= 27d567bc2d48d9f40e9ccc20ea370b15)
Upvoted. The heuristic "uncertainty exists in the map, not in the territory" is in the first place meant to be an heuristic against frequentist statistics. One can argue that probabilities are properties of the things itself also in situations of purely epistemic randomness. The argument "uncertainty exist in the map, not in the territory" is used in this context to show that thinking probablilities existing as "thing itself" can lead to weird conclusions.
In response to
Map:Territory::Uncertainty::Randomness – but that doesn’t matter, value of information does.
In response to
Rationality Reading Group: Part R: Physicalism 201
If someone is interested about the book, i highly recommend the audiobook version. You can find it from: http://castify.co/channels/53
It took me 5 days to listen the whole book (volume 1-3).
Ok, lets say you are right that there does not exist perfect theoretical rationality in your hypothetical game context with all the assumptions that helps to keep the whole game standing. Nice. So what?
Speaking as someone who finds that placebos cure my headaches as well as painkillers do, I dispute that that is the question. When placebos work, they are a method of healing. Of course, you may not consider headaches to an illness, in which case I will not object, but will only point out that medicine does deal with headaches, so shifting that part of your caseload to a placebo dealer should help maximize your healing of sick people.
"Speaking as someone who finds that placebos cure my headaches as well as painkillers do, I dispute that that is the question.."
You should consider regression to the mean and illusion of correlation.
In response to
Nature publishes an article about alternative therapy
The article isn't so much about Reiki as about intentionally utilizing the placebo effect in medicine. And that there is some evidence that, for the group of people that currently believe (medicine x) is effective, the placebo effect of fake (medicine x) may be stronger than that of fake (medicine y) and (medicine x) has fewer medically significant side effects than (medicine y).
Sorry the misleading title and thanks for downvoting :D.The author goes much further than just ”utilizing the placebo effect”. The article is basically about endorsing alternative medicine. You can easily see this from the following quotes .
There are many shady arguments in the article:
”Conventional medicine, with its squeezed appointment times and overworked staff, often struggles to provide such human aspects of care. One answer is to hire alternative therapists.”
--> Just because there are challenges in medicine like overworked stuff does not mean alternative medicine practicioners should be hired.
”Critics say that this is dangerous quackery. Endorsing therapies that incorporate unscientific principles such as auras and energy fields encourages magical thinking, they argue, and undermines faith in conventional drugs and vaccines. That is a legitimate concern, but dismissing alternative approaches is not evidence-based either, and leaves patients in need.”
-->Dismissing alternative approach does not mean that the patient is leaved ”in need.” If the patient is in need the answer is not necessarily alternative medicine.
I have problems seeing the problem of utilizing placebo using evidence based medicine and at the same time NOT "hiring alternative therapists".... While acknowledging the limits of placebo.
Nature publishes an article about alternative therapy
Very interesting decision from the one of the leading scientific publications to publish an article about Reiki therapy.
http://www.nature.com/news/consider-all-the-evidence-on-alternative-therapies-1.18547
[Edit: should be Nature publishes an article about alternative therapy]
Help with relevant rational decision making research ideas
Im researching rational medical decision making and trying to find right angles to approach the subject of probabilistic decision making in medicine.
My first paper will be about the probability updating after the medical tests when we know the prevalence of the patients possible disease and the tests sensitivity and specificity.
Second paper will be (at the moment) about the decision to test if we have certain treatment treshold and know the prevalence of the disease and the tests sensitivity and specifity.
So im reasearching how well the descriptive medical decision making (usually intuitive) matches the normative ideal when in the decision making situation we have all the available info to actually calculate the best action.
If you have some ideas regarding this topic I would love to hear them. Thanks a lot!
In response to
Welcome to Less Wrong! (8th thread, July 2015)
Hi. Just leaving a few comments about me and what I have been doing in terms of research people here will find interesting. I joined just a couple of days ago so I am not so sure about styles, this seems to be the proper place for a first post and I am guessing the format and contents are free.
While I was once a normal theoretical physicist, I was always interested in the questions of why we believe in some theories, I think that for a while I felt that we were not doing everything right. As I went through my professional life, I had to start interacting with people from different areas and that meant a need to learn Statistics. Oddly, I taught myself Bayesian methods before I even knew there was something called hypothesis tests.
Today, my research involves parts of Opinion Dynamics (I am still a theoretical physicist there, somehow), and I have been starting to make more and more use of results of human cognition experiments to understand a few things as well as a Bayesian framework to generate my models. I have also been doing some small amount of research in evolutionary models. But my real main interest in the moment can easily be seen in a paper that I just put online at the ArXiv preprint site. Indeed, while I already knew the site and found it interesting, time limits meant I never really planed to write anything here. So, the reason I actually joined the site now is because I think you will find the whole discussion in the paper quite interesting. I do think that my main conclusion there about human reasoning and its consequences is so obvious that it always amazes me how deep our instincts must be for it to have remained hidden.
There is a series of biases and effects that happen when we decide to support an idea. And those biases make us basically unable to change our minds or, in other words, to learn. In the paper I inspect the concept of choosing an idea to support from what we know about rationality. I conduct a small simulation experiment with different models that suggest that our desire to have one only idea is behind extremist points of view, and I finally discuss the consequences of it all for scientific practice. There is a book planned, with many more details and aimed at the layperson, the first draft version is complete, but it will still take a while before the book is out. The article is in drier prose, of course.
Anyway, while I am still submitting it for publication, the preprint is available at
http://arxiv.org/abs/1508.05169
The name of the article is "Thou shalt not take sides: Cognition, Logic and the need for changing how we believe", I do think you people here will have a lot of fun with it.
Best, André
Thanks for the link! Very nice publication!
In response to
The Heuristic About Representativeness Heuristic
Thanks for the post. I love it.
My comments:
First sidenote that dont assume that if something is a heuristic it is automatically a wrong way of thinking.(sorry if i misinterpret this, because you dont explicitly say this at all :) In some situations simple heuristics will outperform regression analysis for example.
But about your mainpoint. If I understood right this is actually a problem of violating so called "ratio rule".
(1) The degree to which c is representative of S is indicated by the conditional propability p (c | S)- that is, the propability of members of S have characterestic c.
(2) The propability that the characteristic c implies membership S is given by p (S | c). (Like you write)
(3) p (c | S) / p (S | c) = p(c) / p(S)
This is the Ratio Rule= Ratio of inverse propabilities equals the ratio of simple propabilities. So to equate these two propabilities p(c|S) and p(S|c) in the absence of equating ALSO the simple propabilitis is just wrong and bad thinking.
Representative thinking does not reflect these differences between p(c|S) and p(S|c) and introduces a symmetry in the map (thought) that does not exist in the world.
For example: "Home is the most dangerous place in the world because most accidents happen in home. So stay away from home!!!" --> This is confusion about the propability of accident given being home with propability being home given accident.
= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
= f037147d6e6c911a85753b9abdedda8d){kind=logo}
And why do we need one of those? Most academics think religious warfare been B-ism and F-ism is silly, and you should use whichever is the most appropriate.
If you mean me and you...well we dont. I agree. But maybe one should ask that having Ronald Aylmer Fishers ideas about Bayesian statistics in mind: "the theory of inverse probabilities must be fully rejected"
Let me reprhase my quote: The heuristic "uncertainty exists in the map, not in the territory" is in the first place meant to be an heuristic against dismissing Bayesian concept of probability."