Last Weekend’s Quantified Self Conference

Last Saturday and Sunday there was an international Quantified Self Conference at Stanford. I attended. In Gary Wolf’s introductory talk, he said there are 70 Quantified Self chapters (New York, London, etc.) and 10,000 members. I was especially impressed because I recently counted about 50 chapters. One new chapter is Quantified Self Beijing. It has its first meeting — in the form of a day-long conference — in nine hours and I haven’t quite finished my talk (“Brain Tracking: Why and How”). Please indulge me while I procrastinate by writing about the Stanford conference.

Here are some things that impressed me:

Office hours. A new type of participation this year was “office hour”, meaning you sit at a table for an hour. My office hour, during which two people showed up, was the most pleasant and informative hour of the whole conference for me. I thank Janet Chang for suggesting I do this.

Robin Barooah
used a measure of how much he meditated, which he collected via an app he made, to measure his depression. When he was depressed, he didn’t meditate. Depression is half low mood, half inaction. It is very rare that the inactive side of it is measured. It is so much easier to ask subjects to rate their mood, but this has obvious problems. Robin inadvertently found a way to measure level of activity over long periods of time. He also found that participation in an experiment that tested a PTSD drug caused long-lasting improvement, another idea about depression I’d never heard before. At dinner, Robin told me that his partner, when they’re at a restaurant, has sometimes said “God bless Seth Roberts” for allowing her to eat butter without guilt.

Steve Jonas, from QS Portland, told me that he spent a long time (many weeks) doing some sort of mental test. During one of those weeks, he consumed butter a la Dave Asprey, in coffee. Much later he analyzed the results, computing an average for every week, and noticed that during the week with butter his performance was distinctly better than performance on other weeks. I hope to learn more about this. Steve also gave a talk about learning stuff using spaced repetition. He noticed that learning new stuff increased his curiosity. After he used spaced repetition to learn stuff about Mali, for example, he became more interested in reading news stories about Mali. I think this is an important conclusion about education, the way rote learning and encouragement of curiosity are not opposites but go together, that I have never heard before.

Larry Smarr, a computer science professor at UC San Diego, gave a talk called “Frontiers of Self-Tracking” centered on his Crohn’s disease. I was struck by what was missing from his talk. He began self-tracking before the Crohn’s diagnosis and clearly the self-tracking helped establish the diagnosis. However, you don’t need to self-track to figure out you have Crohn’s disease, roughly everyone who has gotten this diagnosis did not self-track. I couldn’t figure out how much the self-tracking helped. Crohn’s is generally associated with frequent diarrhea, which is exactly the opposite of hard to notice. Larry said nothing about this. Later he talked about massive amounts of personalized genetic data that he was getting. I couldn’t see how this data could possibly help him. Isn’t self-tracking supposed to be helpful? If I had a serious disease, I would want it to be helpful. At the same time, judging from his talk, he seemed to be ignoring the many cases where people have figured out how to better live with their Crohn’s disease. I would have liked to ask Larry about these gaps at his office hour but I had an eye problem that caused me to miss it.

I asked Nick Winter, cofounder of Skritter, what he thought of the recent Ancestral Health Symposium at Harvard (August 2012), which we both attended. He didn’t like it much, he said, but it more than justified itself because Chris Kresser’s talk about iron led him to get his iron checked. It turned out be off-the-charts high. Partly because oysters, partly because of red meat. I think he said he has since donated blood and it came down. I hadn’t previously heard of this danger of eating red meat. Again I discussed with Nick why he found that butter had a bad effect on his cognitive performance, the opposite of what I found. One possibility is that the butter slowed digestion of his lunch, thus reducing glucose in his blood at the time of the cognitive tests. But this does not explain why a certain drug eliminated the effect of butter.

In his talk, Paul Abramson, a quant-friendly San Francisco doctor, said that mainstream medicine is “riddled with undisclosed conflicts of interest”. I hope to learn more about this.

Jon Cousins contributed a neat booklet about what he had learned and not learned from starting Moodscope. What he hadn’t learned was how to make a sustainable business out of it. I suggested to him that he might be able find professors who would apply for grants with him that would use Moodscope as a research tool. The grants would pay Jon a salary and might include money for software development. Mood disorders are a huge health problem — depression is sometimes considered the most costly health problem of all, worldwide — and Moodscope is a new way to do research about them. Paying Jon a salary for a few years would cost much less than assembling a similar-sized sample (Moodscope has thousands of users) from scratch. I wonder how professors who do research on mood disorders will see it.

2 Responses to “Last Weekend’s Quantified Self Conference”

  1. Jim Purdy Says:

    I hope you will publish more about Paul Abramson and undisclosed medical conflicts of interest.

    Surely there are no conflicts of interest in medicine.

    Sarcasm off now.

  2. Josh Says:

    It was great chatting with you during office hours seth. I hope we can continue the conversation soon. Also great summary of qs.