Month: August 2013

Criticism of My View of Education: My Answer

Seth Roberts2 Comments

My criticism of college education can be boiled down to this: It is too much one-size-fits-all. It takes too little account of differences between students. Those differences are no accident. They reflect the fact that a good economy needs to produce many different things. Human nature has been shaped to provide exactly that.

Bryan Caplan posted about this, and one reader (Tim of Angle) replied:

Roberts is criticizing colleges for not doing something that they aren’t really trying to do. . . . Our educational model is built around hiring teachers who are (supposedly) good at thing X and paying them to train other people to do thing X. Nobody claims that the way the teacher does thing X is the only way to do thing X, nor even the best way to do thing X; what colleges do claim is that the way the teacher does thing X is a successful way to do thing X, and it hopes that the teacher can train students to do thing X competently at least the way the teacher does thing X.

I was discussing undergraduate education at Berkeley. Berkeley professors are hired mainly based on their ability to do research. Undergraduate classes are not about training researchers (= the next generation of professors at research universities, such as Berkeley); that’s what graduate school is for.

In most Berkeley undergraduate classes, professors aren’t teaching students to “do” anything, at least anything that most of us would recognize as “doing”. (Engineering, art, architecture, foreign language and perhaps statistics classes are exceptions.) In most classes, students are introduced to an important fraction of an academic field. In a social psychology class, for example, they learn about social psychology research. The class is not about how to do social psychology. It is about what has been done and what has been learned. If the class consisted entirely of students who wanted to become psychology professors, that would be fine. In fact, only a small fraction of Berkeley psychology majors (5%?) go to graduate school in psychology. The students in most Berkeley classes (outside of the more vocational areas, such as engineering) will go on to do many different jobs. Few in any class will become professors.

I think one theory of higher education is close to what Tim of Angle says. The practice, at least at elite universities such as Berkeley, is quite different.

A different theory of higher education revolves around signalling. College performance provides a useful signal to future employers, that’s why it exists in present form. At Berkeley, I never heard this motivation (will this provide a good signal to employers?) brought up in discussions about grading or anything else. It’s utterly clear, on the other hand, that where you go to college (Harvard versus College of Marin) is indeed a powerful signal to employers and, yes, if you can go to Prestigious College X, you really should. How many “axes of excellence” there should be — how many separate categories or dimensions we should use to rank colleges — is a different discussion.

Rewarding Criticism Put Nicely Produced Long-Lasting Change

Seth Roberts2 Comments

Eliezer Yudkowsky, I’m told, used to be a not-nice critic. The problem was his delivery: “blunt, harsh, not sufficiently tempered by praise for praiseworthy things” (Alicorn Finley). However, this changed about a year ago, when Anna Salamon and Alicorn Finley decided to try to train him to be nicer. Alicorn describes it like this:

Me, Eliezer, Anna, and Michael Blume were all sitting in my and Michael’s room (where we lived two houses ago) working on, I think it was, a rationality kata [= way of doing things], and we were producing examples and critiquing each other. Eliezer sometimes critiqued in a motivation-draining way, so we started offering him M&Ms when he put things more nicely. (We also claimed M&Ms when we accomplished small increments of what we were working on.)

Eliezer added:

Some updates on that story. M&M’s didn’t work when I tried to reward myself with them later, and I suspect several key points:

1) The smiles/approval from the (highly respected) friends feeding me the M&Ms probably counted for more than the taste sensation.

2) Being overweight, M&Ms on their own would be associated with shame/guilt/horror/wishing I never had to eat again etc.

3) Others have also reported food rewards not working. One person says that food rewards worked for them after they ensured that they were hungry and could only eat via food rewards.

4) I suspect that the basic reinforcement pattern will only work for me if I reward above-average performance or improvement in performance (positive slope) rather than trying to reward constant performance, because only this makes me feel that the reward is really ‘deserved’.

Also:

  • Andrew Critch advises that ‘step zero’ in this process is to make sure that you have good internal agreement on wanting the change before rewarding movements in the direction of the change
  • The Center for Applied Rationality (CFAR) has some experience learning to teach this.
  • CFAR has excellent workshops but not much published/online material. A good mainstream book is Don’t Shoot the Dog by Karen Pryor.

I like this example because the change was long-lasting and important.

Assorted Links

Seth Roberts7 Comments

Thanks to Alex Chernavsky.

Autoimmune Disorder Improved With Fermented Food

Seth Roberts5 Comments

From a recent story in the Santa Cruz Sentinel:

[Kelly] Dearie turned to fermented foods in a moment of despair.

Her husband Charlie, who suffered from an autoimmune disorder that attacked his platelets, was told by doctors that he needed a spleen removal and a hip replacement. That would mean Charlie, an active 32-year-old man, would never be able to run or mountain bike again. . . .

The family decided to seek an alternative, and consulted Santa Cruz clinical health coach Craig Lane from Health Alkemy. . . . He checked Charlie’s temperature, blood pressure and lab results, and listened to Charlie talk about his diet, sleep and exercise. Instead of the surgeries, Lane recommended some dietary changes such as taking out coffee, wheat and sugar, and adding beet kvass, a traditional Russian fermented tonic.

Within three weeks, his platelet numbers were almost normal. Within two years he was running again, said Dearie. . . . Inspired by her husband’s healing, Dearie opened Creative Cultures and sells the beet kvass.

Dutch University Fires Unnamed Researcher

Seth Roberts5 Comments

If you google “Ranjit Chandra” (a famous Canadian nutrition researcher), the second result is this page, created by me, which lists many articles about a scandal that Saul Sternberg and I did a lot to to uncover. We pointed out that several details of one of Chandra’s papers were impossible. I did not create the page to harm Chandra, but it does: For the rest of his life, anyone curious about him will find out about the scandal. It is a scarlet letter with capital S and capital L.

I suspect this is why Leiden University recently fired a scientist without naming him/her.

Leiden University Medical Centre (LUMC) has fired an employee who has committed fraud in the collection of research data. An internal inquiry showed that the employee deliberately manipulated laboratory research. The employee has confessed and accepted the dismissal. Additionally, the LUMC withdraws two scientific publications by this employee. The fraud was discovered by immediate colleagues at the Rheumatology Department.

A deal was struck. The employee won’t contest the firing, the medical center won’t name the employee in the press release. The employee didn’t want the scandal to follow him/her for the rest of their life.

I disagree with this deal. As a result of the employee’s fabrication, a clinical trial was started in which sick people ingested or had injected a powerful drug. The university claims no one was hurt (“It is clear that at no time a dangerous situation has arisen for patients”). I have no idea if anyone was hurt, but the potential for damage was great. Last night a friend told me about a Traditional Chinese Medicine drug that a friend of hers took. It worked for years and then one day stopped working. It came from China. It turned out the Chinese manufacturer had run out of the crucial ingredient and had substituted an animal tranquilizer. Her friend was really damaged by this. Chandra’s data might have caused people to take too many vitamins.

The medical center employees who handled this case (presumably very high up in medical center administration) treated the rest of us — who deserve to be warned about the fabricator — not so differently than the fabricator did: as people who don’t matter. Who don’t deserve protection.

More A comment at Retraction Watch says the anonymity is Dutch tradition: “The names of the people are not published so that these people have a chance of rebuilding their lives in the future. In the Netherlands even people who have committed serious crimes do not have their full name or photo published in the press.”

Organic Pollutants Associated With Diabetes

Seth Roberts4 Comments

Everyone knows that diabetes is associated with obesity, probably because obesity causes diabetes. However, thin people also become diabetic. A clue to why is provided by the correlation between diabetes and what are called “persistent organic pollutants” (POPs). POPs are man-made organic compounds, usually pesticides, such as polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans.

A 2006 study using NHANES (National Health and Nutrition Examination Survey 1999–2002) data found very strong associations between levels of these chemicals and diabetes. For example, a risk ratio of 30. These associations persisted even when the data was stratified in all sorts of ways. The scariest result came from people who had BMI < 25. Looking only at such people, those above the 90th percentile for amount of POPs had 16 times the risk of diabetes as those below the 25th percentile. Here is something associated with thin people getting diabetes.

Does the association exist because POPs cause diabetes? You might argue that POP exposure is correlated with poverty (poor people are more exposed), poor people exercise less than rich people, and lack of exercise causes diabetes. However, Agent Orange exposure among soldiers is associated with diabetes. That is unlikely to be due to confounding with poverty or lack of exercise.

Everyone has these chemicals in their body, but almost no one knows how much. I don’t know if I’m in the 10th percentile or the 90th percentile. If I’m in the 90th percentile, what can I do about it? A good place for self-measurement and tracking.

Showers and the Ecology of Knowledge

Seth Roberts2 Comments

In a recent post, I said a well-functioning system will produce both optimality and complexity. I meant important systems like our bodies, economies, and formal education. If you look at the nutrition advice provided by the United States Department of Agriculture — the food pyramid, the food plate, the recommended daily allowances, and the associated reports — you will find nothing that increases the complexity of metabolism inside our bodies (in particular, the diversity of metabolic pathways). The advice is all optimality — for example, the best amounts of various micronutrients. The people behind the USDA advice, reflecting the thinking of the best nutrition scientists in the world, utterly fail to grasp the importance of complexity. Half of nutrition research — or more than half, since the topic has been so neglected — should be about how to increase internal complexity. In practice, almost none of it is. It’s obvious, I think, that the microbes within us are very important for health. They are mostly in our intestines and must be heavily influenced by what we eat. How did they get there? How can their number be increased? How can their diversity be increased?

The absence is especially striking because the point is so simple. To solve actual problems, you need both optimality and complexity. Showers — what we use to take a shower — provide an example. You want to adjust the water temperature. If you try to do this while taking a shower, it can be hard because of the delay between changing the hot/cold water proportions and feeling the effects. It is better to use the bathtub (lower) tap to set the temperature (measuring it with your wrist) and only after you’ve optimized the temperature, shift the water to the shower head. The bathtub tap produces simple output (a single stream of water) that is easy to optimize. The shower head produces more complex output that is harder to optimize but does a better job of washing (an actual problem). You need both bathtub tap (for optimization) and shower head (for complexity) to do a good job solving the problem. Likewise, we need both an understanding of necessary nutrients (Vitamin A, etc.), which can be optimized, and an understanding of microbes, which cannot be optimized but can be made more complex, to make good decisions about what food to eat. Ordinary food is the hardware, you might say; and microbes are the software.

 

Wild-Fermented Wine and the Ecology of Knowledge

Seth Roberts4 Comments

I learned about wild-fermented wine from Shana Reade, who teaches wine sellers about wine. She works for a New York wine distributor called Empire Merchants.

Before the 1950s, almost all wines were made with wild ferments. Only then did cultured (store-bought) yeasts start to be used on a large scale. The new wines surely tasted worse, but it was the era of TV dinners. The first cultured yeasts were especially popular in Australia, where less tradition blocked their adoption.

Nowadays wild-fermented wines are made in many places, including California, France and Germany. They are more expensive than cultured-yeast wines but you can buy one as cheaply as $15. Wild yeast is free, but the overall process is more costly because it needs more space and time. When you do wild fermentation, you put out vats of wine open to the air. The vat-to-vat variability goes way up and some vats will have to be thrown out. Wild fermentation also varies much more in how long it takes. Wineries rarely harvest all their grapes at once. With cultured yeasts, but not wild yeasts, they can be sure that one batch will finish before the next batch arrives.

Scientists have found that the yeasts in wild-fermented wines have thicker cell walls than the yeasts in cultured-yeast wines. This is an example of the general observation that microbes (and other living things) grown by man have fewer functioning genes and metabolic pathways (such as the metabolic pathways that build cell walls) than the wild type. Wild yeast, of course, has a more stressful and variable environment than cultured yeast. Cultured yeast loses functioning genes over generations because it does not encounter the problem they solve. There is no selection against deleterious mutations. Because wild yeast has more functioning pathways, it produces more metabolic byproducts, making a more complex flavor. This is a tangible version of the idea that we should use all our metabolic pathways. (A better version is use as many metabolic pathways as possible — fermented foods help with that.) So wild fermentation is (a) more diverse in terms of strains of yeast than cultured yeast and (b) individual strains of wild yeast have more functional metabolic pathways than individual strains of cultured yeast. (Cultured wine yeast starter includes several strains of yeast.)

The ecology of knowledge isn’t simple. Cultured-yeast wines (in the 1950s) were made possible by earlier wild-fermented wines. With cultured yeasts you can do wine experiments you could never do with wild yeasts, thereby learning how to make better wine in general. Today’s wild-yeast wines benefit from that knowledge. They also benefit from a mass market created by cheap (cultured-yeast) wine. An ecosystem that includes both sorts of wine spreads much further and produces much better wine than an ecosystem that includes only one sort of wine.

Personal science is like cultured-yeast wine in the sense that it allows far more experiments. Personal scientists can do experiments that professional scientists would find almost impossible. (For example, the effect of standing 8 hours/day on sleep.) A scientific ecosystem that includes both personal and professional science is going to solve problems far better than an ecosystem with only one of them.

More broadly, the story of wild-fermented wines illustrates how you need complexity and optimality — not just one of them — to solve actual problems (in contrast to artificial ones). Wild yeasts are complex, but not optimal; cultured yeasts are optimal but not complex.

A well-functioning system produces both complexity and optimality. This not-very-difficult idea is almost absent from modern thought. In nutrition, economics and education, for example, there has been almost no study of how to produce complexity.

Nutrition scientists have had little interest in fermented foods, which increase our inner complexity. Yes, as nutritionists say, we need good amounts of a long list of nutrients and micronutrients (optimality). In addition, however, we need inner complexity to solve actual problems, such as digesting food and fighting off pathogens. You can’t make a list of all the metabolic pathways we need to be healthy — it might be in the hundreds of thousands. You’d never learn our need for complexity from any nutrition book, as far as I can tell.

The science of economics revolves around optimality (e.g., most profit). I believe the current stagnation of the American economy is partly due to the poor understanding of economists of how to produce economic complexity. If they don’t know, neither will anyone else. Interest groups, rich and poor, have no interest in complexity. (Illustrating The Stupidity of Crowds. The Wisdom of Crowds is about optimality.) All sorts of policies are too narrowly evaluated. Their effect on optimality is assessed (how will this affect growth of GDP? or how will this affect percent unemployed?), but not their effect on complexity.

As for education, it is a good idea to push students to be better (push them toward optimality, e.g., be better at math). But a single-minded emphasis on optimality (e.g., No Child Left Behind), with no value placed on complexity, is a disaster.

Stagnation in Psychiatry

Seth Roberts6 Comments

A recent New York Times article lays it out:

Fully 1 in 5 Americans take at least one psychiatric medication. Yet when it comes to mental health, we are facing a crisis in drug innovation. . . . Even though 25 percent of Americans suffer from a diagnosable mental illness in any year, there are few signs of innovation from the major drug makers.

The author has no understanding of the stagnation, yet is opinionated:

The simple answer [to what is causing the stagnation] is that we don’t yet understand the fundamental cause of most psychiatric disorders [what does “fundamental cause” mean? — Seth], in part because the brain is uniquely difficult to study; you can’t just biopsy the brain and analyze it. That is why scientists have had great trouble identifying new targets for psychiatric drugs.

The great increase in depression has an environmental cause. Meaning that depressed brains (aside from the effects of depression) are the same as non-depressed brains. Someone who knows that would not talk about biopsying the brain.

You come to a room with a door. If you don’t know how a door works, you are going to do a lot of damage getting inside. That is modern psychiatry. I described a new explanation for depression in this article (see Example 2).

Thanks to Alex Chernavsky.

Assorted Links

Seth Roberts4 Comments
  • The increasing popularity of kvas. “We ferment with ginger and, I believe, longer than other people – for seven to 10 days.”
  • Giving up wine (and other alcohol) for a month. Before this he drank 2 glasses of wine/day.
  • Wellness Mart (in California) makes it easy to get basic medical tests. “ In California, you are required to have an order from a doctor for blood tests, but WellnessMart, MD stores all have medical doctors on staff. Our doctors allow their license to be used for basic screening tests because there are some things that really shouldn’t be that difficult to find out. If you don’t have a doctor’s order and you want to run tests that aren’t a part of our standard screening packages, you will be charged a MD Consultation Fee of $25. Our doctor will help you to put together a panel that will accomplish the goals you are looking to accomplish. If the doctor determines that it is not appropriate for you to run the tests you want to run at WellnessMart, MD there will be no charges.”
  • Riding a bike while learning Polish. It helps.

Thanks to Casey Manion and Adam Clemens.