The Skill of Noticing Emotions

post by Neel Nanda (neel-nanda-1) · 2020-06-04T17:48:28.782Z · LW · GW · 6 comments


  The Skill of Noticing Emotions
    should you care about noticing?
    to apply it
      Motivation behind the Algorithm
  Appendix: Personal Examples

The Skill of Noticing Emotions

(Thanks to Eli Tyre and Luke Raskopf for helping teach me the technique. And thanks to Nora Ammann, Fin Moorhouse, Ben Laurense, Daniel Hynk, Nathan Young, Toby Jolly, Michael Ng, James Walsh and Jaime Sevilla for feedback on various drafts!)

(For those familiar with the core idea, you might find it more interesting to skip to the long list of personal examples in the Appendix)


I was introduced to a technique called Noticing at my CFAR workshop back in October, to become better aware of my emotions and better at productively reacting to them. I’ve since found this a really powerful technique, and it’s become a pretty key tool for solving problems in my life. My goal in this post is to give my take on Noticing, explain how I actually go about using it, and hopefully convince you to give it a try!

What do I mean by Noticing? I first want to introduce the idea of an emotion or mental state being noticeable: something that feels important and urgent and is immediately promoted to conscious attention. An excellent example of a noticeable experience is hearing my name. When I hear someone say “Neel” it immediately feels important and captures my attention. Even if I’m busy and focused on something, hearing my name can easily break that focus and cause me to change what I’m doing.

Being noticeable breaks down into two parts: something that I’m aware of and something that feels important. For example, I’m often aware that I’m procrastinating, but can’t muster the motivation to actually do anything about it. It’s something I am aware of, but not something that feels important. I like to think of this as spending most of my life on autopilot, focused on the present moment. A noticeable event or emotion will turn off autopilot and engage my conscious mind. Hearing my name does disable autopilot, but if I’m aware that I’m procrastinating and continuing to do so, I’m still on autopilot.

Noticing, then, is a technique to take a specific emotion or mental state[1] and deliberately make it noticeable. This essentially installs automatic triggers to turn off my autopilot. This is extremely useful, because a lot of problems in my life dissolve when I can just be more self-aware at the right times! For example, I’ve had a lot of success with noticing the feeling of defensiveness in arguments. By default, I’ll often lash out and stop arguing in good faith when defensive. But by making the feeling of defensiveness noticeable, I can recognise in the moment where the urge to lash out comes from. This doesn’t stop me becoming defensive, but once I’ve identified the problem I can take steps like removing myself from the situation.

Why should you care about noticing?

In general, fixing a problem involves figuring out the right thing to do, the right time to do it, and then actually doing it in the moment. And I often find it easy to do the first two, but then forget to do the right thing in the moment. One symptom of this is that I find it much easier to give other people advice than to apply it myself. For example, it’s easy to understand a bias like the planning fallacy in the abstract, and even to identify it in friends. But it’s much harder to notice in the moment when I'm falling prey to it and to put in the effort to correct for this. I find Noticing valuable for bridging this gap, and helping remind myself in the moment to actually apply the ideas I already understand. I spend much of my life on autopilot, focused on what I'm doing, and Noticing helps me be self-aware at the times when it’s most useful to be.

A concrete example of how I’ve used this: While working, unrelated ideas often pop into my head and I feel an urge to explore them. Eg an impulse to check my messages or google a random fact. This always feels justified in the moment, and like it’ll be a quick detour. But in practice it’s incredibly hard to stop procrastinating, often starting a half hour rabbit hole that leaves my original task long forgotten. This makes it near impossible to maintain focus and deep work. With Noticing, I’ve been able to make the feeling of those fleeting urges feel noticeable. And by becoming self-aware at the moment of distraction, I can recognise that the urge just feels important, rather than actually mattering. This doesn’t work perfectly, but it triggers several times a day, and has made it much easier to focus.

I’ve found a pretty wide range of things can be improved by better control of my autopilot. A few general categories:

Further, by practicing Noticing for a while I’ve found myself developing a more general skill of being self-aware. I've found it notably easier over time to be aware of how I'm behaving and why, and better at tracking my emotions. And it's much easier to start noticing something new!

How to apply it

I’ve hopefully convinced you that Noticing is useful, so how to actually use it? My approach is to install a mental reflex, where every time I feel the mental state I take a simple physical action, like snapping my fingers. I call these actions markers, and their purpose is to take the experience of Noticing outside of my head and make it harder to ignore.

I’ll now outline the exact algorithm I use to install these reflexes[2], but in general I expect this to be a pretty personal process. I expect the best approach to vary a lot between people, so I encourage you to adapt it to whatever feels most natural and helpful!


  1. Choose the mental state
    1. This should be as specific as possible. It doesn’t have to be easy to put into words, but should be eg one you have clear memories of
    2. Normally I first identify a problem which could be resolved by being self-aware at the right times, and then identify a relevant mental state. Good prompts:
      1. When do I think “I should have known better”?
      2. What do I often regret?
    3. A good litmus test for whether Noticing fits your problem: “If I could set an alarm to go off in my head at the right time, would this problem feel solved?”
  2. Choose your marker action
    1. This should be a small and subtle physical action. It’s important that it’s something you can always do, eg snapping your fingers or tapping your foot
  3. List 10 previous examples where you've felt this mental state
    1. Ideally ones that are recent, and that feel visceral - where they've really stuck in your mind
    2. 10 is an arbitrary number, the point is that more examples are always better. I recommend setting a 5 minute timer and spending the full 5 minutes brainstorming. It’s easy to list examples off the top of your head and then feel stuck, but it’s surprising how many more examples you can find with more time.
  4. Mentally simulate each example, and take the action when you feel the emotion
    1. Try to really relive the experience, in as much visceral detail as possible. The goal is that it’s something you feel rather than something you're describing
    2. Add as many details as possible to flesh out the scene
      1. Eg what were you saying? What did it feel like? Where were you? What could you see?
    3. Look out for cues associated with the state, eg physical sensations, thought patterns you have, common contexts
  5. Actively practice this over the next 2 weeks - I call this the learning period
    1. Keep it in the back of your mind that you're practicing Noticing on this mental state
    2. Leave yourself regular reminders, eg post-it notes next to your bathroom mirror, daily email reminders, a list you check as part of your morning routine
    3. It can be helpful to track the times you successfully notice over this period, eg incrementing a counter, or writing it down

Motivation behind the Algorithm

The following is my model for why this algorithm works. All models are wrong, and so this is almost certainly incorrect in important ways, but I find this useful for motivating the algorithm. When tweaking the algorithm, I think it’s more important to keep to the spirit of this model than to keep to the letter of the algorithm.

When I’m focused on a task, on autopilot, most of my conscious attention is going towards it. But some part of my subconscious mind, my awareness, is still aware of what’s going on around me. There’s always going to be a lot of unimportant background stimuli, so my awareness will ignore things by default. But it needs to be able to promote important things to my attention, so it has a list of a few important things. And when it detects something with a strong association to something important, it flags that in my conscious attention. This is the experience of something being noticeable. So, to make a mental state noticeable, I need to make it feel important and to create strong associations with it.

How does the algorithm actually help with this?

  1. Choosing as specific a mental state as possible is valuable, because if it feels clear and concrete in my mind, the associations will be stronger. There’s a clear target to latch on to.
  2. The marker action is useful to help it actually feel important in the moment.
    1. It’s important that it’s physical, because it’s easy to be somewhat aware of the emotion but for it not to feel important and be ignored. Just as I can be aware that I’m procrastinating but it doesn’t feel important enough to be able to stop. Taking a physical action, even a simpler one, makes it much harder for my mind to implicitly ignore the emotion
      1. It’s easy to skip this part, and think you can ‘just notice’ - I highly recommend having a physical marker
    2. The marker should be a simple action that takes minimal willpower and can always be performed. This makes it easier to build the reflex of paying attention to this mental state. The goal is to help your mind focus, rather than to directly solve the problem
  3. Mentally simulating historical examples helps make it feel important, because my mind is extremely good at pattern spotting. These simulations give it a bunch of visceral data points of “when I feel this emotion, it’s important and I react to it”.
    1. It’s normal for this to feel a bit over-the-top, you want to really drill in that this is a reflex. The point is to do it enough times for my subconscious mind to spot a pattern, rather than stopping when I feel like my conscious mind gets it.
  4. When simulating, it’s also useful to look for as many cues for the mental state as possible. Ie things that correlate with it, like physical sensations and associated contexts and emotions.
    1. This increases the probability that my awareness notices one of the associated cues, and flags it to my attention.
    2. I call this increasing the surface area on the emotion, I want to understand it and what it looks like in as much detail as possible.
  5. The learning period is important because it’s hard to build this artificial association that the emotion is important. In the short term, this association will be quite weak, and keeping it in the back of my mind helps me respond to it. As it becomes more familiar, the association becomes stronger and is more likely to stick in the long term.
    1. A useful framing: The default state of the world is that I will forget about all new habits I develop. This isn’t something I can resolve by just “trying harder”, I need to take action and create external reminders to help it stick long term



Ultimately, I'm extremely excited about Noticing because a lot of my problems boil down to not being self-aware at the right time. I think the bottleneck for learning a lot of good mental habits is doing the right thing in the moment, and Noticing has given me an extremely powerful tool for actually doing this. If you're reading this and empathise with the kinds of problems I've outlined, I urge you to take some time to try it out yourself!

Noticing is a very general technique, so I’ve given a long list of specific ways it’s worked for me in the Appendix. I find it easy to get excited about the potential of a new technique, but fail to come up with any clear direction for how to apply it, and end up forgetting about it. So hopefully this list can provide some inspiration for specific ways it could be useful to you.

I've found it especially interesting to practice Noticing in the longer term. It's felt like there's a general meta-cognitive skill of "being aware of how I’m thinking and what I’m feeling" that I've been developing. I've both found it much easier to begin noticing something new, and found it generally easier to be self-aware and pay attention to what I'm feeling. So if you feel excited about the idea of Noticing, I highly recommend just trying it out on something, even if it doesn’t feel like a perfect fit. You aren’t just solving that specific problem, you’re training the general skill of self-awareness!

It's extremely easy to be Typical Mind Fallacy-ing when talking about techniques like this, and I expect different things work best for different people. I'd be extremely interested in hearing about other people's experiences with Noticing, or other effective tools for these kinds of problems!

Appendix: Personal Examples

A few notes:




Mental health

  1. I’m using mental state very broadly here, this technique can be used for a range of things: emotions (like guilt, anxiety), mindsets (like defensiveness), internal experiences (like ‘failing to plan’), sensory experiences (like ‘hearing birds chirping’). It’s all about finding common themes between events. ↩︎
  2. This algorithm is pretty heavily based on CFAR’s Trigger-Action Pattern framework. I call these Empty TAPs, because the purpose of the action is just to highlight the trigger better, rather than having a direct purpose ↩︎


Comments sorted by top scores.

comment by Mark Xu (mark-xu) · 2020-06-04T21:58:07.477Z · LW(p) · GW(p)

I like this post and think that I have personally gotten really large gains by just noticing various sensations.

A modification of the algorithm is to purposefully place yourself in situations where you experience the mental state instead of just imagining it. For instance, something I found valuable was opening reddit and just kind of sitting in the state of wanting-to-mindlessly-consume-content. The caveat is to not do this in a way that might be damaging.

comment by Pattern · 2020-06-07T19:27:45.304Z · LW(p) · GW(p)

The links too and from the footnotes are broken.

Replies from: habryka4
comment by habryka (habryka4) · 2020-06-07T23:23:05.908Z · LW(p) · GW(p)

Alas, I tried to fix it for the author, but it's not super easy. We don't (yet) have first-class support for footnotes in the WYSIWYG editor, so I can't really make the footnotes work in the way they were probably intended.

comment by athom · 2020-10-01T01:29:06.312Z · LW(p) · GW(p)

Weird question, but if your marker action is doing something with your hands, what do you do if you're holding something at the time?

Replies from: neel-nanda-1
comment by Neel Nanda (neel-nanda-1) · 2020-10-01T05:54:56.036Z · LW(p) · GW(p)

Huh. Somehow that has never come up before...

My default reaction when holding my phone was to either put the phone down, or just to tap my fingers together

comment by Roshan (roshan-1) · 2020-06-04T21:04:07.388Z · LW(p) · GW(p)

Hi Neel,

Have you thought about physical habits as well as mental? Have you tried to ingrain any behavioural habits in order to benefit your life?

For example, some (perhaps odd) personal examples include trying to improve my tongue posture and body posture. For habits in general, do you have an opinion on how much time should be allocated for focused and 'unfocused' work?

PS I really enjoyed the post, I have been trying to many of the same things, however I haven't been tracking/tallying occurrences which I will try! I like the idea that given our neuroplasticity, we could potentially engineer mental and physical habits for long-term happiness!