Search Episodes
Listen, Share, & Support
Listen to the latest episode
Subscribe via iTunes
Subscribe via RSS
Become a fan
Follow on Twitter

Support Us:

Please consider making a donation to help make this podcast possible. Any contribution, great or small, helps tremendously!

Subscribe to E-Mail Updates

Related Readings
  • Answers for Aristotle: How Science and Philosophy Can Lead Us to A More Meaningful Life
    Answers for Aristotle: How Science and Philosophy Can Lead Us to A More Meaningful Life
    by Massimo Pigliucci
  • Nonsense on Stilts: How to Tell Science from Bunk
    Nonsense on Stilts: How to Tell Science from Bunk
    by Massimo Pigliucci
  • Denying Evolution: Creationism, Scientism, and the Nature of Science
    Denying Evolution: Creationism, Scientism, and the Nature of Science
    by Massimo Pigliucci

RS125 - The Quantified Self

Release date: January 11, 2015

"Quantimetric Self-Sensing" apparatusPeople have been keeping track of their moods, sleeping, dietary habits and more for hundreds of years -- Benjamin Franklin famously recorded instances of his virtues and vices. But only in the last decade has the rise of smartphones and fast computing created the new "Quantified Self" movement in which some people are trying to mine their own data for insights about how to be happier and more effective. In this episode, Massimo and Julia discuss self tracking -- what you can learn from it, and what its pitfalls might be.

Julia's pick: "Rethinking Positive Thinking: Inside the New Science of Motivation"

Massimo's pick: "Free Will: The Basics"

References (2)

References allow you to track sources for this article, as well as articles that were written in response to this article.

Reader Comments (8)

"Father of wearable computing" Steve Mann compared wearable devices to a personal black box, and recounted being the victim of a hit-and-run that he caught on his wearable camera.
He cited a San Jose professor who continuously webcasts his location after a false arrest by the FBI.
He coined the term "sousveillance" which means to watch under, in contrast to surveillance, which means to watch over.
Russian drivers install dashcams in their cars to prevent insurance fraud, and they've recorded some incredible stuff like the Chelyabinsk meteor.
A New York woman said she started making purchases with credit cards to leave a paper trail after she was framed for armed robberies by her ex-boyfriend. Cell phone records gave her an alibi and showed that her ex-boyfriend phoned the false witnesses. So records were good for the good girl and bad for the bad guy.

January 12, 2015 | Unregistered CommenterMax

Some of these activity trackers have to make a lot of assumptions and inferences from indirect measurements. The wrist has got to be the worst place to put a pedometer, since the pedometer will confuse hand waving with footsteps. Then, it might use the number of footsteps to estimate the number of calories burned, but that depends on the walking speed, the incline, the carried weight, etc. Do they ever validate that the estimations are correct, or is it all hand-waving?

January 12, 2015 | Unregistered CommenterMax

It is sad how afflicted Massimo is with 'good intentions bias'. He constantly assumes that selfish or amoral behaviour will lead to bad outcomes. He talks about 'companies' gathering data etc as if that is the final condemnation. What if their selfish amoral motivations lead them to serve their customers!? lol. Just so weird to hear someone running a rationalist podcast so deep into a cognitive bias.

January 13, 2015 | Unregistered CommenterMartin S

This could change the way medicine is done. Like, how does a doctor tell if you have a low-grade fever? He could compare your temperature with that of similar people, and it could be within the normal range yet be abnormal for you. A more useful comparison is with your own baseline temperature at that time of day. So it's a sample of one person, but lots of measurements at different times, kind of how climate scientists take the Earth's temperature to know when it has a fever.

Doctors already tell patients to keep food diaries and sleep diaries, but who has the time and expertise to sit down and analyze them? You can calculate simple stuff, like how many calories you eat and how many hours of sleep you get, but discovering causality is harder, like which food gives you gas. Objective data can help you overcome confirmation bias, like you might realize that taking vitamin C does not reduce the duration of your cold, but if you don't know what you're doing, you can find all sorts of spurious correlations. It would be nice if the analysis could be done automatically by the software, but if it were that easy we wouldn't need as many statisticians.

January 14, 2015 | Unregistered CommenterMax


Could you please clarify something you said in this podcast. It appears as if you just endorsed the "it works for me" argument. At minute 28 of the podcast you seemed to undermine skeptical arguments against the "It works for me!" reply.

To paraphrase you: "Even though there is a lot wrong with collecting your own personal data, it's ok when it's in the area of personal behavior because certain things that make you feel or perform better may never be significant in a large scale randomized controlled test." I hope I didn't misrepresent what you said but that seemed to be the gist of it. You mentioned specifically the areas of motivation, mood, energy and focus.

So, according to this argument, if ginkgo biloba "work's for me" to improve my focus, then I can ignore those large scale randomized and controlled tests that demonstrate it is ineffective. If healing crystals give me energy then it doesn't matter that they are implausible. If using acupuncture to align my Qi improves my mood then nothing should stop me from doing it every week.

Please explain.


January 26, 2015 | Unregistered CommenterDavid Y

David, I haven't seen Julia comment here, but here's my take on it.

Clinical trials can tell you what's worth trying, but patients still need to experiment to find out what works for them. People are unique, and any physician will tell you that medicine is not an exact science. If clinical trials find that something is no better than placebo, then you can assume it's not effective, BUT if the clinical trials find that it's effective for 50% of patients and causes nasty side effects for 10% of patients, then you have to figure out which patient you are and what your optimal dose is. For example, some drugs like sleeping pills are first prescribed in a low dose, and then the dose is gradually increased as necessary. And if one drug doesn't work for the patient, then it might be replaced with another drug.

January 27, 2015 | Unregistered CommenterMax

On the subject of mood tracking and how it may affect the mood, I've been recording my mood for about a year as part of my [hoped for] recovery from clinical depression.

I don't use an app. I use a scale of my own creation, centred at zero with 0.5 increments to 2.0 and -2.0. I gave it a lot of thought and I'm sure it's a very personal thing, but I found necessary to perceive the scale as non-linear, so that 1.0 is more than twice as good as 0.5 (more like 4 times better). From the start I strove to minimise subjectivity of the recorded figures, but perhaps my depression is harder to allow for than I expected. My readings tend to cluster around the neutral 0.0 mark and rarely go above 0.5. They mostly are between 0.0 and -1.0.

I don't find the recording of my mood affects my mood significantly, but does occasionally send me into a chain of negative thoughts.

A person's general peronal mental state must be significant in this. I'd guess that optimistic people tend to boost their mood by conscious acknowledgement of the mood.

February 21, 2015 | Unregistered CommenterThomas C.

I'm pretty late to the party on this one, but -- yes, basically what Max said. Large scale RCT can tell us whether an intervention is helpful for *people in general*, or for certain demographics of people (based on age, presence of certain other medical conditions, etc).

But you should expect there are many interventions that will work for some people and not others, and that the reasons for the different results are not going to be as obvious as age or gender or something. So if an intervention is helpful to some subsets of the population, and and harmful to others, an RCT won't be able to detect an effect -- it'll just look like random noise unless you know the right things to control for. Which we usually don't.

Psychology and nutrition, in particular, have this property: very complex, and a lot of variation between individuals. So interventions in these areas seem most likely to have interventions that work for some people but not others, which RCTs can't detect because we don't know the reasons for the differing outcomes.

March 15, 2015 | Unregistered CommenterJulia Galef

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
All HTML will be escaped. Hyperlinks will be created for URLs automatically.