My Fear Heuristic

post by lsusr · 2020-12-01T08:00:57.608Z · LW · GW · 9 comments

My friends and family call me "risk tolerant". I wasn't always this way. It is the result of a 3-year-long scientific experiment.

You must do everything that frightens you…Everything. I’m not talking about risking your life, but everything else. Think about fear, decide right now how you’re doing to deal with fear, because fear is going to be the great issue of your life, I promise you. Fear will be the fuel for all your success, and the root cause of all your failures, and the underlying dilemma in every story you tell yourself about yourself. And the only chance you’ll have against fear? Follow it. Steer by it. Don’t think of fear as the villain. Think of fear as your guide, your pathfinder.

The Tender Bar by J.R. Moehringer

When I was 18 I discovered a useful heuristic. Whenever I didn't know what to do I would pick whatever not-obviously-stupid[1] option frightened me the most.

My indecisions always centered around choosing between a scary unpredictable option and a comfortable predictable option. Since the comfortable option was always predictable, I always knew what the counterfactual would have been whenever I chose the scary option. If I chose the scary option then I could weigh the value of both timelines after the fact.

As an experiment, I resolved to choose the scarier option whenever I was undecided about what to do. I observed the results. Then I recorded whether the decision was big or little and whether doing what scared me more was the right choice in retrospect. I repeated the procedure 30-ish times for small decisions and 6-ish times for big decisions. If I were properly calibrated then picking the scary option would result in the correct choice 50% of the time.


  • For my 30-ish small decisions, picking the scary option was correct 90% of the time.
  • For my 6-ish big decisions, picking the scary option was correct 100% of the time.

The above results underestimate the utility of my fear heuristic. My conundrums were overwhelming social. The upsides earned me substantial value. The downsides cost me trivial embarrassments.

I terminated the experiment when my fear evaporated. The only things I still feared were obviously stupid activities like jumping off of buildings and unimportant activities like handling large arthropods. I had deconditioned myself out of fear.

I didn't lose the signal. I had just recalibrated myself.

  1. "Stupid" includes anything that risks death or permanent injury. ↩︎


Comments sorted by top scores.

comment by niplav · 2020-12-01T13:13:50.865Z · LW(p) · GW(p)

This post would be strongly improved by 3 examples of decisions you made differently due to this heuristic.

comment by Yitz (yitz) · 2020-12-02T07:27:47.941Z · LW(p) · GW(p)

Seconding this, examples would be extremely helpful here (they could be anonymized if you don't want to share personal details)

comment by lsusr · 2020-12-03T03:53:46.676Z · LW(p) · GW(p)

I agree. Someone else should replicate my experiment. I give prior advance permission to self-promote the results in this comments section.

comment by Matt Goldenberg (mr-hire) · 2020-12-01T21:43:57.347Z · LW(p) · GW(p)

I also had this heuristic for a while. It helped me a lot and gave me a lot of growth. However, it also made me make some bad social moves that didnt turn out well for me.

Eventually I stopped using this heuristic, and instead started viewing things that I feared as things to process and integrate instead of face.

comment by Vanilla_cabs · 2020-12-01T11:10:35.745Z · LW(p) · GW(p)

How do you make the difference between something you fear and something you suspect will be detrimental? Like, say, befriending someone who freaks you out, or entering a shady scheme?

comment by lsusr · 2020-12-01T12:00:44.333Z · LW(p) · GW(p)

Befriending someone who freaks me out and entering a shady scheme are both obviously stupid. If something is obviously stupid then this heuristic is overruled.

I use this heuristic only for marginal cases where my calculated expected value is equal for both options. (The net expected detriment of either choice compared to the other is zero.) Only if the previous condition is satisfied do I then check which of the two options triggers greater fear. Fear is an emotion. I imagine both futures and observe my emotional reaction.

comment by Gunnar_Zarncke · 2020-12-01T19:56:05.893Z · LW(p) · GW(p)

I think you should make this explanation part of the article. It's crucial.

comment by AllAmericanBreakfast · 2020-12-03T03:40:19.717Z · LW(p) · GW(p)

I'm exploring a practice of what I'll call "peer preview," trying to assess the value a post provides to its audience, what other valuable projects could emerge from it, and how it might be viewed a year from now.

There's a lot of advice and vivid descriptions of anxiety out there, and the benefits of overcoming it. But it's rare to get a quantitative perspective like this. The documentation seems like a lightweight technique with real potential benefits. I wonder if it's a common prescription.

It makes me wonder how many people who don't perceive themselves as neurotically timid or anxious have actually just managed to shield themselves from ever encountering anxious triggers. I can think of several things that seem anxiety-provoking, that might be a happy addition to my life, and which don't actually cause me anxiety because I so thoroughly avoid them. I can also see how a tacit part of many relationships is identifying what makes each other anxious, and securing tacit agreements that you will avoid those anxious triggers forever.

I'd be interested in writing about relationships that delves into the nuances of how to cultivate friendships and romantic relationships in which there's a conscious mutual understanding of that "anxiety avoidance" dynamic, and an explicit agreement to find ways to shift into a "do what scares us" paradigm.

Note: In response to feedback, I'm removing a portion of this comment that might not be constructive.

comment by George (George3d6) · 2020-12-02T20:24:08.305Z · LW(p) · GW(p)

The crux of the matter is determining what "stupid" means.

I'm afraid of boredom, darkness and prolonged effort.

Yesterday I decided a 8km treck from 2300 to 3800m (having no acclimatization in the last 5 months) was something I should do, in part to counteract these fears.

I had done similar feats of mouteneering before and I had equipment for 4k+ altitudes in the Alps (horrible wind, -10-20 temp and the like). My climb was with moderate-high wind with 10- -5 temperatures.

My "rational" self told me all of my fears were the kinds I should face, I was so over-equioed my one risk was heaving due to altitude sickness and having to turn back, something safe at that low elevation. Armed with the info I had yesterday I'd still agree.

Yesterday, however, I lacked knowledge about:

  • Just how much glasses help regulate face temperature
  • How hellish 90% air humidity feels with 100km/h winds in freezing temperatures
  • How inaccurate wind forecasts can be
  • How much havock the stress caused by darkness can reack upon my body's ability to allocate energy efficiently.

90% air humidity + 80km/h wind + -5 degrees on an exposed mountain face at 3000m with no acclimatization is a horrible experience, even with very good gear. By horrible I mean, I'm fairly sure I'd have risked long term injury or death had I turned back 1 hour later.

At the outtest the fear I had seemed like the kind that might stop you from flirting with a pretty girl at a restaurant.

In hindsight it was that kind of fear, but if the restaurant was in a very conservative region of saudi arabia and the girl was the favorite wife of a regional judge.

Overall I agree with the approach though. Just keep in mind fear is there in part to guard against incomplete information and bets that have much greater downsides.

Note: If you want a more suitable example think fear of doctor/dentist stopping people from going in for minor procedures (usually small upside) and the fact that, in many cases, this might actually be the correct choice, since even minor procedures carry a risk of death or chronic pain (cosmetic tooth extractions, vasectomies, IUDs are great examples). The above just came to mind due to proximity, but maybe it doesn't illustrated my point well, since most people would blanket ban high mountain climbing as "stupid" without room for appeal.