3 ms·

I think one of the most interesting things that have come from my study of the decision making literature is that there is a second field that Kahneman respects

I think one of the most interesting things that have come from my study of the decision making literature is that there is a second field that Kahneman respects, that stands opposed to his heuristics and biases (H&B) tradition. The field is called Naturalistic Decision Making (NDM), and it's mostly built around the study of intuition and expertise. The argument that the H&B camp makes is that heuristics and biases get in the way of proper judgments and decisions because they are shortcuts that our brain makes to conserve resources. The argument that NDM makes is that heuristics are how our brain works, and we should spend more time utilising how it naturally works to make better decisions, instead of fighting it and trying to build intuitions for probability (which are unnatural, and not what our brains are built for). The disagreement led both of Kahneman (H&B) and Klein (NDM) to write a joint paper: https://www.fs.fed.us/rmrs/sites/default/files/Kahneman2009_ConditionsforIntuitiveExpertise_AFailureToDisagree.pdf https://www.fs.fed.us/rmrs/sites/default/files/Kahneman2009_... the key takeaway of which is that the H&B approach is best suited for domains where expert performance is bad, and the NDM approach is better suited for domains where expert performance is reliable. The paper describes the conditions of the two domains. I've written a longer post summarising the findings from NDM at https://commoncog.com/blog/putting-mental-models-to-practice/ https://commoncog.com/blog/putting-mental-models-to-practice..., which exists as part of a longer series on putting mental models to practice. (Earlier parts cover the foundation of the judgment & decision making field, which is mostly built around the H&B tradition at the moment).

sah2ed 8y ago
You did a solid job outlining the 2 oppposing sides between heuristics and biases (H&B) and Recognition-primed decision making (RPD) in the blog article you authored and linked to. I’d like to invert your key takeaway by looking at it a bit differently by framing it in terms of the environment being observed rather than in terms of the expert’s performance. I’m reminded of this Murray Gell-Mann quote: “Imagine how hard physics would be if electrons could think" which I came across while reading an Economics piece. The writer used the quote to illustrate how much harder Economics as a field is, relative to a field like Physics. I agree with that characterization because, in general, electrons and by extension inanimate objects are not susceptible to behavior change from external stimuli, but humans are. Which is why expertise can be considered reliable when the environment consists of inanimate entities (e.g. astronomers, accountants) or animate but motiveless entities (e.g. livestock judges, grain inspectors). How reliable an expert performs in an environment largely depends on the level of autonomy enjoyed by the entities that constitute the environment under observation. An expert’s ability to reliably understand an environment correlates inversely with the number of entities that enjoy autonomy in that environment, because it is infeasible to model a real-world environment in your head. What experts do is to mentally model a tiny slice of reality from which to draw inferences from. Humans are entities that can think and act independently before, during and after an expert has been asked to intervene, which is why Doctors, Nurses and Auditors appear in both columns of reliable and unreliable performance for experts. (PS: I encountered 2 occurrences of the typo: “Marlie Chunger“) EDIT: reworded my comment significantly to make my point more clear.
- shadowsun7 8y ago
  I’m assuming that you didn’t read the section of the essay where I summarise Shanteau, Kahneman and Klein’s findings? Complexity of systems is one aspect, certainly, but valid and regular causal inferences are by far the dominating factor. Doctors, nurses, and auditors appear on both sides because they belong to fractionated fields of expertise. I have included a definition of ‘fractionated’ (their terminology, not mine) in the essay. (Also, Marlie Chunger is a fictitious character invented to illustrate some of the ideas in the series.)
  - sah2ed 8y ago
    > I’m assuming that you didn’t read the section of the essay where I summarise Shanteau, Kahneman and Klein’s findings? I did read your summary in the section titled A Personal Take, unless there is another summary? Link? (I’m wondering if your blogging tool allows a way to jump directly to sub-sections using anchor-links. This would be a time-saver for long-form essays.)
    - shadowsun7 8y ago
      Ahh, my apologies then. Yes, that’s the summary. You’re not the first person to point this out to me, but it turned out that the other reader stopped at ‘A Personal Take’ — which was what I thought happened here! (I was going to look at the essay again and ask myself if I made a mistake, since everyone seems to stop at that heading. It also meant my structure was badly chosen. And then I saw your reply!) Then, to address your idea more directly: I can see where you are coming from, but I think Klein and Kahneman’s definition of ‘has reliable causal cues’ + ‘provides opportunities to learn those cues’ are better filters. Klein studies some incredibly high performing individuals — experts who are able to perform well even in very complex environments. So it’s not always true that complexity of environment determines existence of expertise.

sah2ed 8y ago

You did a solid job outlining the 2 oppposing sides between heuristics and biases (H&B) and Recognition-primed decision making (RPD) in the blog article you authored and linked to. I’d like to invert your key takeaway by looking at it a bit differently by framing it in terms of the environment being observed rather than in terms of the expert’s performance. I’m reminded of this Murray Gell-Mann quote: “Imagine how hard physics would be if electrons could think" which I came across while reading an Economics piece. The writer used the quote to illustrate how much harder Economics as a field is, relative to a field like Physics. I agree with that characterization because, in general, electrons and by extension inanimate objects are not susceptible to behavior change from external stimuli, but humans are. Which is why expertise can be considered reliable when the environment consists of inanimate entities (e.g. astronomers, accountants) or animate but motiveless entities (e.g. livestock judges, grain inspectors). How reliable an expert performs in an environment largely depends on the level of autonomy enjoyed by the entities that constitute the environment under observation. An expert’s ability to reliably understand an environment correlates inversely with the number of entities that enjoy autonomy in that environment, because it is infeasible to model a real-world environment in your head. What experts do is to mentally model a tiny slice of reality from which to draw inferences from. Humans are entities that can think and act independently before, during and after an expert has been asked to intervene, which is why Doctors, Nurses and Auditors appear in both columns of reliable and unreliable performance for experts. (PS: I encountered 2 occurrences of the typo: “Marlie Chunger“) EDIT: reworded my comment significantly to make my point more clear.

shadowsun7 8y ago
I’m assuming that you didn’t read the section of the essay where I summarise Shanteau, Kahneman and Klein’s findings? Complexity of systems is one aspect, certainly, but valid and regular causal inferences are by far the dominating factor. Doctors, nurses, and auditors appear on both sides because they belong to fractionated fields of expertise. I have included a definition of ‘fractionated’ (their terminology, not mine) in the essay. (Also, Marlie Chunger is a fictitious character invented to illustrate some of the ideas in the series.)
- sah2ed 8y ago
  > I’m assuming that you didn’t read the section of the essay where I summarise Shanteau, Kahneman and Klein’s findings? I did read your summary in the section titled A Personal Take, unless there is another summary? Link? (I’m wondering if your blogging tool allows a way to jump directly to sub-sections using anchor-links. This would be a time-saver for long-form essays.)
  - shadowsun7 8y ago
    Ahh, my apologies then. Yes, that’s the summary. You’re not the first person to point this out to me, but it turned out that the other reader stopped at ‘A Personal Take’ — which was what I thought happened here! (I was going to look at the essay again and ask myself if I made a mistake, since everyone seems to stop at that heading. It also meant my structure was badly chosen. And then I saw your reply!) Then, to address your idea more directly: I can see where you are coming from, but I think Klein and Kahneman’s definition of ‘has reliable causal cues’ + ‘provides opportunities to learn those cues’ are better filters. Klein studies some incredibly high performing individuals — experts who are able to perform well even in very complex environments. So it’s not always true that complexity of environment determines existence of expertise.