Adversarial Collaboration

Kahneman hedges just a bit in his talk: &quot; To a good first approximation, people simply don&#x27;t change their minds about anything that matters.&quot;<p>Trying to explain the small number of people who do change their minds, David McRaney has an interesting book out called How Minds Change.[0]<p>And I imagine there are people who change minds, even outside his model. People who change their minds <i>when it doesn&#x27;t seem to help them</i> strike me as important people to hear from.<p>[0] <a href="https:&#x2F;&#x2F;www.econtalk.org&#x2F;david-mcraney-on-how-minds-change&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.econtalk.org&#x2F;david-mcraney-on-how-minds-change&#x2F;</a>

I was lucky enough to be in the group that ran Dr. Kahneman&#x27;s talk at Google, so I was at lunch with him &amp; 8 or so others. I asked him,<p>&quot;Dr. Kahneman, you&#x27;ve been writing about thinking for 40 years. Do you think you&#x27;ve changed how people think at all?&quot;<p>He said, &quot;No, not even me.&quot; He proceeded to tell a story where he fell into the &quot;eloquence trap&quot; that he, himself, wrote about: a doctor said something, and he said to his daughter, &quot;that doctor is very impressive!&quot;<p>She&#x27;d learned his lessons better than he had: she said that what mattered was how much experience the doctor had in this particular area, not how good she sounded.

One of the most disappointing things that I have learned is that most people hold opinions in order to be part of some group. If someone is a member of a group, it is almost not worth listening to their arguments, especially arguments in support of views held strongly by that group. They are arguing in order to maintain their group membership, not to find the truth. It appears this is true of academic and scientific disciplines as much as anywhere else.

&gt; Let&#x27;s start from the main domains where we know people don&#x27;t change their minds—politics or religion.<p>&gt; In politics and in religion, the main driver is social. We believe what the people we love and trust believe.<p>Except I&#x27;ve changed my mind on both, despite enormous pressure by people I love not to.<p>Also this article seems to be self-defeating: if we don&#x27;t change our minds, why attempt to convince us of that fact? If we truly don&#x27;t change our minds, no amount of evidence will help us change our minds about this topic.

&gt; The power of reasons is an illusion. The belief will not change when the reasons are defeated. The causality is reversed. People believe the reasons because they believe in the conclusion.<p>This is one of my pet peeves with even casual conversation. Someone goes &quot;I don&#x27;t like X, because Y.&quot; You point out Y is false, and the response is &quot;Well, I still don&#x27;t like X&quot;.<p>It used to annoy me because wanted to help people not base their views on bad information, as that&#x27;s what I would want people to do for me. But now, it mostly just annoys me because of how predictable it is. You know the only good answer is to agree with them, so what&#x27;s the point in even conversing?

There&#x27;s a simple trick that allows professionals in a given field to change their minds about something fundamental in their field - basically, it&#x27;s the Feynman approach, where he trained himself to accept the data and analysis even if he didn&#x27;t like the result, from an aesthetic viewpoint or whatever. A key strategy here is to have <i>something else</i> to put your ego and emotions into, such as a very subjective area like art, music, literature, etc.<p>Politics and religion might be trickier, as people tend to have external forcing factors (family and work opinions) that respond negatively to a fundamental change of some sort or other, up to being sent to prison for a decade for apostasy (see Saudi Arabia). In such situations, even if people do change their minds, they may not broadcast that change to anyone over fear of retaliation at family gatherings or in workplace environments. Even in cases where a particular member of a political party or religious group is shown to be criminally corrupt, many people will still embrace the politics or religion, on the basis that the individual in question is an outlier, not a representative sample - even after dozens of such examples are exposed. (Practically, this is why I&#x27;ve made it a rule to avoid political or religious discussions at work or at family gatherings, there&#x27;s just not much value to be found there).

&gt; To a good first approximation, people simply don&#x27;t change their minds about anything that matters.<p>I love the model of adversarial collaboration, and I don&#x27;t dispute the extremely strong influence of social bonds on knowledge formation, but Kahneman is just wrong about this. I know he&#x27;s wrong because I change my mind relatively frequently, about things of at least some consequence.<p>For a recent example, I was fairly sure that at the beginning of the pandemic, in the US, widespread, cheap testing would enable us to drive COVID cases near zero, and I wasn&#x27;t shy about telling everyone I met. Obviously, I was wrong, for a variety of reasons - so I updated.<p>That intimate experience with uncertainty and updating my own beliefs makes me wonder about Kahneman&#x27;s research methods. It makes me doubt whether this question is even tractable or whether people are even legible enough to researchers to draw conclusions about this.<p>Interestingly (and disarmingly) Khaneman is very forthright about the role his own experiences have played in convincing him that people in general don&#x27;t change their minds. He writes:<p>&gt; I was also impressed by the fact that Anne and I didn&#x27;t change our minds. I had read Kuhn and Lakatos about the robustness of paradigms, but I didn&#x27;t expect that minor theories would also be impervious to evidence.<p>also:<p>&gt; I will now share a personal experience of belief perseverance that I cannot shake ... However, it turns out that I only changed my mind about the evidence. My view of how the mind works didn&#x27;t change at all. The evidence is gone, but the beliefs are still standing. Indeed, I cannot think of a single important opinion that I have changed as a result of losing my faith in the studies of behavioral priming, although they seemed quite important to me at the time.<p>I think the most likely explanation for this is 1) social desirability bias has a dramatic influence on what information people make accessible about their cognition and 2) Kahneman is unusually stubborn, and his generalization from his own personality to all humankind is a manifestation of the typical mind fallacy. [0]<p>[0] <a href="https:&#x2F;&#x2F;www.lesswrong.com&#x2F;tag&#x2F;typical-mind-fallacy" rel="nofollow">https:&#x2F;&#x2F;www.lesswrong.com&#x2F;tag&#x2F;typical-mind-fallacy</a>

I know this isn’t exactly a counterpoint, since “The belief will not change when the reasons are defeated. The causality is reversed. People believe the reasons because they believe in the conclusion.” — but I have changed my political views dramatically and repeatedly over the course of the past decades. Of course, I <i>could</i> disprove the conclusion by accepting it based on the evidence, but I don’t want the universe to collapse in on itself due to paradox, so I won’t.

&quot;A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it.&quot; -- Max Planck

At least in religion I think the idea that reasons follow belief rather than precede it is quite acknowledged. I think that&#x27;s part of what in Christianity it meant when it says that faith is a gift of God. In my own conversion I felt it very intensely - my belief came over me in a sudden, external (or miraculous) way. I then spent 6 months learning everything there was to learn about my new found faith to placate my reason that I hadn&#x27;t simply gone mad, but the temporal flow was clearly belief -&gt; reason rather than the reverse.<p>In some ways it&#x27;s like falling in love. Most of us don&#x27;t evaluate potential partner on a bunch of metrics and then after they reach a sufficiently high ranking we declare ourselves in love. Rather we fall in love instinctively, then evaluate the person rationally (and sometimes reject them, like when the circumstances aren&#x27;t right or we can see their flaws even through the rose tinted glasses, etc.)

I&#x27;m not sure what he means by the naive scientific method. It has always been a fundamentally adversarial process, nothing new is being discovered here.<p>The scientific method as it has always existed requires two heuristics from you, the scientist: a hypothesis generator, a test generator (and an all important simplicity measure). The point of the method is that by making these things work adversarially, we can turn our very messy, fallible creativity, ie these heuristics, into facts about what the model isn&#x27;t. Moreover no additional hypothesis of what the model is, or test which contradicts it, can derive a false fact of this form, relegating the entire concept of adversary to the meta who-gets-their-name-on-the-paper game.<p>If Daniel has only recently started &quot;using science adversarially&quot;, then he has only recently started doing science.

If you haven’t read The Undoing Project, you should.

I believe this all boils down to mental energy an efficiency.<p>Our brains are great at pattern matching. We want to look at the landscape, see that there’s no hiding lion, and continue hunting-gathering. Constantly rechecking that there’s no lion is draining, and it also leads to less hunting-gathering. Adult humans really dislike having to do that (we even consider it a sign of mental illness)<p>The “back and forth” between husband and wife supports this theory. Every time one of them designed one experiment, “the ball was on somebody else’s field” and they could temporarily forget about the problem until the results came back. I am sure they were two very busy individuals and being able to “move on to other things” after designing the next experiment was gratifying.

This whole article reads like a cope for the fallout caused by Thinking Fast and Slow.<p>I purposely keep my copy faced forward on the bookshelf as a reminder of the dangers of reading too much into anything published in the past 15 years.

<a href="https:&#x2F;&#x2F;eprints.soton.ac.uk&#x2F;252301&#x2F;1&#x2F;El%2520Know%2520from%2520Exp%2520OBHD.pdf" rel="nofollow">https:&#x2F;&#x2F;eprints.soton.ac.uk&#x2F;252301&#x2F;1&#x2F;El%2520Know%2520from%25...</a><p><a href="https:&#x2F;&#x2F;citeseerx.ist.psu.edu&#x2F;viewdoc&#x2F;download?doi=10.1.1.982.3132&amp;rep=rep1&amp;type=pdf" rel="nofollow">https:&#x2F;&#x2F;citeseerx.ist.psu.edu&#x2F;viewdoc&#x2F;download?doi=10.1.1.98...</a>

I think what author describes and attributes to &quot;everyone&quot; is a trait of character. Yes, a common one, but not universal. From my own experience to that of (some of) my friends I can say with certainty people do change their minds.

Somewhat surprised to see Jeffery Epstein&#x27;s &quot;charity&quot; still showing up on the front page of HN

Much better title than &quot;People don&#x27;t change their minds&quot;, which is false

Fun fact, edge.org and the Edge Foundation was generously funded by Jeffrey Epstein -- without him, Edge probably wouldn&#x27;t exist as it does. <a href="https:&#x2F;&#x2F;www.buzzfeednews.com&#x2F;article&#x2F;peteraldhous&#x2F;jeffrey-epstein-john-brockman-edge-foundation" rel="nofollow">https:&#x2F;&#x2F;www.buzzfeednews.com&#x2F;article&#x2F;peteraldhous&#x2F;jeffrey-ep...</a>. From 2001 to 2017, &quot;foundations associated with Epstein provided $638,000 out of a total of almost $857,000 received by Edge over this period.&quot; So basically Epstein funded 80% of Edge&#x27;s budget.