The Robot Diet


The robot works by talking to you about how much you’re eating and exercising. It helps people stick to their diets by verbally asking dieters to input data about what they ate on a touch screen. The robot then provides encouragement and advice.

More. Social facilitation effects are powerful. For example, if you are riding a bike with someone else you will be able to go faster and farther than if you ride alone. Next, perhaps: A robot that judges your appearance (praise only!).

Thanks to Dave Lull.

Interview with Kamran Nazeer (part 2)

ROBERTS I got the sense reading your book that the autistic persons got more pleasure from focusing on small details of things. Those details were comforting. Like a favorite piece of music, or whatever; anything you enjoy is comforting. So when they were uncomfortable, they would go to an unusual place. If somebody likes small details, they’ll pay more attention to them, simple as that. Because if you don’t you’ll do something else with your attention.

NAZEER We get taught that it’s more socially useful not to focus on the small details. Sometimes we get taught that in explicit ways, or sometimes that’s just the way things are. So people forget about small details, because they’re focusing on something that’s more socially useful.

ROBERTS When you say focus on the small details, you gave an example. The opposite of a small detail is where the bus is going; that’s sort of a big thing. So what’s another example that’s opposite of the small detail?

NAZEER I’ll give you an example of Craig, who’s the speechwriter, also in the book. Craig will often find that he’s sitting in a meeting and he’s supposed to be focusing on the political issue or the speech that is the topic of discussion at that meeting. That is the thing that he’s supposed to be focusing on professionally. But he’ll often find difficult is that he’s noticing lots and lots of other small details, as well. He tries very hard to keep his focus where it ought to be, but because he keeps noticing these other small details, they can drag him away from where he’s supposed to be focusing.

ROBERTS Details of the meeting, or details of the argument, or what?

NAZEER It could be that. So he could be seeing a level of nuance that actually isn’t all that useful. Because sometimes when you’re in a meeting, you have to ignore certain nuances to get the bigger points. Or it could just be sometimes that he’s focusing on the fact that somebody’s missed a button when they buttoned up their shirt, or that their cufflinks are unusual, or that they’re flicking their pen. He might notice and focus on things that are completely irrelevant to the conversation he’s supposed to be having.

Interview directory.

Interview with Kamran Nazeer (part 1)

Send in the Idiots: Stories from the Other Side of Autism (2006) by Kamran Nazeer is one of my favorite books. Nazeer works as a policy advisor to the British government. When I found a reason to interview him, I took it.

ROBERTS At a conference of experimental psychologists, I heard about some test results that found that autistic kids did better than non-autistic kids. The researchers were expecting the opposite. They expected the autistic kids to have deficits in processing of faces, how well they can perceive faces. But they found the opposite. That’s what prompts this. I spoke to the researcher afterwards, and he said it wasn’t the only example. Another researcher has several findings along these lines and parents are fond of the idea that autistic kids have a different set of skills.

NAZEER I’m not convinced that we need to think of these things as polar opposites. I think what’s going on is that autistic kids have equal or even higher attention to particular details, or particular kinds of details. There can be two issues; one issue is sometimes that their sense of hierarchy about sense data is different from what we regard as normal. So it may be that autistic kids will regard particular sense data as being more interesting to them than sense data that might be more socially useful. So they might well pick up just as much, or more, information about people’s faces, but it’s just not the thing that they focus on. They might focus on something completely different instead. So then, it’s a question of how do you change the kid’s focus so that the data that all the time they’ve been taking in is the data that they actually use to form judgments about the world. So I think that that’s one thing that happens. I think the other thing that often happens is that, because of language difficulties, even though autistic kids might be picking up equal or higher levels of sense data, they’re just not able to articulate to other people, and hence probably not even that well to themselves, what it is that they’re perceiving.

ROBERTS You’re saying that autistic kids favor some kinds of sense data over other kinds of sense data?

NAZEER Right. To give you an example that I use in the book, which is about Elizabeth, who you might remember is the only girl that I write about. There is this scene in which her parents took her along to a bus stop. It’s not that she wouldn’t notice that there was a bus coming, and it’s not that she wouldn’t notice what the number on the bus was; it’s that she would also notice who in the queue for the bus had their nails cut, or what color people’s sneakers were, or if there was a missing apostrophe in the advertisement on the side of the bus. So, you know, it’s not that she was missing out on the crucial piece of sense data, which is “where is this bus going,” but she was not realizing that this was the most important piece of sense data for her at that time, to be paying attention to. So in that sense, she had a different hierarchy.

ROBERTS So you’re saying that for other people, where the bus is going would be higher on their hierarchy?

NAZEER Exactly. That’s because the non-autistic have a better social sense of what the relevant piece of sense data is at any particular time, whereas an autistic person might have a different hierarchy, or might have no hierarchy at all of sense data. That’s what often happens with autistic people when they feel overwhelmed by their surroundings. It’s because they’re not forming a hierarchy of sense data, it’s because they’re taking on all the sense data, it’s random, and as you can imagine, we’re always overwhelmed by sense data. But the reason why we don’t feel overwhelmed is because we have a hierarchy for sorting them out. So, when we’re sitting and reading the newspaper, we realize that it’s the words on the page that are at the top of the hierarchy. When we’re standing at a bus stop, we realize that it’s whether or not the bus is coming in, what the destination at the front of the bus is that’s at the top of the hierarchy. I think that what often happens with autistic people is that they don’t hierarchize. Either they don’t hierarchize in the same way, or they don’t hierarchize at all?

ROBERTS What does it mean, to not hierarchize at all?

NAZEER It means that you just feel overwhelmed by what you see around you, and so you don’t know, what if it is useful to you? And so you don’t know, what if it is useful to you, so you experience it all as being sort of alien and unsettling. That, I think, is why a lot of autistic people display what I and many other people have called desire for local coherence. So because they’re not forming a hierarchy of sense data, which ultimately is the only way in which we can stop ourselves from feeling overwhelmed in the world, what they do instead, instead of forming the hierarchy, they ache for some simple way of bringing order to the chaos around them. So rather than sorting out the sense data, they just pick one thing to focus on, so they pick a pen, or the edge of the table, or they start rocking, or they walk on the soles of their feet. So they take one random thing and put it on the top of the hierarchy, so that everything else that’s under it doesn’t overwhelm them any more.

How Things Begin (I Got UGGs!)

Mohamed Ibrahim, the New York schoolteacher who does Behind The Approval Matrix (which I have blogged about) also has a blog called I Got UGGs!. I asked him how the Ugg blog began. Here’s what he said:

I have a fetish about Uggs. Whenever I see a girl wearing Uggs, it’s the sexiest thing in the world to me. It drives me crazy. You know how they say “do what you love and the money will come later”? I read an article in Time about bloggers and blogging. One of the blogs they profiled was by two ladies who post pictures of kittens and cats and write little blurbs about them. This gave me an idea: I’ll do the same thing about girls in Ugg boots. They got $5-6000/month from ads and all they do is post pics and write blurbs about them. I’ll take pics of girls wearing Uggs. Not only will I enjoy it but maybe I can also make some money. I went to Best Buy, got the cheapest digital camera, and hit the streets. The first place I went was Times Square. Initially I would approach people and ask them if I could take their pic for the blog. I discovered later it’s better to just take the pic and put it up. That’s what I do now. Now I get people sending me pics — they take a picture of their friends or they send me pics of celebrities. We’re getting over 500 page views/day. It’s only been about 4 months.

The Gawker link Mohamed got by telling them some crazy guy was taking Ugg pics and blogging about it.

Buried Treasure (part 2)

Before the invention of statistical tests, such as the t test, science moved forward. People gathered data, computed averages, drew reasonable conclusions. As far as I can tell, modern ways of analyzing data improved the linkage between data and conclusion because they reduced a big source of noise: How the data were analyzed. Procedures became standardized. Hypothesis testing improved. Hypothesis formation, however, did not improve. Knowing how to do a t test and the philosophy behind it will not help you come up with new ideas. Yet data can be used to generate new ideas, not just test the ones you already have.

Our understanding of outliers is in a kind of pre-t-test era. People use them in an unstructured way. As Howard Wainer’s analysis of his blood sugar data indicates, better use of them will improve hypothesis formation. A kind of standardized treatment should help generate ideas, just as the t test and related ideas helped test ideas. Here are some questions I think can be answered:

1. Cause. What causes outliers? It’s a step forward to realize that outliers are often caused by other outliers. Howard has found that unusually high blood sugar readings are caused by eating unusual (for him) foods.

2. Inference. I’m fond of saying lightning doesn’t strike twice in one place for different reasons. The longer version is if two outliers could have the same explanation, they probably do. I think this principle can be improved.

3. Methodology. To test ideas, you want variation to be low. To generate ideas, you want outlier rate to be high. Howard could make progress in understanding what controls his blood sugar by deliberately testing foods that might produce outliers. In genetics, x-rays and chemical mutagens have been used to increase mutation rates; mutations are outliers. (Discovery of a white-eyed mutant fruit fly led to a wealth of new genetic ideas.) In physics, particle accelerators increase the outlier rate in order to discover new subatomic particles. There are no comparable procedures for psychology. Self-experimentation increased my rate of new ideas because it increased my outlier detection rate. It increased that rate for three reasons: 1. I kept numerical records. 2. I analyzed my data using the same methods as Howard. 3. I did experiments. Travel is like experimentation; there too it helps to keep numerical records and analyze them. The question: What are the basic principles for increasing outlier rate?
Part 1.

More from Holland

My friend in Holland wrote again:

Last year, the Dutch Supreme Court ruled that it was OK to have sex with animals, as long as the animals enjoyed it.

She attached a newspaper article in the Hague/Amsterdam Times dated 20 March 2008 that began:

Under a new law being debated by the government, sex with animals will be allowed as long as people don’t enjoy it.

It ended:

The Animal Party was mainly disappointed about the fact that the new bill does not refer to the animals’ dignity.

Buried Treasure (part 1)

Not long ago, Howard Wainer, the statistician I mentioned recently, learned that his blood sugar was too high. His doctor told him to lose weight or risk losing his sight. He quickly lost about 50 pounds, which put him below 200 pounds. He also started making frequent measurements of his blood sugar, on the order of 6 times per day, with the goal of keeping it low.

It was obvious to him that the conventional (meter-supplied) analysis of these measurements could be improved. The conventional analysis emphasized means. You could get the mean of your last n (20?) readings, for example. That told you how well you were doing, but didn’t help you do better.

Howard, who had written a book about graphical discovery, made a graph: blood sugar versus time. It showed that his measurements could be divided into three parts:

measurement = average + usual variation + outlier (= unusual variation)

Of greatest interest to Howard were the outliers. Most were high. They always happened shortly after he ate unusual food. Before a reading of 170, for example, he had eaten a pretzel. He had not realized a pretzel could do this. He stopped eating pretzels.

When Howard told me this, it was like a door had opened a tiny crack. Recently a deep-sea treasure-hunting company found a shipwreck off the coast of Spain. They named it Black Swan, apparently a reference to Nassim Taleb’s book. Shipwrecks are black swans on the ocean floor; black-swan weather had sunk the ship. For Howard, outliers were another kind of buried treasure: the key to saving his sight.

It isn’t just Howard. Outliers are buried treasure in all science. They are a source of new ideas, especially the new ideas that lead to whole new theories. The Shangri-La Diet derived from an outlier: Unusually low hunger in Paris. My self-experimentation about faces and mood started with an outlier: One morning I felt remarkably good. My discovery that standing improved my sleep started with a series of days when I slept unusually well.

Modern statistics began a hundred years ago with the t test and the analysis of variance and p values — very useful tools. Almost all scientists use them or their descendants. Almost all statistics professors devote themselves to improvements along these lines. However, conventional statistical methods, the t test and so on, deal only with usual variance. (Exploratory data analysis is still unconventional.) As Taleb has emphasized, outliers remain not studied, not understood, and, especially, not exploited.