Group Rationality Diary, August 1-15

by therufs1 min read1st Aug 201361 comments


Personal Blog

This is the public group instrumental rationality diary for August 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!

Immediate past diary:  July 16-31

Next diary:  August 16-31

Rationality Diaries archive

61 comments, sorted by Highlighting new comments since Today at 8:43 PM
New Comment

It's a small thing, but a friend and I were having coffee in a park, and he offered me the rest of his muffin, since he was full. I didn't really want it, and we looked guiltily at it for a bit before spiking it into the garbage can, yelling "Take that, Sunk Cost Fallacy!" and doing a brief high five/victory dance.

It's nice to have a social network that gives you positive feedback for spotting and short-circuiting biases.

What, no hungry pigeons around?

There are surely hungry birds at the dump as well.

What, no hungry people around?

That, too.

Universalizability problem - I don't want to be in a park littered with the odds and ends of food people didn't eat.

I've always had problems dealing with negative emotions, in that once I am in the negative emotion it changes my decision making such that it is hard to break out of it. For example, I get angry, and even though I know I will feel stupid about it later on, it feels so good to be angry that I stay angry. And then I feel stupid about it.

This last week, I got really annoyed at something trivial enough I no longer even remember what. For the first time ever, I just asked myself what the point of getting annoyed was, why would I want to inconvenience myself with annoyance, and what I could do to ensure that this annoyance wouldn't occur again. Almost immediately, my annoyance went away and I felt good again.

At work, one of our customers is known to be "unique"/"challenging". The customers normally talk to a person (let's call this the middle person) who used to work in the customers' field and translates their requests for software developers. The customer had gone through >3 e-mail rounds with the middle person and I as we tried to build an item to her liking.

After the last e-mail complaint, I printed out all of the e-mails related to the topic, all of the example cases she pointed at, and tried to figure out what state of mind would make her so frustrated and have so much trouble getting us to do what she wants as opposed to simply taking each e-mail at face value. I made one more tweak to the product, e-mailed the middle person, and held my breath.

This happened about 2 weeks ago.

My company hasn't received any more e-mails about the product so it seems to be a great success.

I post this anecdote here because the idea of approaching the problem like that was almost certainly taken from HPMoR. Quote is below:

Harry kept his expression blank, and realized one second too late that it might as well have been a signed confession. Professor Quirrell didn't care what your expression looked like, he cared which states of mind made it likely.

Caution: This probably only works if the person you're modeling is sane.

Updating towards many animals I eat being more similar to humans than I had previously thought, based mostly on this link from Why Eat Less Meat?. Still unsure what to do about it. Feeling guilty seems unproductive and so does rabid vegetarian advocacy.

Also, I conducted an informal poll on Facebook among my friends about how much they would have to be compensated for going vegetarian for a year, and I got answers generally hovering around $3,000 (I included instructions not to look at other people's answers first, although I don't know how well they were followed), although one respondent seriously answered $20,000. At the very least this reflects a general impression that vegetarianism is very inconvenient / unpleasant which could be the target of effective animal altruists. Although...

Another interesting question is to ask current vegetarians how much they would pay to stay vegetarian.

I can't answer this question without knowing what the scenario where I don't cough up the requisite amount looks like. (I have this problem with a lot of "how much would you pay" questions.) Even I assume for the sake of argument that I am in a position where I have to part with money to continue not eating meat, that doesn't tell me who has me in this situation or what they're going to do about it. Force-feeding, legal consequences, health consequences, social consequences, if I don't eat some minimum amount of meat? Meat will be teleported into my stomach on a routine basis without my intervention should I fail to make quota? How much is that minimum?

This is a difficult question. By analogy, should rich cannibals or human child abusers be legally permitted to indulge their pleasures if they offset the harm they cause with sufficiently large charitable donations to orphanages or children's charities elsewhere? On (indirect) utilitarian grounds if nothing else, we would all(?) favour an absolute legal prohibition on cannibalism and human child abuse. This analogy breaks down if the neuroscientfic evidence suggesting that pigs, for example, are at least as sentient as prelinguistic human toddlers turns out to be mistaken. I'm deeply pessimistic this is the case.

I wasn't speaking at all about "moral offsets". I was attempting to counter Qiaochu_Yuan's point that a high value put on eating meat by meat eaters indicates that being vegetarian is difficult.

I want to answer about $3000, but I am pretty sure that's almost entirely because of priming. I think the honest answer is that I'm not sure I'm capable of eating meat anymore. Emotionally, I find it disgusting and repulsive. I almost certainly don't have the enzymes to digest meat anymore, as I've been a vegetarian for over two years. The resulting combination is... gastrically unpleasant.

Wow at first I couldn't understand at all how someone would claim that much money just for not being vegetarian. I think there's two important biases here though.

  1. "Willingness to pay vs. willingness to accept". This should decrease your actual figures a lot. For example consumers will ask much more money to accept a bad pixel on their computer screen than they would pay to have it repaired. "Willingness to accept" is not really interesting to economists. Basically ask how much they would pay to eat meat for a year if they had a vegetarian diet by default to get around this.

  2. Maybe not a bias, but as others said here, vastly underestimating the inconvenience of being vegetarian. It's really not that much of a hassle. Add to that the lower cost and health benefits and the environmental benefits if you care about those and being vegetarian seems like a win-win.

One more thing I'd like to add is that only after turning vegetarian I was first able to understand how severely I had been biased toward neglecting suffering in other species. Speciesist bias you could say. While still eating meat I was simply not able to care about animal suffering enough.

Another effect related to 2 is that many people probably think they can't find vegetarian dishes that taste as good as their preferred meat dishes. (I mean, I also think this. I don't currently know of any good substitutes for my preferred meat dishes.)

One more thing I'd like to add is that only after turning vegetarian I was first able to understand how severely I had been biased toward neglecting suffering in other species.

As long as we're talking in terms of biases, this could itself be regarded as a bias, namely a consistency effect.

Still unsure what to do about it. Feeling guilty seems unproductive and so does rabid vegetarian advocacy.

You could try to eat fewer animal products (which doesn't require rabid advocacy), especially chicken, eggs, and fish; or you could donate to effective animal charities. I agree that "rabid vegetarian advocacy" is probably ineffective.

I agree that aiming to eat less meat rather than none is a worthy goal for many reasons, I just want to add (anecdotally) from the perspective of someone who was a strict veg for 10 years and recently started eating some meat again, that you have to be very strict and planned if you want it to be effective. During the decade I didn't eat meat I didn't think about it or miss it because in my mind it wasn't an option. Now that I have the goal of just eating small amounts of it, it's much more mentally taxing to actually make the decision at each meal as to whether I should or shouldn't eat meat this particular meal.

So my, again anecdotal, advice if someone is to try and reduce meat consumption but not eliminate it: be extremely strict about how you're going to do it, as in say you will only eat meat on weekends for example, or one meal a day, so that you don't have to consider at each meal whether it's appropriate or not, which I find draining on willpower.

Not sure that's a good way of asking. The pain of being veg seems to be inversely correlated with knowing reasons for going veg, and there's a lot of loss aversion. I got anecdotal evidence from quite a few people telling me that going and staying veg is actually much easier than they anticipated. (It's important to take the time to learn about health effects, look at meat alternatives and find veggie restaurants). Reversal test: Imagine you're vegetarian and earn $3000 more than you currently do. Would you pay $3000 a year ($8 a day) to eat meat again?

It's not only more ethical but also healthier to be veg.

It makes me happy to see you and others taking action due to Peter's post.

EDIT: Concerning the "although", I recommend to also read Brian's comment.

I got anecdotal evidence from quite a few people telling me that going and staying veg is actually much easier than they anticipated.

I would expect this, but my point is that if non-vegetarians have inaccurate impressions about how hard it is to go vegetarian then that could be a useful misconception to clear up.

Agreed. Do you have an idea how to go about this?

You should read the results of the first study you posted more carefully:

Cohort studies of vegetarians have shown a moderate reduction in mortality from IHD but little difference in other major causes of death or all-cause mortality in comparison with health-conscious non-vegetarians from the same population.

The other links don't contradict this study, and only look at deaths from specific causes, and not general mortality.

You should read the results of the first study you posted more carefully:

Good point, thanks. My statement is not exactly wrong, but I should have written "healthier than average diets".

The other links don't contradict this study and only look at deaths from specific causes, and not general mortality.

That's quite wrong, examples:

Key 1999:

Total mortality and longevity also differed according to vegetarian status in California Seventh-day Adventists. After adjusting for age and sex, Seventh-day Adventist vegetarians had a relative risk for total mortality of 0.80 (95% CI: 0.74, 0.87) compared with those who ate any meat products. Using a multivariate, multiple-decrement-lifetable approach (19), we showed that vegetarian Seventh-day Adventist women live 2.52 y longer than their nonvegetarian (meat ≥ 1 time/wk) counterparts (P < 0.001), and a similar comparison in men showed a 3.21-y difference in longevity (P < 0.001).

McEvoy 2011 (review):

Overall, vegetarians tend to be slimmer, appear to be in better health, with reduced risk of chronic diseases and greater longevity when compared with omnivores

In that analysis, no significant differences were observed for stroke mortality or overall mortality between vegetarians and non-vegetarians(12).

(...) but no significant differences were observed for overall mortality rates between vegetarians and omnivores in these cohorts. One possible explanation may be that overall mortality was low in the cohort populations compared with the general Western population.

I deliberately only quoted very conservative and reliable sources, and although the effects are not really large, they are statistically significant and positive.

Sorry, I meant in regard to the "compared to health-conscious non vegetarians" part.

The fact that vegetarians tend to be more concerned about eating well is a huge factor that almost every "vegetarian diets are healthier!" study I've seen ignores. The first one you posted is the only one that tries to control for that, and they ended up with seeing no significant difference.

Updating towards many animals

When I read this as "updating towards many-worlds", I have probably spent too much time on LW.

Vegetarian diets are cheaper, you'd get about $1000 just from that, no other considerations involved.

$1000 seems pretty high/optimistic to me. Sometimes vegetarian meals are more expensive due to lack of options. Also, preparing veggie food usually takes more time. As a result (after having improved my cooking / food preparing speed) I still get a small monetary benefit though, maybe $500 a year.

Can't you get almost all of that cost reduction without becoming fully vegetarian? For that matter, could one get some portion of almost all of the befits of being completely vegetarian by becoming a part-time vegetarian?

Yes. Being a part-time vegetarian is a good choice.

Yes, except the benefit of not hurting sentient beings, I'd say. And probably except the benefit of not being biased towards hurting animals.

Is the benefit of not hurting sentient beings at all significantly different from the benefit of not hurting sentient beings as much?

Treat me as though I don't understand the moral value in not hurting animals...

If the part-time vegetarian still eats significant amounts of meat and eggs, then yes, there will also be a significant ethical difference.

If you're just interested in cutting down the cost of your diet, you also might switch to different products such as cage eggs. The cheapest production often is also the most cruel. But I assume that's not what you meant (and it's not what I meant either).

Yeah, I meant e.g. adopting a vegetarian diet for three days a week and an unchanged diet for the other four; it would seem to offer 3/7 of the benefits of being fully vegetarian.

Okay, now I see what you meant. I assumed that since you'd optimize for financial benefit you want to start with a reduction of the most expensive meat options and thus get more than 3/7 of the financial benefit when adopting it three days a week.

If I wanted to optimize for financial benefit, I'd be completely agnostic about eating meat, and I suspect I might end up eating mostly oils for calories but buy bulk grains to grow vitamin-rich yeasts.

Is this still true after accounting for nutritional and health changes necessary to get a vegetarian diet with the same quality as an omnivorous diet? I could cut out all meat, and just replace it with more of the vegetarian things I typically eat, but I would get less protein, and healthy fats, creatine, etc... which I would have to compensate for.

Protein deficiency is very rare even among long-term vegans and it's pretty hard to miss out on essential amino acids. As for "healthy fats, creatine, etc...", those can be easily supplemented, which is particularly important for vegans. Also note that meat eaters usually don't get enough healthy fats either.

Vegetarians have higher life expectancies (1-9 years), as also stated here.

Most vegetarians probably get the majority of their protein from nuts, beans, and grains, which tend to be a lot cheaper than meat. The same goes for fats, etc.

Having been spending $900/year to eat an omnivorous diet, you'd have to be eating a huge amount of meat to save $1K/year by going veg. Something like eating 16oz steak every day.

Having been spending $900/year to eat an omnivorous diet, you'd have to be eating a huge amount of meat to save $1K/year by going veg.

I spend roughly four times that, eating an omnivorous diet which, as it happens, does not include any large slabs of meat, and I'm curious to know where the difference lies. If you were specifically working to economise on food outlay, then it would not be a surprise to you that people not trying to economise have room to cut down by $1000 a year. But if you are not, how does your cost come in at less than $3 a day?

I'm in the UK, where food is more expensive than the US, but not four times more expensive, and while I generally shop at the better supermarkets rather than the cheap ones, I have no inclination towards "luxury" goods, and rarely eat out.

If you were specifically working to economise on food outlay

I was.

The surprising thing to me was not that people would spend more than $1K on food per year, but that the projected savings by eliminating meat would be $1K.

I was generally under the impression that eating on less than $3/day was not particularly typical beyond poverty. I expected roughly $100/month for food, $200 if you indulge in anything like eating out/etc, if you were trying to save money.

A quick search pops up $~150/week as the average American food bill (four times what I came up with), and I found this page which gets more detailed.

What I take away from this is that most people probably could benefit financially from reducing meat intake, and most people seem to be spending more on food than is strictly necessary in general, at least in the US.

ETA: Unclear if we're talking about specific people or general groups. The $3000 for avoiding meat in Qiaochu_Yuan's informal survey tells me people value eating habits that come with lots of ridiculous drawbacks a bit more than I thought (and I previously thought people's values regarding eating were kinda screwy).

under the impression that eating on less than $3/day was not particularly typical beyond poverty

You're right that no one else I surveyed was spending that little, so it is unusual. I was just surprised to see expected savings larger than what I'm used to spending.

Your link says that "typical family of four in the United States making the median household income would have to double its food expenditures in order to eat what USDA nutritionists consider a healthy diet" but bases that claim on the cost estimates from the Thrifty Food plan. The problem is that these estimates come from a ridiculously limited optimization process where they group food into about 60 categories ("whole fruits", "orange vegetables", "whole grain cereals"), then assign costs and nutrition information assuming that people are eating from the category in the proportion people do on average. So if the "whole fruits" people tend to eat are 40% apples, 35% oranges, and 25% bananas, then the cost will be a weighted average of those three. Once they've assigned costs and nutrition, they run some optimization to figure out how cheaply one can get the needed nutrients. The problem with this is that they can't say "eat bananas to get more potassium" because the only knob their optimization thing can tweak is the "whole fruits" one. This means that if a category is not homogeneous in terms of either nutrition or cost, they'll not be able to optimize well.

I think most people spend much more than that on food. The internet seems to think it's about $6000 a year.

I can definitely believe going veg would save $1K off of $6K, but I have trouble imagining how you could spend that much without eating out for most of your meals.

[-][anonymous]7y 1

I guess it involves buying more stuff than you'll be able to eat before it goes bad, and hence having to throw lots of stuff away.

I realized that I have a myriad of interests, and that I put a lot of effort into pursuing and studying them, but ostensibly make very little progress in any of them.

I realized this is mostly due to how I will try to do too much at once, and learn too many things at once, changing which interest I want to pursue daily or hourly etc.

To combat this two months ago I began forcing myself to choose one interest or goal at a time, and only allow myself to use my free time on that particular goal or interest for a minimum of two weeks. So far the results have been excellent, I've made actual significant progress on the goals and interests I've chosen so far. Not only has this been objectively beneficial in the sense that I'm making more progress than I normally do at something, but I find it to be much more satisfying as well, since I can really dig into something for a while before moving on.

After a year and a half of living in China I got a Mandarin tutor for an hour six days a week. I'm making much more progress, faster, than I did in my previous half assed way of learning.

Congrats man. I also highly suggest you use Anki if you aren't already. It reminds me of the words & phrases I learned months ago & keeps them fresh in my memory. Not that it is my sole motivation, but you can really impress native speakers when you pull out some obscure saying someone mentioned in passing a long while back.

I am using Anki though sub-optimally. Until recently I was just using one of the shared decks, New Practical Chinese Reader. Now I'm catching up with myself inputting the phrases and vocabulary from the book I'm using, "Chinese for Foreigners". Making cards is a massive PITA but worth it for things you actually care about learning. How long did it take you to do a textbook's worth of cards when in university?

I've actually built & abandoned several decks. When I was studying in university I was focusing more on the specific usage of certain words in context, so my cards were ridiculously complex. It was good for studying for the exams, but exhausting to keep up with. Now my cards are pure vocabulary. I don't go through lists & add all the words. I wait for the word to come up in daily life, then add it. In the past I studied a lot of words I never used, now I mostly study words I'm encountering regularly. I do think this depends on where you are though. I can usually follow a conversation but just one or two words will be new to me. If you're still in the early beginning phase, you need to collect all of those super common words first.

These two weeks I have succeeded at avoiding web procrastination from Monday to Friday. (This is what I have decided in advance, because I don't want to give up the web completely -- especially LW -- I just don't want it to harm my work, or take away all the evenings during the weekdays. Then I use the weekend to read all the new stuff.) As a proof, see my disappearence from the "top contributors, 30 days" list. :D

The technique is susprisingly easy (for me; that is not an evidence it would work for anyone else). Each week I write on a piece of paper "I will not browse the web during this week (Monday to Friday)", and I keep the paper always with me in my pocket. However, there is an exception: I can browse the web as much as I want, assuming I will eat that paper first. Every morning, and every time I feel the temptation, I touch the paper with my tongue to remind myself (in near mode) about the conditions. At the end of the week I happily tear the paper to pieces and throw it away, as a symbol of success.

Some speculation about why it works: The punishment for going to internet is precisely calculated: It is not too big to become unrealistic. For example if instead I decided "if I go to internet, I will kill myself", I know that I really wouldn't; I can't precommit to a serious self-punishment. However, I don't really have a good excuse for not eating a piece of paper. So this threat cannot be removed by some clever reasoning. On the other hand, licking the paper really is unpleasant. Doing it when I feel the temptation probably helps associate the unpleasant feeling with web browsing, which should help. It is easy to keep the paper always with me.

See you next Saturday!

After "dying" twice on Habit RPG and reading the first few chapters of Don't Shoot the Dog (via a Reddit post from EY, the evidence seems to indicate that I'm trying to make too big of a change too quickly (or, too many small changes).

I'm going to look for a way to pause or temporarily hide items on Habit RPG. I'm also open to similar sites that are more stable. Habit RPG tries really hard with good fundamentals but it doesn't work reliably in Internet Explorer or Firefox. I'm aware that they're rewriting the site but I'm trying to change things now and I suspect that using "Habit RPG is down, why bother today?" as an excuse isn't a good idea.

I haven't tried it, but dailies are set by default to be active every day of the week; you might be able to unselect all the days.

Is there any reason not to just delete the tasks that aren't working for you, though? It seems like if the task isn't something you actually want/need to do in the immediate future, it's more of a goal than a task.

I'm reluctant to delete things when it was so hard to get the items into HabitRPG to begin with due to the site's habit of ignoring user input and dying.

That being said, I ended up doing exactly what you suggested because there is no way to pause things. At this moment, HabitRPG's developers have posted on Tumblr reporting theirp rogress so I'm still using the site via a third party Android application that's been much more reliable than the website.

I've been a lurker here for a while now, but thanks to Beeminder I'm finally posting!

In an effort to increase my luminosity I decided to start positively rewarding any thoughts about my internal state with a smile. For a few days it seemed to be working wonderfully. I would think about my resolution, smile and feel good about it very often. After a few days I started responding to this smile by trying to gauge my emotions.

Turns out this was a bad idea. The whole reason I'm doing this is that I find this type of purposeful introspection difficult and frustrating. So I turned the positive reinforcer into a negative one. As a result, I haven't been nearly as introspective the last couple days. I'm switching back to simply rewarding self-aware thoughts, no more trying to provoke them.

Recently, when considering why I've done something, I've been consciously distinguishing between historical cause and justification. I can't seem to find the sequence post that prompted this effort. In the process I've come across some moderately vile or at least disappointing motivations. I experience considerable cognitive dissonance when I find myself doing something that appears to be right for what I consider to be the wrong reasons. Usually that involves doing something for signaling purposes that I really should have done anyway; I can't justifiably change my behavior, but I despise signaling as a motivation.

On a separate note, when considering my personal future, I've been trying to make explicit predictions with probability estimates about what will happen, at least in my own head. The intent was to stem my tendancy to catastrophise, but it seems to have backfired; in cases where I was catastrophising it turned out to be correct catastrophising, and I walked around for a few days thinking things wouldn't really be that bad when they actually were that bad. That was depressing. I intend to continue the mental predictions anyway.

In both cases I think I've sacrificed hedons for bayes points. I'm comfortable with the tradeoff but it's still a tradeoff.

I despise signaling as a motivation.

I'm curious why. After you've thought about it: va zl rkcrevrapr, gur uvfgbevpny pnhfr nccrnef gb or fvtanyyvat sbe zbfg crbcyr.

Sorry for taking so long to respond. I've been thinking about this off and on since you asked; it's a topic on which I don't trust my brain. I've come up with three possible responses, without reading your ROTed text:

  1. I treat esse quam videri as a terminal value, and signaling seems to thwart it.

  2. I think of signaling as inherently dishonest. (this is how it feels from the inside, but see below)

  3. I am subconsciously trying to signal that I am a non-signaler, or that I am exceptionally honest.

The specific form of signaling I find so distasteful is Impression Management, if it's relevent. It feels deceptive for Bottom Line reasons. I've been unable to expunge it from my behavior; trying to do so just ends with signaling something different on another level. It's a bit like trying to inform someone "I don't care what you think"; taking action inherently loses.

Now, unrotting: "in my experience, the historical cause appears to be signalling for most people."

I interpret this as suggesting that #3 is the correct cause; am I right, or do you mean something else?

I interpret this as suggesting that #3 is the correct cause; am I right, or do you mean something else?

"The correct" is too strong; I was shooting more for "a plausible." But it is mostly 3, but I don't think 3 and 1 are exclusive. If esse quam videri is a neat thing for other people to think about you, then it's worthwhile thing to signal! (I found amusing that one of the first Google Image Search results was of a tattoo of the phrase.)

My impression is that one has to do some sort of signalling, and you seem to have the same impression. It seems to me that doing it deliberately is probably better- especially if the result is deliberate white-hat signalling rather than only doing the black-hat signalling that you fail to notice is signalling.

Reporting a prediction: Two weeks ago I went to a talk on the Oregon Health Study. I predicted in advance a 70% chance of a result in favor of health insurance. While listening to the talk I adjusted this to 60% (in the talk they discussed the complications of extracting information from the study and I believed this would add noise). The results ended up being mostly in favor of health insurance.