from Depression.*[2]

More generally, I think we’ll unearth that music theory is actually neuroscience in disguise, and we’ll need to borrow liberally from the mathematics of music to accurately model the mind. To put this slightly differently– music is mathematics, and our brains’ frequency dynamic is mathematics, and I suspect the mathematics of each are similar. E.g., the mathematics of a happy song are similar to the mathematics of a happy mood.

What kinds of mathematics might neuroscience end up borrowing? I’d point to

(1) the specific, mathematical ratios which define and constrain music (which may also generate identifiable emotion when in the context of neural activity);

(2) the mathematics of resonance, acoustics, and dissonance (which may richly describe brain region system dynamics and the neurophysical dynamics of mood);

(3) the modeling of musical structure as geometric constraints within a non-euclidean space (which may also describe bounds on the sorts of neural activity which give rise to emotion).[3]

Very speculative, I know. It’s a testable hypothesis, so we should know eventually.

*Caveats: a great deal of how we interpret a chord is context (both musical and cultural), and our emotional reaction is not determined by the chord. “Major=happy,” “Minor=sad” is of course a gross simplification of a complex, multivariate system— a more nuanced theory would include e.g., dissonance, rhythm, the expressive necessity of improvisational imperfections, chord progression, and novelty, since the brain has myriad anti-feedback loops. Just as context is key in how we interpret music, so too it will be in neural patterns. Maher (1976), among many others, suggests that the physiology of musical interpretation involves some cultural plasticity. However, our emotional reaction is not independent of a chord either, and I’m building my hypothesis of a structural homologue around this core of physiologically-dependent interpretation.[4][5]

Toward a psychophysical law?

To go even further, putting on my Philosopher’s hat, this is a possible start on a psychophysical law. What is a psychophysical law? Essentially it would be a translation function between the brain and qualia. “When X (in the material world), then Y (in the world of sensation/emotion/thought)”. We don’t really know any yet, but we know they’re out there.*

*I tend to agree with David Chalmers (and perhaps he’d even agree with my paraphrase) that ‘qualia arise from computation within systems which are above a certain threshold of a certain sort of organizational complexity’. My hypothesis is an attempt to quantify how the frequency dynamics of such a system constrain the associated qualia.

With something this abstract, examples are really important. Consider the following questions which we lack the answers to:

To answer these sorts of questions, we need psychophysical laws (like the one I’m proposing). Moreover, such laws are absolutely necessary for such tasks as

At any rate, I’d like to extrapolate my ‘chord theory of emotion’ to all computational systems, and subsume the human-specific stuff under this. I.e., given that this chord theory IS right, then… it’s not a quirky biological coincidence that certain relational frequency structures lead to qualia with certain properties, or that coherence feels good and dissonance feels painful, but rather a specific realization of a more general, fundamental, invariant, and universal psychophysical law.

Yeah, I know it’s a weird topic. But it’s important!

*I think understanding the universe’s psychophysical laws is as important to designing a friendly AI (FAI) as understanding aerodynamics and gravity is to designing a safe aircraft. You can muddle through for a while without a good, formal theory, but it’ll probably result in a terrible accident at some point. I don’t think people in AI research (or even FAI research) see this point.

[1] What about atonal societies? I suspect, following this theory, that they experience emotions differently than those from tonal societies. Having said that, (1) atonal music isn’t devoid of structure, and (2) perhaps there are cognitive-functional reasons why most societies (and/or the most successful societies?) are tonal.

[2] Other factors play a part in depression too, of course. E.g., the ‘dullness’ of a depressive mood could come from dissonance within a brain, or the general inability to carry a strong signal. I would say, more accurately, minor chords will resonate much more strongly than major within a Depressive’s head. Generally speaking, I suspect the sorts of music that resonate with us are pale shadows of the types of complex, ordered patterns which are resonating in our heads.

[3] On (3), ‘music structure as geometric constraints within a non-euclidean space,’ I’m reminded of this post by John Hawks:

The lead report in Science this week was this paper by Dmitri Tymoczko, titled “The geometry of musical chords”:

A musical chord can be represented as a point in a geometrical space called an orbifold. Line segments represent mappings from the notes of one chord to those of another. Composers in a wide range of styles have exploited the non-Euclidean geometry of these spaces, typically by using short line segments between structurally similar chords. Such line segments exist only when chords are nearly symmetrical under translation, reflection, or permutation. Paradigmatically consonant and dissonant chords possess different near-symmetries and suggest different musical uses (Tymoczko 2006:72).

Like a lot of things mathematical, the mathematical description of this is fairly distant from everyday experience. Cosmic Log provides a pretty good summary of the mathematical connections. This is pithy:

For years, string theorists have used music as a metaphor for fundamental particles, and now Tymoczko is using the mathematics of string theory to understand the fundamentals of music.

The next couple of paragraphs capture the essence of the work:

The math makes it easier to understand objectively what great musicians and composers do in their head. “When you sit down to interact with a piano, you’re actually interacting with a non-Euclidean space, because there are many different ways you can play a C-major chord on a piano,” Tymoczko said.

He said orbifolds capture the multidimensionality of music: how harmony interacts with counterpoint, how chords are connected with each other, even how notes are arranged “to minimize the amount of effort that musicians have to make when moving from chord to chord.”

I think it helps to read a few different descriptions, and so I’m also linking the perspective in Science by Julian Hook, which includes some history, showing why Tymoczko’s paper is part of a long tradition of mathematical application to music:

Mathematical music theory, although terra incognita to practicing musicians and even to many professional music theorists, has in recent years blossomed into a sizable and multifaceted industry. Pitch-class set theory (3), the study of a discrete 12-note quotient space, was developed as a means of confronting the analytical challenges posed by “post-tonal” music of the 20th century, whose harmonic materials are more varied and complex than those in most earlier music. Diatonic set theory (4, 5) investigates the subtle and beautiful relationship between the 12-note chromatic scale and diatonic scales such as the C major scale, with seven unequally spaced notes per octave (a scale type of great importance in many styles of music). Scale theory (6, 7) studies structural properties of scales and their subscales more broadly, allowing variation in both chromatic and diatonic cardinalities and occasionally engaging considerations of tuning and acoustics.

…

A particularly active area is neo-Riemannian theory, which synthesizes modern group-theoretic techniques with inspiration drawn from the work of the prolific German musicologist Hugo Riemann (1849-1919) and his contemporaries. In its basic form (9, 10), neo-Riemannian theory investigates certain transformational relationships among the 12 major and 12 minor triads in ways that are algebraically elegant, musically suggestive, and readily visualized in various forms of a graph known as a Tonnetz (tone network), in which the harmonic path traced by a musical composition may be plotted (Hook 2006:49-50).

In other words, musical progressions form paths or shapes in multidimensional spaces. Music that is part of the classical Western tradition actually falls within a fairly restricted set of possible paths; other musical traditions also form paths that to a greater or lesser extent overlap (although the dimensionality of the spaces may be different for different systems).

[4] An alternate formulation of the general form of this hypothesis might deal with a structural homologue between consonance and dissonance in music and in the brain– e.g., the same patterns of consonance and resolved dissonance in music that we find pleasurable also obtain in our brain frequencies when we experience pleasure.

[5] What about people with autism who don’t experience any emotional expression from music? I would suggest that the music may get ‘summarized’ by their neural circuitry, but unlike ‘normal’ people, the musical pattern doesn’t actually enter the brain in any deep fashion where it can resonate.

Further Musings:

- A NYT article explores what makes music ‘expressive’ and touches on the concept of there existing natural, inherently pleasurable body tempos:

Anders Friberg, a music scientist at KTH Royal Institute of Technology in Sweden, found that the speed patterns of people’s natural movements — moving a hand from one place to another on a desk or jogging and slowing to stop — match tempo changes in music that listeners rate as most pleasing.

“We got the best-sounding music from the velocity curve of natural human gestures, compared to other curves of tempos not found in nature,” Dr. Friberg said. “These were quite subtle differences, and listeners were clearly distinguishing between them. And these were not expert listeners.”

I wouldn’t be surprised to find these same tempos in other physiological contexts. Moreover, I suspect these tempos will be elegantly derivable from what music we find enjoyable, thus supporting my hypothesis of a structural homologue between thought and music.

- I’m open to the idea (and perhaps metaphysically forced into) that the qualia we experience are only a fraction of the qualia associated with the mass that makes up our bodies. Does a cell have qualia? Likely so, though of a simple sort since it lacks a sense of self, but if it does we can’t feel it. Does this change anything? Perhaps the qualia associated with the cellular strain involved in a heavy-drinking night out on the town is ethically significant. Who knows.

Mapping the connectome is a great step toward understanding the brain. The problem is, what do we do with a connectome once it’s built? There’s a lot of important information about the brain’s connectivity packed into a connectome, but how do we extract it? Read on for an approach to broad-stroke, comparative brain region analysis based on frequency normalization. (Fairly technical and not recommended for a general audience.)

The challenge: turning connectomics data into knowledge.

There’s a worm called c. elegans that scientists study a lot because it’s so simple– its whole body has only 302 neurons. We’ve made a lot of progress on understanding it, but we’re still working on figuring out what each neuron does and mapping the logic of its neural circuits. With the human brain having a hundred billion neurons, it’s clear that we need some big-picture tools to make sense of all the wonderfully combinatorial complexity in our brains.

It’s not an easy task: what sort of analytical lens can we use to study the brain’s neural networks that elegantly illuminates system properties and differences between people, simplifying without being too simplistic? What sort of analysis ‘carves our brains at the proverbial joints’, to paraphrase Aristotle?

A thought: frequency is the internal language of the brain.

The brain is horrendously emergently complex— but all the really interesting signal encoding and processing in the brain is frequency based. Most likely, the texture of our cognition and emotion derive from the high-level frequency dynamics in our brains. Even more strongly and less vaguely, I’d argue that most other elements of our neuro ontology (neurotransmitters, action potentials, fMRI activity, etc) are functionally important only insofar as they influence, or are proxies for, the frequency dynamics of the brain.

It’s a strong statement! But I think it’s warranted. Following this, if we want to understand neuron activity on large scales we’ll eventually need to enable comparative frequency analysis. To normalize neural networks somehow to allow apples-to-apples frequency comparisons between neural circuits or brain regions or people.

An approach: can we construct a normalized framework for analyzing neural networks by flattening them into a one-dimensional bundle of interconnected pathways?

In the previous section I said we’ll need to enable comparative frequency analysis to really understand and predict neuron activity on a large scale. But this is a very hard problem. It’s hard because frequency analysis (e.g., performing a fourier transform) depends on a certain sort of structural consistency in the medium under measurement. There’s a difference in kind between sound waves bouncing around a room and electrical pulses bouncing through a neural network. They’re both based on frequency, but neural networks have complex topology and state*, whereas sound does not. If we want to analyze network frequencies on scales larger than individual neurons (which do not shed much light on overall frequency dynamics), and particularly if we want to apply non-trivial amounts of audio terminology (which seems like a no-brainer to me), we need to control for these differences.

Leaving aside the question of internal state of neurons for now, I have a suggestion on taming the topological issue such that we could attempt structure-normalized, apples-to-apples comparisons of high-level frequency dynamics between neural circuits or brain regions or people.

In short, we could try to computationally flatten out the pathways of a neural network (e.g., a brain region’s connectome) into a one-dimensional bundle of interconnected pathways. Essentially it’d involve trying to flatten the network into a semi-redundant, 1d, massively parallel, linear dataset, of which we could take cross-sections to measure frequency profiles, and which can be easily compared between neural networks.

There would be lots of technical hand-waving required… and lots of pruning too, given the combinatorial nature of the dataset. E.g., perhaps we’d need to filter pathways via minimum threshold of feedback loops. Fully appreciating the benefits, drawbacks, and process of this transform would take much more mathematics and expertise with complex topologies than I have. But I believe it could be an approach that drastically reduces the complexity of a neural network while still preserving many important network properties, and enables comparative frequency analysis on the network.

Ultimately I think the best way to understand large neural network will not be in terms of neuron connection statistics, but in terms of the frequencies and rhythms the network can generate and support. To go a bit further, I believe this structural transform also makes it easier to quantify resonance inside neural networks (and treat a brain region as an acoustic chamber, in which certain patterns resonate much more strongly than others), something I think is very important.

*A little more on the complex topology and state of neural networks: neural networks have ‘state’ in that each neuron is a unique device with its own internal variables. We can’t understand exactly how the network works, and what frequencies and rhythms it can support, unless we can decipher the internal settings of each neuron in the network (and this changes over time). Contrast this with sound, where the signal medium doesn’t have state.

Likewise, neural networks have ‘complex topology’. Sound waves travel in straight lines in three dimensions (“simple/euclidean topology”). However, signals in neural networks travel along the connections between neurons, which are definitely not arranged in neat, straight lines. It’s more like high-dimensional, non-euclidean space, an “Alice in Wonderland” style area where things don’t move in straight lines so much as between a tangled web of nodes. This is NOT to say that there is NO frequency to be analyzed, nor that other elements of audio theory (resonance, harmonics, constructive/destructive interference) aren’t present. There is and they are. It’s just that we need to find SOME way to simplify the topology before we can start to quantify the frequency data and apply concepts from audio theory. And if we ignore the problem and don’t bother with this sort of simplification/normalization, we’ll never get good frequency data, period.

At any rate, this is a possible approach, and it may or may not end up workable. But however we enable frequency analysis inside the brain, I hope more people start thinking seriously about the normalization problem soon. Frequency is of central importance in neural function, and having a method to normalize frequency within the brain will be central in understanding how it all works.

Image: Neurons from the limbic area of the Central Nervous System. Image copyright and courtesy of Bryan W. Jones.

[1] I’m reminded of this curmudgeonly letter regarding the ‘omics’ situation:

It is an old maxim that if you want to get on, invent a new word for your particular niche in an old area, and so become an instant expert. This process seems to have gone mad. A recent article in The Scientist that referred to “nutri genomics” [1] prompted me to see just how many -omics had now been coined. Well over 100 neologisms are listed at http://www.genomicglossaries.com/content/omes.asp. A few of the more ghastly examples are foldomics, functomics, GPCRomics, inomics, ionomics, interactomics, ligandomics, localizomics, pharmacomethylomics and separomics. None of these refers to areas of work that did not exist before the coining of the new word. Perhaps, as an electrophysiologist working on recombinant ion channels, I should dub myself an expert on ohmomics.

This habit of coining fancy words for old ideas might be thought harmless, merely a source of endless mirth for thinking scientists. I’m not so sure though. Apart from reinforcing the view of scientists as philistine illiterates (at least when it comes to etymology), actual harm is done to science as the public becomes aware that some among us seem to prefer long words to clarity of thought.

David Colquhoun

University College London

Indeed, playing against Watson turned out to be a lot like any other Jeopardy! game, though out of the corner of my eye I could see that the middle player had a plasma screen for a face. Watson has lots in common with a top-ranked human Jeopardy! player: It’s very smart, very fast, speaks in an uneven monotone, and has never known the touch of a woman. But unlike us, Watson cannot be intimidated. It never gets cocky or discouraged. It plays its game coldly, implacably, always offering a perfectly timed buzz when it’s confident about an answer.

Perhaps somewhat less funny from Ken’s perspective, a question asked during his reddit interview:

How’s it feel to be owned by something that asked “What is leg” ?

Richard Hamming used to go around annoying his colleagues at Bell Labs by asking them what were the important problems in their field, and then, after they answered, he would ask why they weren’t working on them. Now, everyone wants to work on “important problems”, so why are so few people working on important problems? And the obvious answer is that working on the important problems doesn’t get you an 80% probability of getting one more publication in the next three months. And most decision algorithms will eliminate options like that before they’re even considered. The question will just be phrased as, “Of the things that will reliably keep me on my career track and not embarrass me, which is most important?”

And to be fair, the system is not at all set up to support people who want to work on high-risk problems. It’s not even set up to socially support people who want to work on high-risk problems. In Silicon Valley a failed entrepreneur still gets plenty of respect, which Paul Graham thinks is one of the primary reasons why Silicon Valley produces a lot of entrepreneurs and other places don’t. Robin Hanson is a truly excellent cynical economist and one of his more cynical suggestions is that the function of academia is best regarded as the production of prestige, with the production of knowledge being something of a byproduct. I can’t do justice to his development of that thesis in a few words (keywords: hanson academia prestige) but the key point I want to take away is that if you work on a famous problem that lots of other people are working on, your marginal contribution to human knowledge may be small, but you’ll get to affiliate with all the other prestigious people working on it.

And these are all factors which contribute to academia, metaphorically speaking, looking for its keys under the lamppost where the light is better, rather than near the car where it lost them. Because on a sheer gut level, the really important problems are often scary. There’s a sense of confusion and despair, and if you affiliate yourself with the field, that scent will rub off on you.

Academia does plenty of good things– but the opportunity cost of our systemic incentives toward ‘safe’ research (I include both the derivative and the esoteric) is rather staggering.

Edit, 8-13-11: A friend blogs,

The answer comes down to ethics. Service as an ethic is alien to so many academics. “I serve.” They don’t get it. Some do. A few. But a number of my friends have gone into the academy for longer or shorter periods of time, and the observations have always been similar – it’s not a place of scholarship and diligent service, but rather of all sorts of politics and backbiting where you desperately try to carve out your own private sphere in a confusing bureaucratic jungle.

I think academia used to have a strong shared sense of duty (Sebastian uses the term ‘warrior ethic,’ where service to a noble cause is its own reward), but for several reasons this has largely eroded or isn’t sustainable in today’s academy. It’s still present, but it’s much weaker. We could point to institutional factors, a changing demographic of who goes into academia, a crossover from our increasingly mercenary private-sector culture, or getting more of what we pay for, but at the end of the day– it seems like many people in academia think of it primarily as a career, not as service. It’s a big loss.

Modern neuroscience is increasingly saying that a great deal of a person’s personality, pathology, and cognitive approach is encoded into which of their brain regions are activity-coupled together. (In alternate terms, which regions are more vs. less wired together in a person’s brain.) Right now such coupling is largely invisible and unquantifiable. If we could test for this brain region activity coupling it would be invaluable- perhaps transformative- for psychiatric diagnosis and the study of individual differences. A combination TMS+fMRI alternated pulse device- as it could stimulate a specific brain region/network, and measure how it affected the activity in other regions- may very well be able to test for this and provide an objective basis for psychiatric diagnosis and treatment recommendations.

The following is a somewhat technical writeup of the idea. Not into detailed neuroscience stuff? Click here.

Technology:

A Transcranial Magnetic Stimulation (TMS) device able to perform precisely targeted regional stimulation paired with a functional Magnetic Resonance Imaging (fMRI) device, calibrated such that one will be active while the other is dormant, and able to toggle swiftly between the two. Stimulate one brain region (TMS), then observe which other regions light up (fMRI).

TMS devices function by lowering neuron activation thresholds by rapidly oscillating a magnetic field through the target area. They are commonly found in research institutions and are used for both research into brain region function (e.g., disable the function of a specific brain region by overstimulating it via TMS, and observe which cognitive functions are and are not affected), and in treatment for depression.

fMRI devices function by using the proxies of bloodflow and oxygenation to measure regional brain activity. More active regions need more bloodflow to function, and this increased level can be measured via applying a magnetic pulse and measuring the changes in local magnetic permeability.

Existing approaches:

Most current types of brain scans are ill-suited to gathering regional activity data, as they either tend to focus on physical neuroanatomy (CAT, MRI) or involve poor spacial resolution (PET, EEG, MEG). fMRI is currently the tool of choice for looking at brain region activity, and researchers have used it to look into whether, how, and why brain regions may be linked, and how variations in these linkages may affect cognition. However, the passive nature of fMRI complicates drawing any firm conclusions about the causality involved: e.g., one might show subjects pictures of spiders, or naked women, or christmas presents, and record their fMRI response, but what is really being measured? Are the results clear, reproduceable, and useful, e.g., enough such that they could form the basis of a diagnosis or treatment recommendation? Very often not. On the other hand, TMS Sonar bypasses the ambiguity involved in trying to induce a standardized abstract processing state and instead tests for objective, content-neutral activity coupling between brain regions, something fMRI cannot. In short- TMS Sonar could tell us new and interesting things about regional dynamics that existing brain scanning approaches will never be able to.

This approach in a nutshell:

Systematically ‘ping’ different brain regions with TMS and map which other regions light up via fMRI.

Types of data this could gather:

This device seems promising for exploring at least two different neural phenomena at the macro level: linkages and leakages.

Linkages:

An important part of how the brain works is the layout of which regions of tissue become activity-coupled. There seems to be significant natural variation in this, and it also seems to lie at the center of many sorts of brain disorders: i.e., a region should be strongly activity-coupled to a nearby region yet isn’t, or is activity-coupled to a region it shouldn’t be. These abnormal coupling patterns are for the most part currently invisible to us, but would likely show up under TMS Sonar.

Two particularly good examples of cases where TMS sonar could add to our existing analysis would be in PTSD and Depression, where the causal core may involve certain regions of the brain undergoing abnormal activity linkages or delinkages. Specifically, not only will many types of Depression likely center around activity coupling problems, but presumably different types of depression will have different and indicative activity coupling signatures, which can then inform more specific treatment recommendations.

The general hypothesis here is that activity linkages between brain regions vary considerably between people, and this will always have functional implications. Presumably these activity linkages will often correlate to a fair degree with interesting macro-level conditions, including some forms of psychopathology, and may be useful when diagnosing and designing individual treatments.

Leakages:

The other key use of this technology would be to identify and map microstructural variation as it pertains to information movement in the brain: i.e., how much information ‘leaks’ between different brain regions. Preliminary data indicates this varies across individuals and is implicated in certain types of pathology, and may inform larger, non-pathological differences in cognition.

A focal point for thinking about this issue is synesthesia. Examples of synesthesia include seeing numbers as inherently colored, or experiencing sounds in response to motion, in an automatic, involuntary way. It’s thought that this emerges from deficits in neural pruning between functional brain regions leading to cross-talk between them, and we know this condition manifests itself through sensory perception in roughly 1 out of every 23 people. However, it’s quite possible that this figure underestimates the prevalence and cognitive significance of synesthesia, since when synesthesia exists between two non-sensory regions of the brain there will likely be no consistent indicative symptoms (though it may still play a significant role in cognitive function). A pulse scanning method such as TMS sonar could likely map synesthesia across the entire brain.

The general hypothesis here is that everyone has a little bit of synesthesia, and a method to map individuals in terms of how much and which regions are affected could lead to great leaps in understanding certain forms of psychopathology and individual differences.

Hurdles:

- TMS is generally used to disrupt a brain region’s function by overloading it. The assumption that is made here is that either

1. a sufficiently low dose of TMS can be found to still effectively stimulate regional activity without disrupting regional function; or,

2. stimulation via TMS, though it may disrupt regional function, will still be an effective trigger for a representative sampling of activity coupling.

In other words, will TMS, as a general regional stimulation, evoke a coherent-enough activity response so as to be indicative of functional linkages?

3. fMRI doesn’t have great temporal resolution. I’m not certain this would matter for the purposes it’s being used, but it’s a potential caveat.

- Though both of these devices are currently in use, there may be engineering challenges in creating equipment that can handle both TMS and fMRI in rapid succession, particularly since they both use magnetic pulses in the course of their operation. The solution to this may involve some temporal latency in the fMRI measurement.

Discussion/Open Questions:

- This idea has a lot in common with Giulio Tononi’s approach for analyzing coma patients: stimulate their brains with TMS, and see how long it takes for the ‘ringing’ to die down. If it takes a while (like it does in normal people), they probably still have their higher cognitive functions intact. If it doesn’t, they’re probably vegetables.

- What would this process feel like?

- Could sonic stimulation work better than TMS in some cases?

- If, in the course of region coupling tests, specific abnormal brain couplings/decouplings were found, could this lead directly to a cure? E.g., use a TMS conditioning treatment where multiple regions were conditioned to activate, or not to activate, together via TMS pulses (fire together, wire together, etc). But this is quite speculative. More likely, knowledge of what brain regions were affected would allow traditional types of therapy to be more precisely targeted.

- How close is this sort of evoked Brain Region Activity Coupling (BRAC) to how BRAC works ‘in the wild’? Relatedly, how contextual is BRAC: do which regions are coupled together change significantly when one is in different moods, or will there be significant-enough commonalities such that this usually won’t be a problem in diagnosis?

- What are the underlying causes of BRAC? Relatedly, how plastic is it?

- Could this type of challenge-response hybrid be able to measure such things as relative dominance between brain regions, or a particularly high or low tendency for a region to fall into a coherent pattern given stimulation? What would the likely functional/psychological implications of this data be?

- It’s really unknown how central BRAC is in neural function. There’s not a lot of data out there because we can’t currently measure it with any degree of rigor or precision. Straight-up differential regional activity appears very important right now– but it’s difficult to see how it wouldn’t appear important for brain dynamics, as we model things based on what we can measure, and that’s one of the few functional measurable attributes.

- Could we compare evoked BRAC with coupling topologies derived from connectomes and diffusion-weighted fMRI?

Have you heard of the game NetHack?

It’s been in continuous development for the last 20 years or so. It’s all text-based graphics, very spartan in that sense, but those limited graphics make for extremely rich and deep gameplay and interaction with the world.

Oh, and it’s really, really fucking hard.

When you die, you’re dead forever. And it’s really easy to die.

Basically, every time you touch a key on your keyboard, a turn passes.

If you hold a direction key on your keyboard, 20 turns will pass and you’ll have moved 20 squares.

It’s very, very possible to have a threat emerge and kill you in 2-3 turns. The game requires extreme patience, caution, and planning to get through. Even that might not be enough, but it’s definitely required.

Beating NetHack is called “ascending” – I finally did it after a few years of playing.

And afterwards, I thought to myself – you know, I bet it’s easier to start a bank in the real world than it is to ascend in NetHack.

… Anyways. I haven’t started a bank yet. But I really seriously suspect it’s easier than beating NetHack. If you took 200,000 perfectly normal people and split them into groups of 100,000 – and half of them were instructed to beat NetHack and the other half were instructed to start a bank, and it was a really big deal if you succeeded or failed… I bet you’d get more new banks than NetHack ascensions.

His points:

1. If you can win a hellishly difficult video game, you should be able to do almost anything.

2. If you can structure your life like a video game– e.g., forgiving learning curves, point-based progression systems, rewards for difficult accomplishments, carefully selected addictions– it can really help better yourself.

I basically agree. Regarding #1, I think video games are somewhat akin to very broad IQ tests (probably a much better IQ test than the “standard” psychometric suite!), and as such don’t test for everything. There’s more to achievement than IQ. But if you can beat, say, any of the Civilization games on the hardest difficulty, it’s good evidence that you can handle complexity much, much better than most people. (Maybe you should start a bank!)

Some enterprising scientists (and Roy Baumeister, who has had his fingers in other interesting research) decided to check– if all pain is the same, would Tylenol help the sting of social rejection, too? They ran two experiments: in the first, they had volunteers take Tylenol for three weeks, right after waking up and just before going to sleep. In the second, they simulated social rejection by playing a game and not passing the ball to certain people. In both experiments, the group taking Tylenol reported less emotional pain than the placebo group. NIH paper; Pop writeup.

So next time you do something that could be painful- ask a girl out, get into a fight, do your taxes– take some Tylenol beforehand. It’ll help.