This post is the first of a series on happiness and subjective well-being. In this one, co-written with my co-author Greg Kubitz, we discuss the result of a paper we have recently released showing that loss aversion may be though of as a feature of our cognition helping us make good decisions.
It’s really interesting. Your visualisation of the curve makes it intuitive to see the goal as a trajectory & this got me reading about the principle of least action. The curve made me think about energy in motion & I realised that you could see loss aversion as an adaptation to overcome inertia to set the system (goal seeking) in motion. Once the system is in motion, the energy input needed will quickly decay past a certain reference point. Hence the kink/inflection. You can see it in the trajectory of a ball that has a shallow rounded curve on the upwards journey & then a steep decay as a reference point is passed. We intuitively seek the path of least action & rule out paths that would impede our success (too much aspiration or too much anxiety).
Thanks for your comment Claire. In our case, I am not sure thinking about it in terms of motion fits the model. The curve is meant to help you make good decisions by allowing you to discriminate between the options you face. I'll discuss more these intuitions in the next posts in this series.
Sorry - my exposition was clunky. Can't promise this will be better but let me try to explain the metaphor. Goals are always paths - how do I get from point A to B? Even if a decision doesn't involve actual motion, there is still neuronal activity as we imagine ourselves at different points along that path. To imagine an outcome higher than our well calibrated, likely range of possibilities is psychologically demanding. We want to get from A to B but the more our imagined decision outcomes deviate from what appears possible / likely (the reference point), the more psychological effort is expended. There is some relationship between the effort expended in thinking about losses and the energy expended in thinking about gains. Loss related thoughts require effort (kinetic energy) whereas gain related thoughts are like potential energy - which do not require effort to be maintained. I was reading Feynman's lecture on path of least action and it got me thinking about what you had written. https://www.feynmanlectures.caltech.edu/II_19.html. If you imagine the effort required to think through the decision curve (i.e. to visualise every point on the curve) you can imagine how the initial part of the curve requires more energy / effort to overcome a fear of loss or failure. But once you have moved beyond that fear (the loss aversion), you are left with aspiration (which is all potential energy not effort energy). At this point, the curve decays quickly to satisfy the principle of path of least action. So when we make decisions, we put a lot more effort into thinking through the risk/loss aspects than we put into thinking through the aspirational/gain aspects. You could imagine the psychological effort as kinetic energy that decays quickly (to satisfy the path of least action) as we reach the aspirational bit. So loss aversion is not only adaptive, it also kind of aligns to laws of physics??
But if there is no air resistance, that trajectory will be symmetrical around the highest point. Once the ball leaves whatever threw it up in the air, there is no more energy input. It is "in free fall".
In addition to the explanation itself regarding loss aversion and a series on the science behind happiness that I find absolutely fascinating, I think an interesting contribution of this article is also the way it is written. For me, who writes my newsletter about the psychological, social and behavioral aspects of AI, by reading and talking above all about what researchers talk about in their papers, it is an excellent comparison and inspiration of clarity, of alternation with technical insights at different levels and alongside graphs, in addition to the intriguing world in which everything is told. I strongly recommend reading it, thanks for sharing.
It’s really interesting. Your visualisation of the curve makes it intuitive to see the goal as a trajectory & this got me reading about the principle of least action. The curve made me think about energy in motion & I realised that you could see loss aversion as an adaptation to overcome inertia to set the system (goal seeking) in motion. Once the system is in motion, the energy input needed will quickly decay past a certain reference point. Hence the kink/inflection. You can see it in the trajectory of a ball that has a shallow rounded curve on the upwards journey & then a steep decay as a reference point is passed. We intuitively seek the path of least action & rule out paths that would impede our success (too much aspiration or too much anxiety).
Thanks for your comment Claire. In our case, I am not sure thinking about it in terms of motion fits the model. The curve is meant to help you make good decisions by allowing you to discriminate between the options you face. I'll discuss more these intuitions in the next posts in this series.
Sorry - my exposition was clunky. Can't promise this will be better but let me try to explain the metaphor. Goals are always paths - how do I get from point A to B? Even if a decision doesn't involve actual motion, there is still neuronal activity as we imagine ourselves at different points along that path. To imagine an outcome higher than our well calibrated, likely range of possibilities is psychologically demanding. We want to get from A to B but the more our imagined decision outcomes deviate from what appears possible / likely (the reference point), the more psychological effort is expended. There is some relationship between the effort expended in thinking about losses and the energy expended in thinking about gains. Loss related thoughts require effort (kinetic energy) whereas gain related thoughts are like potential energy - which do not require effort to be maintained. I was reading Feynman's lecture on path of least action and it got me thinking about what you had written. https://www.feynmanlectures.caltech.edu/II_19.html. If you imagine the effort required to think through the decision curve (i.e. to visualise every point on the curve) you can imagine how the initial part of the curve requires more energy / effort to overcome a fear of loss or failure. But once you have moved beyond that fear (the loss aversion), you are left with aspiration (which is all potential energy not effort energy). At this point, the curve decays quickly to satisfy the principle of path of least action. So when we make decisions, we put a lot more effort into thinking through the risk/loss aspects than we put into thinking through the aspirational/gain aspects. You could imagine the psychological effort as kinetic energy that decays quickly (to satisfy the path of least action) as we reach the aspirational bit. So loss aversion is not only adaptive, it also kind of aligns to laws of physics??
I think I get a bit more your intuition. I am not sure it can be mapped onto our model. But that's something I'll think about.
That’s my point? And the symmetry of the trajectory is irrelevant unless I’ve missed something?
But if there is no air resistance, that trajectory will be symmetrical around the highest point. Once the ball leaves whatever threw it up in the air, there is no more energy input. It is "in free fall".
Thanks. Very interesting article.
Have you seen this article?
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7538829/
Thanks, I'll check it out.
In addition to the explanation itself regarding loss aversion and a series on the science behind happiness that I find absolutely fascinating, I think an interesting contribution of this article is also the way it is written. For me, who writes my newsletter about the psychological, social and behavioral aspects of AI, by reading and talking above all about what researchers talk about in their papers, it is an excellent comparison and inspiration of clarity, of alternation with technical insights at different levels and alongside graphs, in addition to the intriguing world in which everything is told. I strongly recommend reading it, thanks for sharing.
Thanks a lot, Riccardo, I am very glad you like this aspect of the post.