The idea for this came to me when reading nyan_sandwich's "An akrasia case study." I outlined the idea itself in my comment in that thread.

So here's the plan:

  • In a comment reply to this, describe a specific instance of akrasia in your own life. Place an emphasis on the specificity. Focus on a specific task, either positive (i.e., that you judge to be good to do but self-sabotage from doing, like writing a paper) or negative (i.e., that you know you shouldn't do but do anyway, like buying more Frosted Flakes for breakfast and continuing to eat them). The more specific, the better.
  • You can share multiple instances, but please create new comments for each.
  • If you tried an anti-akrasia technique in this specific context, explain what you did and what effect it had. If you have some way in which you measured its effects objectively, please share that. If not, though, that's okay; we can still learn something from what various attempts to tackle different manifestations of akrasia feel like from the inside. The goal here is not to propose solutions; instead, it's to see what different things that feel like solutions seem to do to different kinds of akrasia. So even failed attempts are useful.
  • If you tried multiple approaches or if your approach requires some explanation, you might consider describing it and its effects in a reply to your description of the instance of akrasia you applied it to.

Let me emphasize one more time that we are not looking for solutions in this thread. Please don't give each other suggestions! If you think you're on to something hot in terms of the "kicking" aspect of the Art, please show us with a description of how the technique worked for you on a specific instance - but share the instance first. The goal here is not to demonstrate that you have a clever anti-akrasia technique. The goal instead is to see what different instances of akrasia and attempts to tackle it actually look like.

If at all possible, please share both successes and failures. This is especially helpful if we can see successes and failures of the same technique. This helps to balance out positive bias and gives us a better idea of the parameters within which different techniques work. Be especially wary if you have a favorite anti-akrasia technique because of the subconscious desire to attempt to change reality by pretending your favorite technique is actually perfect. If you do have a favorite technique, please actively seek out its true weak points.

Let's crack this thing!

New Comment
21 comments, sorted by Click to highlight new comments since: Today at 9:43 AM

INSTANCE: Checking "stuff" online

I have a number of things I like to check online: Google Reader, email, Less Wrong, my friends' blogs, etc. I find that if I don't make some kind of conscious effort to avoid the entropic slide, I'll default to checking them all impulsively in an irregular cycle. That is, I'll check email, then look at Google Reader, then Less Wrong, then check Google Plus, then Facebook, then think "Hey, I bet someone sent me an email by now" and then go check email again, etc. If I've slipped into this failure mode and honestly believe there isn't anything more for me to check, sometimes I'll start looking for new stuff to look at, like pulling up "Damn You Auto Correct."

This is especially insidious if I do it first-thing in the morning. If I get up and check my email right away, my day stands an unreasonably high chance of being totally unproductive in terms of my dissertation, job applications, or even getting household chores done.

One factor making this especially bad used to be that I'd have a nagging feeling when I was in "check my stuff" mode that I was forgetting something I usually want to check. This would prompt me to waste time exploring random junk on the internet until I either remembered something or gave up trying to remember. After a while I started developing a habit of feeling like I was forgetting something. So, I made a list of all the things I normally check in the form of hyperlinks. That cured the "Am I forgetting something?" problem - so much so that I don't need the list anymore since I can just visualize it and notice if I've missed anything. This did nothing for the larger problem, but at least it helped curb the entropic death spiral.

I do seem to have solved this problem, though. I'll explain in a reply since the explanation requires some verbiage.

APPARENT SOLUTION: Willpower weightlifting

I'll explain my thinking, but with the understanding that the thinking generated a solution for reasons that might have nothing to do with the thinking that went into the solution-generation.

It occurred to me that since I am godshatter, I should expect that I have many, many different utility functions. I'm also aware of the apparent fact from embodied cognition that physical enactment is a kind of reinforcement. Since I think it makes sense to think of akrasia as what happens when one utility function generates a behavior that another utility function judges as undesirable, it should be possible to eliminate akrasia by maximizing actions that support specific utility functions while minimizing actions reinforcing opposing utility functions.

The main mechanism for being able to do this, as I understand it, is the dorsolateral prefrontal cortex. In short, it's responsible for impulse control. There are three ways to train it that we currently know of, namely (1) mindfulness training, (2) doing novel and challenging things, and (3) encountering and resisting temptation. It occurred to me that I could use #3 in order to apply hormetism to honing willpower.

So here's what I've been doing since November 21st:

  • I started by listing all things that draw me in that I could think of. That includes my checking-a-thousand-things, but also things like wanting to work on my dissertation to get it finished, writing up solutions for my students, writing entries on Less Wrong, and poking at my skin in the mirror when I'm tired. Any regular activity I tend toward pre-reflection went on this list. (These probably weren't all in the same category, but that's okay. I still ended up with all of about 20 things on the list, including some things I'd want to reinforce and some I'd rather do without.)
  • Along with these, I made note of the "triggers." For instance, the "check my stuff" impulse would appear anytime I had access to the internet and had an idle moment.
  • I picked out a few I'd like to decrease and chose one (namely the "check my stuff" impulse) to use for training.
  • Every day I would pay attention specifically to the arising of that specific impulse. Whenever it would arise, I viewed it as an opportunity to train my willpower. I let my mind dwell on the possibility of following through, but I was very careful not to physically start following through at all. (The impulse control mechanism works on motor neurons. Starting something and aborting it turns out to be significantly easier and less intense as a "workout" than catching the impulse when it first appears and preventing it from moving your body is.)
  • In particular, I specifically sought out ways to trigger the "check my stuff" impulse in order to make it arise and then not follow through. Yes, I knew this would have the effect of shaping my impulse away, but that wasn't my main purpose. My main purpose was to strengthen my willpower. I was using the impulse as a "weight" upon which to develop my inner strength in general. The diminishing of the "check my stuff" impulse would be a pleasant side-effect. (This reframing turned out to be immensely useful to me.)
  • Finally, I would let myself do this to mental exhaustion in training sessions, avoid the trigger thereafter, and then give myself permission to check my stuff at predesignated times. For instance, I can check my email at the end of the day. It seems to be important to offer myself some time to recover, much like constantly weightlifting doesn't actually help your muscles nearly as much as having hard training sessions and then resting does.

This has had the effect of decreasing impulsiveness in general for me. While training, though, I find that I have to watch for rationalization rather than for getting overwhelmed by the impulse. Rationalization seems to be what "getting overwhelmed by the impulse" feels like, at least for me.

I should mention, by the way, that I haven't worked out a good way to avoid training the mind to be really good at sending the "I'm overwhelmed with exhaustion and have had enough training" signal prematurely. I haven't noticed this as a big problem, but of course I wouldn't if it were a problem, would I?

The measurable effect is that I now tend to check email twice a day and the rest of my stuff just once a day. I've also started to use rationalization as a signal of a wonderful opportunity for training rather than as something to which I'm overly inclined to trust. My brain keeps getting better at offering more impressive-sounding rationalizations, which is actually pretty useful; it keeps the intensity of training up.

If nothing else, I seem to be able to notice when I'm erring in this particular respect based on a gut-feeling that I guess I would call "guilt" if I had to tag it with something. It's very subtle, but I've learned to notice it because it appears along with a sort of inner "sigh" of relief when I find myself following through on an impulse I decided earlier I wouldn't follow through on. (It's sometimes surprising what my subconscious mind considers to be "starting to follow the impulse." I thought that opening a new tab, putting in the email address for one of my "things to check," and then putting my finger on the "Enter" key but choosing not to follow through would make the training more intense, but it actually feels like following through on the impulse even if I stop there.)

INSTANCE: I snoozed a lot, almost never woke up on time unless I had to, and often ended up sleeping way more than is reasonable or necessary.

TRIED SOLUTIONS: pretty much everything most people would recommend in this situation. More alarms, different alarms, alarms further away, rewarding myself with cookies when I got up on time, practicing getting up, try to feel good about waking up, try to behave as if I was the kind of person that loves mornings, on and on.

WORKING SOLUTION: I got really angry at myself, and I just decided to use the most extreme solution I could. I put up $1000 on for a pro-life charity (which I do not support). I wrote up a contract for myself, where I said that (with some minor exceptions) I could only get 9 hours of sleep per night for 30 days. If I fail, the money goes to the charity.

I've succeeded with my contract, and since then getting up on time has been a lot easier.

Wow, that is extreme. And potentially dangerous. Do you really think you would have followed through in the event of failure? I don't think I could have.

Regarding your problem: if you haven't taken steps to rule out narcolepsy, I recommend you do.

The symptoms are more likely to be the result of sleep apnea than narcolepsy.

(Incidentally, the earliest description of the symptoms of sleep apnea can be found in, of all things, a Charles Dickens novel...)

INSTANCE: I used to spend hours online on StumbleUpon when I knew I had to get up early for class in the morning. It would be 2 AM, I would think to myself "It is just a really wretchedly stupid idea to keep clicking 'Stumble'", and yet I would keep doing it. Sometimes I would respond to myself "Just one more page" or "Just five more minutes", but sometimes I would just say "Fuck you. You don't deserve sleep." I think part of it stemmed from the perceived hassle of getting ready for bed: take out contacts, brush teeth, take meds, etcetera. It just seemed short-term easier to keep Stumbling, so I wouldn't stop until I either berated myself enough that guilt outweighed inertia, or fear of falling asleep with my contacts in convinced me to get up.

SOLUTION: Got married. My wife is much better at getting me into bed than I was. Would not recommend getting married just to combat akrasia.


INSTANCE: arguing with Boyi

I was being super productive over the last few days, but then there is an interesting discussion going on in how not to lose an argument. It has sucked me off the tracks, manifesting itself like the Mercurial's Checking "stuff" Online: f5 f5 f5 f5 f5.

Now that I'm becoming more aware of it, my thoughts are tending towards "burn it". So fuck it I'm out of here for a while. I have too much to work on to waste my time in comment threads on LW.

I know Boyi in real life and find this really hilarious.

INSTANCE I was supposed to have things done ahead of time for a roleplaying game I was GMing (the GM is the guy who makes up the scenarios, for those who don't know). Frequently this did not happen due to me finding anything else to do during the times I had scheduled myself to work on it. Ended up winging my sessions very frequently, and it showed.


-Remove distractions; this failed miserably. I am apparently capable of distracting myself for hours on end by thinking about physics if the need arises.

-Don't schedule time to work, just do it when I feel like it; I worked a bit more since I felt less forced into it, but plateaued at a lower rate than I hoped.

-Reward myself with yummy snacks; was effective for a time, but did not last; not sure what the psychological effect there is.

CURRENT WORKING SOLUTION In desperation, I decided on a 'fight-fire-with-fire' strategy. I have the typical aversion to performing menial tasks (and occasionally my homework as well), so rather than scheduling two separate blocks of time for 'housework' and 'GM work' I scheduled a larger time block for 'Housework and/or GM work'. Turned out to be fairly effective, it allowed me to alternate between two nasty tasks without burning out (since I could simply walk away from the dishes when I got sick of them and go play with steampunk spaceships, or vice versa).

Note, however, a couple potential pitfalls to this trick: you need to have enough recurring nasty tasks that you won't run out of them, and depending on how fast you get sick of things you may switch between tasks at an unacceptable rate (for the record, my switch rate seems to be slowly increasing over time, so this may not work as a permanent solution).

-Reward myself with yummy snacks; was effective for a time, but did not last; not sure what the psychological effect there is.

I understand that this is called an extinction burst.

I would probably make an excellent actual case study in akrasia. I'll try to quickly summarize a few issues.

instance:Weight lifting.

After a bad break-up, I pursued it rigorously for about 6 months, with great success, creating a postitive feedback loop. I was fueled by the progress, which inevitably plateaued, leading me to stagnation. In an attempt to get back on track, I purchased a squat rack and barbell, so that I could work out whenever was most convenient. In retrospect, this was a really bad idea. I find it virtually impossible to put in an adequate workout at home, because it is too easy to be distracted. My average workout duration dropped from over an hour to probably less than 15 minutes. Attempting to precommit to not to interact with anything that wouldn't be at a gym failed, because my mind won't accept artificial constraints like that. I speculate that it failed for another reason: working out in a public gym probably triggers the status seeking parts of my brain, incentivizing me to look good by working harder now, as well as reminding me what I'm working for in the long-run (not that anyone cares at all about what anyone else does in the gym).

  • Do: Work out in public, measure progress in an objective way, reward yourself after a good work out
  • Don't: Make it too convenient, fail to take into account diminishing returns

instance:Undergraduate research

Much of my akrasia stems from anxiety. I had originally impressed the professor of a class I was taking, and later in informal research. In both cases I enjoyed the work, which was occasionally greuling, but where the price of failure was nonexistent - I was in no danger of receiving less than an A in the class, and the research consisted of occasional, casual meetings and setting my own pace and direction. In time, they offered me a payed position, which came with deadlines, frequent status reports, and so on. I noticed that both the quantity and quality of my work decreased, for two reasons. The original work had convinced me to go to graduate school, which subsequently made my supervisors opinion of me suddenly important (letters of reccomendation). Because of this, I started making promises and accepting burdens that were probably not unreasonable, but personally unrealistic. I began to feel incredible anxiety about this, so much so that even doing the work, but especially corresponding with my supervisors caused significant duress. It was perceived as laziness and nonresponsiveness; and I knew this, which made the anxiety worse: so bad in fact that I would avoid checking my email and seeing them on campus. Sometimes I'd be sitting there, knowing that there was probably an unanswered email in my inbox, KNOWING that I would have to read and answer it eventually, and knowing that waiting could only possibly make it worse, and for some reason I still would put it off.

In general, I feel that I have destroyed my ability to precommit. For a particularly ludicrous illustration: I have trouble getting up. My record for most times hitting snooze is probably > 15. I tried literally taping a caffeine pill on top of the button. The next morning, I just peeled it off and went back to sleep. Me_sleeping is an evil bastard.

INSTANCE: Going to an unscheduled job

Details- Job was high-risk/ high-reward. Extremely variable pay range. Was not required to go and did not have schedule, but did have a ~2 hour window of when a good arrival time was. Actual work was boring and stressful. No feelings of accomplishment besides earning money.

Risk-aversion factor: I found it extremely difficult to make myself go to work when there was a small probability (maybe P=.15) that I would make very little money, even though there was a much higher probability (P=.5) that I would make a lot of money. This was worsened by the fact that the possibility of large amounts of money made what would otherwise be considered a lot of money seem terribly tiny by comparison. Example using random numbers- If you expect to make $500/hr, then only making $100/hr seems unacceptably small, even though by any other standard it's really good. Recognizing this affect does not really alleviate it.

Default "strategy": I would earn whatever I needed that month in the first two weeks, and just not go the other two weeks, OR I wouldn't work the first two weeks, and then would work the second two weeks once bills came due.

Generally Unsuccesful Strategies:

1) Tried setting up a point and reward rule system. Rules were that I had to go 3 times a week, including one weekend day. I had points based on how much I made that would allow me to "earn" days off. Worked better than nothing, but not well enough.

2) "Getting in the Groove"- The more I worked the easier it was to go more often. If you get into the habit of going every other day, then it is easy to continue the habit. Of course the reverse of this is being in a "rut"- The longer it's been since you've been to work, the harder it is to get yourself to go back. This is always bound to happen sooner or later, due to scheduling conflicts, injury, etc.

3) Seeing that I didn't have the impetus to get myself to work, I made a reward system with a boyfriend that placed the impetus for getting me to work on him. This actually worked pretty well, but only so far as he had time for it.

Successful Strategy: Switched careers. Current job (child care) is highly-scheduled, low-stress, and set wages. Feelings of accomplishment, and being needed/appreciated. Get to read or surf internet at work (I'm working now!). I have to work about 4x as many hours, but am much more happy and relaxed.

EDIT: This may not seem like a strategy for dealing with akrasia, but I think finding what you have trouble with, admitting that it is a major obstacle, and re-directing your life to avoid it, is a highly effective strategy. It's the same way that I realized I should only take about 2 classes per quarter.

INSTANCE: Applying for academic jobs

I'm in graduate school finishing a doctorate. Last fall (2010) I thought I was going to finish, so I started looking around for jobs. At the time my advisor wasn't sure whether I would graduate that academic year (by summer 2011), so he was hesitant to write a letter saying I'd have my Ph.D. by fall 2011. He decided in December that I wasn't going to finish that year. But in the course of negotiating with him and looking for jobs, I realized that the rhythm for the academic job search required me to put my materials together sometime around September to October the year before the job would start and then send out applications from about October through December.

Fast-forward to this year. Throughout the summer I kept thinking I had plenty of time to work on my job application materials, so I didn't put much effort into them. Around September I started considering getting to work on it "soon." (For those who don't know this: Academic job searching requires putting together a curriculum vitae (kind of like a resumé on steroids), a description of current and planned future research, a description of one's teaching philosophy, a cover letter repeating much of the same content, and usually somewhere around three letters of recommendation.) I was also preoccupied with putting material together for a conference of sorts where I'd be presenting a small chunk of my dissertation work.

At the conference in early October, I overheard some other people talking about their submissions to a conference in Portland in February. One of the ways one gets jobs in my field is by going to conferences and making social impressions while showing off that you can do research. This particular conference is right in my area of specialization, so I felt a thrill of panic for having not attended to this sooner. I immediately put together a submission for a presentation. That took me until mid-October.

That's when I finally started putting my job application materials together. I figured it would take me a weekend to get it all together, and then trying to account for the planning fallacy I gave it a week. It actually took me a solid week of ignoring my dissertation just to get the curriculum vitae put together. (It's quite amazing how mind-numbing that process is!) I also ended up trusting my department's wall of job postings to do my basic search because I felt like I no longer had the time to figure out how to do the search myself.

Most job announcements started reviewing in November, though, so I figured I still had a chance to get most of my job applications in. I talked to three people about getting letters of recommendation, one of whom wanted my CV, research statement, and a set of reminders about work we had done in the past. I decided to get my advisor's feedback on the CV and research statement before giving them to the letter-writer. (It's pretty typical that the advisor needs to approve these things before one submits them.) He, however, took his time (as I would have known he'd do if I had bothered to reflect on it), and as a result I wasn't able to give the letter-writer her materials until late November - at which point most of the application deadlines had passed.

In retrospect, what I should have done is asked multiple people for letters of reference, started working on the job application material in the summer, and given the thorough letter-writer an early draft of the materials she had asked for right away.

INSTANCE: Ooh, shiny! (online)

Often, when I have some specific task to do on the internet, I'll get distracted by something else and forget about my original task, even if it was something that I was genuinely interested in. For example, let's say that I need to check how much a gym membership costs. I might grab my laptop, open it up, and think "I'll take a quick break (since I'm already on the internet)." Next thing I know I'm on LW or another blog engrossed in whatever new shiny popped up lately. The time it takes me to realize that I actually had something to do varies, but probably isn't more than 7 to 10 minutes. However, at this point, I actually have to rip myself away from the shiny, which can take a bit longer.

Instance: Letting my inbox forever grow so that important items get lost in a sea.

Attempted Solutions: A ton of hacks like email snooze features and GTD-like systems. But fundamentally it's a problem of just avoiding items that are sitting on your to-do list.

Actual Solution: Beeminding my inbox!

My situation is so bad that I am akratic concerning thinking about what to do with my akrasia.

You can still help the rest of us, and maybe yourself too, if you describe your situation. You don't need to think about a way to solve it. Just tell us what the situation is and how you know it's akrasic. (Of course, omit details as needed to feel comfortable sharing it!)

Sometimes after a break of a few days or weeks I find it hard to start work on a new painting (art) project. I have found it effective to use contrary thinking, so I tell myself to do no more than one hour of work and then I MUST stop, and I do this, then again the next day, say, an hour and a half, and then I MUST stop, which I do. By about the third day I am engrossed in the project, and stopping work is hard to do, and by then I'm over the initial problem. Works nearly every time and I have been using this technique for some years.

The relevant aspects are so bad that I really don't want to think about it. I know it is akrasia because it's pretty obvious. Discussing details, apart from being unpleasant, also damages my perceived status. Sorry for being too vague. I only wanted to stress that akrasia can easily expand to the meta-level, once you know about it.

The status part can be worked around by using a throwaway account.


Unless it's akrasia that involves generating throwaway accounts on LW. :)