Group Rationality Diary, November 1-15
post by therufs · 2014-11-03T03:58:37.961Z · LW · GW · Legacy · 18 commentsContents
This is the public group rationality diary for November 1-15. None 18 comments
This is the public group rationality diary for November 1-15.
It's a place to record and chat about it if you have done, or are actively doing, things like:
- Established a useful new habit
- Obtained new evidence that made you change your mind about some belief
- Decided to behave in a different way in some set of situations
- Optimized some part of a common routine or cached behavior
- Consciously changed your emotions or affect with respect to something
- Consciously pursued new valuable information about something that could make a big difference in your life
- Learned something new about your beliefs, behavior, or life that surprised you
- Tried doing any of the above and failed
Or anything else interesting which you want to share, so that other people can think about it, and perhaps be inspired to take action themselves. Try to include enough details so that everyone can use each other's experiences to learn about what tends to work out, and what doesn't tend to work out.
Thanks to cata for starting the Group Rationality Diary posts, and to commenters for participating.
Previous diary: October 16-31
Next diary: November 16-30
18 comments
Comments sorted by top scores.
comment by Wes_W · 2014-11-14T06:16:56.258Z · LW(p) · GW(p)
I am confused about weight loss. Specifically, it is past time that I update somewhat away from the (practical usefulness of?) the "calories in, calories out" model.
Some background:
For the last fourteen months, I've been logging everything I eat and tracking calorie intake. I lost about 15 pounds in the first 9 weeks at almost exactly the expected rate. After that, I did bulk-cut cycles (running 250cal surplus for a few weeks, then a 750cal deficit for a few weeks, with transition weeks at maintenance between phases), but have failed to make any further progress to speak of in body composition, whether in muscle mass gained or fat lost. I seem to have hit a wall at 170 pounds and about 20% bodyfat.
The standard response is that self-reported calorie intake is notoriously inaccurate, a problem to which I do not imagine myself immune. But I don't think that's enough to explain the results: when bulking, my weight goes up at the expected rate, and when at maintenance my weight does in fact hold constant. So I appear to be well-calibrated for those. But when running a deficit, I lose only a fraction of the expected amount, and I find it hard to believe that I mysteriously become that much less accurate while cutting, to a degree that almost entirely cancels out the planned deficit. Those are the times when I'm being most scrupulous!
Also, my physical performance during workouts unmistakably tanks during a cut, which is exactly the expected result of a large calorie deficit, but should not happen if I'm nearly at maintenance.
So at this point, I am much less sure that calorie balance is the sole determinant of weight loss. At best, it seems to be "technically true, but useless": a year of specifically focusing on calorie balance has fallen far short, both of theoretical predictions and of what I would consider a satisfactory rate of progress.
I wish I knew what alternative hypothesis to update towards.
Replies from: gjm, ChristianKl, Lumifer↑ comment by gjm · 2014-11-14T13:59:14.225Z · LW(p) · GW(p)
Uninformed guess: When you cut your consumption of calories below your usual baseline, your body adjusts some metabolic parameters so as to use fewer calories (e.g., lower body temperature, less fidgeting, less activity more generally). So there's probably really a range of consumption levels within which, if you consume that much and otherwise do what comes naturally, your weight will be stable. To lose weight, you'll need to cut calories further than you'd otherwise think and/or force yourself to be more active than you feel like being despite the lower intake.
Replies from: Lumifer↑ comment by ChristianKl · 2014-11-15T16:00:40.563Z · LW(p) · GW(p)
Most of the energy expenditure of the body goes towards various maintenance tasks with can be regulated up or down. Brown fat tissue can even simply produce heat without any real process.
If a tissue is under high pressure it kills some of it's cells via apoptosis to lower the pressure under which each cell happens to be. If the tissue is under low pressure it increases cell numbers to fill up the available space. Of course available energy has some effect on this process as well but the body can regulate hunger or some other process when it thinks that some tissue really needs a different number of cells because it's under a lot of pressure or there isn't enough pressure.
↑ comment by Lumifer · 2014-11-14T15:54:46.715Z · LW(p) · GW(p)
I am much less sure that calorie balance is the sole determinant of weight loss
Physics is pretty adamant it is.
a year of specifically focusing on calorie balance has fallen far short
You're measuring your input side, you don't have data for the output side. The calorie expenditure is not stable, besides it depending on the environment, your body will adjust your metabolic rate depending on the calorie availability.
In fact, if you know how many calories you consumed, you can calculate your average calorie spending rate from the changes in your weight. If your weight loss stalled (your weight stays the same), it means your calorie expenditure is equivalent to the calorie input.
Replies from: Eliezer_Yudkowsky, Wes_W↑ comment by Eliezer Yudkowsky (Eliezer_Yudkowsky) · 2014-11-15T05:00:37.730Z · LW(p) · GW(p)
Physics is pretty adamant it is.
Physics is even more adamant that mass-in-mass-out is the determinant of weight balance. Try drinking less water, which is very massy, and eating lighter foods. I hear fat has more calories per gram than protein or carbs, so if you take your calories by fat you can't help but lose weight! Right? If you were previously taking in most of your calories as protein and carbs, changing all your calories to fat can't help but nearly halve your mass-in. As long as you don't inhale more air to make up for it, and keep breathing out and sweating and excreting the same amount, physics says you literally can't help but lose mass! Literally every time someone keeps strictly to the MIMO diet, they lose mass - period. Every time someone gains mass, strict examination shows that they took in more mass than the MIMO diet says they should, given how much they were excreting, exhaling, and sweating. If you keep to the MIMO diet and still have problems with your weight, it's probably because you're not staying in the same gravity and have moved to a heavier planet. If you can't keep to the MIMO diet, you just need more willpower to avoid taking that extra inhalation.
Modulo caloric loss by e.g. ketones in urine, physics says CICO has to be 'true in some sense', but it doesn't say CICO has to be any more useful than MIMO.
Replies from: Lumifer, army1987↑ comment by Lumifer · 2014-11-16T20:15:34.528Z · LW(p) · GW(p)
physics says CICO has to be 'true in some sense', but it doesn't say CICO has to be any more useful than MIMO.
CICO is correct, to quote Yvain, in the best way -- it's technically correct.
As to being useful, as usual, it depends. I know you didn't find CICO useful, but lots of other people did. Your personal experience doesn't generalize well as you should know :-)
↑ comment by Wes_W · 2014-11-14T16:40:37.647Z · LW(p) · GW(p)
Physics is pretty adamant it is.
Yeah, that's why it confuses me. But after posting this and sleeping on it, I think I'm really experiencing one part epistemic confusion to two parts regular frustration.
Discussions of metabolism adjustments are almost never quantitative, for some frustrating reason. My impression has been that 750 cal/day is well beyond a realistic adjustment, especially in the absence of obvious side effects (lethargy, severe hunger, chills, sexual dysfunction, etc). But it occurs to me that I've been too dismissive: supposing e.g. a 250 cal/day depression in metabolism, and that self-reporting inaccuracy has me really overeating by 250 cal/day compared to my target intake, that would leave me at a 250 cal/day deficit, which... would just about match the observed rate of progress.
So okay, yes, those two things together would just about explain it. Then epistemically, I don't have much reason to dispute the model. Now there's just the instrumental problem of actually making progress: attempting a steeper deficit of 1000 cal/day has proven deeply unpleasant before, and rapidly produced the previously-mentioned side effects. Perhaps I need to plan more for hunger management, and/or get really obsessive about not eating things during a cut unless I can measure them exactly? At least this gives me some parameters to experiment with.
Replies from: Lumifer↑ comment by Lumifer · 2014-11-14T16:53:55.518Z · LW(p) · GW(p)
Discussions of metabolism adjustments are almost never quantitative, for some frustrating reason.
The reason is that directly measuring your metabolism is highly inconvenient (you basically spend time in a gas mask) and requires specialized and expensive equipment.
750 cal/day is well beyond a realistic adjustment
No, I don't think so.
For a frame-of-reference adjustment consider reports that the Olympic swimmer Michael Phelps consumed 8-10,000 calories per day while training. He didn't get fat :-)
attempting a steeper deficit of 1000 cal/day has proven deeply unpleasant before
You don't need a steep deficit, you just need some deficit which is visible as weight loss.
Given that you don't know the output side of the calorie balance, I am not a big fan of counting calories to start with. I would suggest setting a reasonable rate of weigh loss (say, 1-2 lbs/week -- if you can measure body fat % it would be even better) and eating at the level which sustains this weight loss regardless of how many calories it takes. If the weight loss stalls, eat a bit less. If it accelerates, eat a bit more. That, of course, does not exclude using whatever "tricks" you find to be helpful like load-up days or intermittent fasting or whatever.
Replies from: Wes_W↑ comment by Wes_W · 2014-11-14T17:08:56.099Z · LW(p) · GW(p)
For a frame-of-reference adjustment consider reports that the Olympic swimmer Michael Phelps consumed 8-10,000 calories per day while training. He didn't get fat :-)
Is that a result of his metabolism adjusting upward, as opposed to simply burning lots of energy on activity and maybe having an unusually high metabolism in general?
Given that you don't know the output side of the calorie balance, I am not a big fan of counting calories to start with. I would suggest setting a reasonable rate of weigh loss (say, 1-2 lbs/week -- if you can measure body fat % it would be even better) and eating at the level which sustains this weight loss regardless of how many calories it takes. If the weight loss stalls, eat a bit less. If it accelerates, eat a bit more.
I'm not sure how to actually apply this. Eating consistently at a certain level without somehow tracking calorie intake seems exceptionally tricky, unless I could standardize my meals to a degree that I don't think I can realistically do.
Replies from: Lumifer↑ comment by Lumifer · 2014-11-14T17:17:29.689Z · LW(p) · GW(p)
Is that a result of his metabolism adjusting upward, as opposed to simply burning lots of energy on activity and probably having an unusually high metabolism in general?
With Michael Phelps it's tricky. There are actually three parts here. Part of his energy balance goes just to do the physical movements of swimming, part goes to increased heat production as he spends hours per day in a cold-water pool, and part is just high resting metabolism.
Eating consistently at a certain level without somehow tracking calorie intake seems exceptionally tricky
Depends on how your food is structured. But if you're counting calories already, you can just continue to do this, just don't have a fixed level in mind. E.g. last week you were eating, say, 1,700 cal/day and lost zero pounds, so this week reduce to 1,500. Don't pay attention to absolute numbers, all you need to know is whether to shift up or down.
comment by NoahTheDuke · 2014-11-07T18:55:03.809Z · LW(p) · GW(p)
This week, my therapist and I talked about the nature and subtleties of my major forms of akrasia, and have begun the conversation that I hope will help me change or break them. After years of trying to deal with these problems by myself, it's oddly freeing to be beholden to someone else and also to have someone who is committed but not emotionally bound to me and my success.
My focus this week is to think of possible mantras that I could use to stay on task and begin (if not complete) tasks I know I need to do but I don't want to do. Lots of rereading of the anti-akrasia threads here.
Replies from: ahelcomment by jackk · 2014-11-03T21:23:07.090Z · LW(p) · GW(p)
Whenever I left home after making a cooked breakfast, I would worry that I'd left the gas on. I'd always have to go back and check, but of course the burner was off. I've fixed this by noticing that I have noticed that I've turned off the gas. Is there a common name for this pattern?
Replies from: polymathwannabe, shminux↑ comment by polymathwannabe · 2014-11-07T02:24:59.395Z · LW(p) · GW(p)
One technique I read somewhere is doing something unusual (like reciting a line from a song or rubbing both ears) while you do the thing you'll want to remember, so that you'll only need to ask yourself "Did I rub my ears?" instead of "Did I turn off the stove?"
↑ comment by Shmi (shminux) · 2014-11-05T00:05:06.735Z · LW(p) · GW(p)
I don't know if there is a name for this, but I've heard of people who take a time-stamped picture of the stove just before they leave, to help them cope with an OCD symptom like that. It is somewhat better than the old approach of carrying an iron with you to avoid worrying whether you turned it off :)
Replies from: jackk