We study a large sample of 656 undergraduate students, tracking the evolution of their beliefs about their own relative performance on an IQ test as they receive noisy feedback. … Subjects (1) place approximately full weight on their priors, but (2) are asymmetric, over-weighting positive feedback relative to negative, and (3) conservative, updating too little in response to both positive and negative signals. These biases are substantially less pronounced in a placebo experiment where ego is not at stake. We also find that (4) a substantial portion of subjects are averse to receiving information about their ability, and that (5) less confident subjects are more likely to be averse. We unify these phenomena by showing that they all arise naturally in a simple model of optimally biased Bayesian information processing … [of] agents who derive utility directly from their beliefs (for example, ego or anticipatory utility). (more; HT Dan Houser)

They also have results on how overconfidence relates to IQ and gender:

We show that agents who are of high ability according to our IQ quiz, and hence arguably cognitively more able, are just as conservative and asymmetric as those who score in the bottom half of the IQ quiz. … In our data women differ significantly in their priors, are significantly more conservative updaters than men while not significantly more asymmetric, and significantly more likely to be averse to feedback. These gender differences are consistent with our theoretical framework if a larger proportion of women than men value belief utility.

We present an evolutionary model showing that, counterintuitively, overconfidence maximizes individual fitness and populations tend to become overconfident, as long as benefits from contested resources are sufficiently large compared with the cost of competition. In contrast, unbiased strategies are only stable under limited conditions. The fact that overconfident populations are evolutionarily stable in a wide range of environments may help to explain why overconfidence remains prevalent today.

They model pairwise contests over a discrete prize, won by the most able side. Each agent chooses if to fight, after seeing the other agent’s ability with error. If both agents fight, they pay a conflict cost. Agents can win by committing to overestimate their own ability, as this makes them fight more, which induces opponents to fight less.

I found that strategic commitment effect in ’06, in a paper that seemed too simple to publish in econ theory journals:

In a simple model of conflict, two agents fight over a fixed prize, and how hard they fight depends on what they believe about their abilities. To this model I add “preagents,” representing parents, leaders, or natural selection, who choose each agent’s confidence in his ability. Depending on the reason for such confidence, I find five different patterns in how confidence varies with ability. Agents who estimate their ability with error have under-confidence when ability is high and over-confidence when ability is low, while strategic commitment incentives induce the opposite pattern. Agents who misjudge their value for the prize, relative to their cost of effort, induce an over or under-confidence that is independent of ability, while cooperating pre-agents choose extreme under-confidence. Agents who use confidence to signal ability have a relatively uniform over-confidence.

I doubt Nature would have published my paper either. My paper used a few lines of math of simple game theory, while the paper Nature published used lots of evolutionary simulations, which don’t seem to add much beyond simple game theory. Based on this and other cases I’ve seen, I conclude Nature doesn’t care what econ theorists think about the social science papers they publish.

Without a doubt, doubt is useful and needed at the margins of any ideology. The world is too complex to know completely. Many of our judgments are, by nature, provisional. Those who are immune to evidence, who claim infallibility on debatable matters, are known as bores – or maybe columnists.

Doubt becomes destructive as it reaches the center of a belief and becomes its substitute. A systematic skepticism may keep us from bothering our neighbor. It does not motivate a passion to fight for his or her dignity and rights. How do ambiguity and agnosticism result in dreams of justice, in altruism and honor, in sacrifices for the common good? What great reformers of American history can be explained by their elegant ambivalence? (more)

Ask yourself this simple question: how confident would you need to be on a moral or political conclusion in order to work passionately for it? 99%? 90%? 75%?  If you have such an action-threshold, and this threshold is high, well then yes, to let your passion flower, you may need to lie to yourself about your confidence. So that you might actually do something.

Would your overconfidence then lead you to do too many things too enthusiastically?  If so, perhaps you’d do better to also allow yourself some other more graded psychological reluctance to passion, to counter this bias.

But it would of course be even better if you could see the nobility and glory in doing your best as a limited but well-meaning creature. You shouldn’t need to be absolutely sure of a conclusion to work sincerely and passionately for it.

In situations where it is empirically optimal to follow others and contradict one’s own information, the players err in the majority of cases, forgoing substantial parts of earnings. The average player contradicts her own signal only if the empirical odds ratio of the own signal being wrong, conditional on all available information, is larger than 2:1, rather than 1:1 as would be implied by rational expectations. … In total, the meta dataset contains 29,923 decisions made by 2,813 participants in 13 studies. All participants observe a private signal and a (possibly empty) string of previous choices made by others in analogous situations. In all decision problems there are two actions and two possible payoffs, but the dataset nevertheless comprises a large variety in environments, instructions, players’ personal characteristics, and histories of other players’ choices. (more)

Of course copying others’ acts sends a bad signal about our confidence in own own info and and analysis abilities. So it can make sense to focus more on one’s own clues to the extent is is important to send a positive signal to observers.  Just beware of assuming too easily that such gains are substantial.

]]> http://www.overcomingbias.com/2010/12/the-wisdom-of-others.html/feed 6 http://www.overcomingbias.com/2010/09/overconfidence-signals.html http://www.overcomingbias.com/2010/09/overconfidence-signals.html#comments Sat, 11 Sep 2010 18:45:46 +0000 Robin Hanson http://www.overcomingbias.com/?p=24231 Continue reading →]]> We test three different theories about observed relative overconfidence. The first theory notes that simple statistical comparisons … are compatible with a Bayesian model. … Data on 1,016 individuals’ relative ability judgments about two cognitive tests rejects the Bayesian model. The second theory … We test an important specific prediction of [self-image concern] models: individuals with a higher belief will be less likely to search for further information about their skill, because this information might make this belief worse. Our data also reject this prediction. The third theory is that overconfidence is induced by the desire to send positive signals to others about one’s own skill. … Personality traits strongly affect relative ability judgments in a pattern that is consistent with this. … Overconfidence in statements is most likely to be induced by social concerns than by either of the other two factors. (more; HT Tyler)

People generally tend to believe they are more competent than they actually are, and this effect is particularly pronounced among poor performers. … One striking demonstration, the illusion of explanatory depth (IOED), arises when people overestimate their ability to explain mechanical and natural processes. For example, people know that a zipper closes because it has teeth that somehow interlock, but they know very little about how the teeth actually interlock to enable the bridging mechanism. Similarly, many people know vaguely that an earthquake occurs because two geological plates collide and move relative to one another, but again they know little about the mechanism that initially produces these collisions. Nonetheless, people believe they understand these concepts quite deeply and are surprised by the shallowness of their own explanations when prompted to describe the concepts thoroughly. …

People who construe a ballpoint pen abstractly are more likely to focus on the pen’s function and perhaps its global appearance. In contrast, people who construe the pen concretely are more likely to focus on how well they understand how its parts work together to enable the pen to function—in this case, the appropriate metacognition. Accordingly, people are less likely to overestimate their understanding of how the pen works when their introspections focus appropriately on the pen’s concrete features rather than its abstract features. …

In six studies, we showed that IOEDs arise at least in part because people sometimes adopt an inappropriately broad or abstract construal style when evaluating their understanding of concrete processes. … Participants … experienced larger IOEDs the more abstractly they construed 13 basic human behaviors. … Participants rated their knowledge of how three mechanical devices worked more accurately when the devices were framed more narrowly according to their component parts. When asked to express how those devices worked, only participants in the broad construal condition were surprised by the incompleteness of their explanations. …

Participants were induced to adopt a concrete or an abstract mindset by expressing how (concrete) or why (abstract) they engage in certain everyday processes, like getting dressed in the morning. Again, participants in an abstract mindset tended to show a significantly greater IOED. … Participants … reported understanding their favored 2008 Presidential candidate’s policies better than they actually did when asked to express those policies in writing. … Participants who adopted a more abstract construal style showed a more pronounced illusion of political sophistication. …

Our findings suggest that … when adopting an abstract construal style, people might therefore be systematically overconfident about what the future holds and how well they understand themselves and others. … The IOED is both similar to and distinct from a range of overconfidence biases documented. … According to one account, egocentric over-confidence effects tend to emerge because people anchor on their own subjective experiences and fail to adequately account for the experiences and abilities of other people. … Other researchers have suggested that people are overconfident because … their memories tend to be overpopulated with successes rather than failures.

I strongly suspect these patterns are driven mostly by customers, i.e., that more accurate professionals would be less successful in inspiring confidence by others in them.  If you are a successful professional, that is probably in part because of your unjustified arrogance.

Added: Carl reminds us of an ’06 post on overconfident software managers.

Imagine that you were thinking of buying or building a house. Now consider various possible hypothesis you might have about your degree of influence over this resulting house.

At one extreme, you might fatalistically assume you had no influence. For example, you might think your spouse will pick the neighborhood, house, and all later home improvements, and that you’d have zero input. If this assumption were mistaken, you might later regret that you’d invested little effort in thinking about what you wanted, or what was feasible.

At the other extreme, you might assume you had budget and approval for a huge estate and mansion anywhere you wanted.   So you might sketch out elaborate designs – the bowling alley goes here, ballroom to the south, the helipad over there, and so on. If your budget was actually far smaller, however, most of this effort might be wasted.

Yes, it can be good to spend a bit of time considering a wide range of influence levels. Sure, sometimes you might think about what you’d do if you won the lottery, or if you were locked in jail for decades. But surely most of your planning should be done matched to the scale of your actual degree of influence. Not much point in shopping for the best private jet if you can barely afford a car.

The same principle applies to our strongest relations, such as romance and friendship. These matter greatly deal to us, and so we’d very much like to control them. We make lists of what we want in our mates and allies, we rehearse what we will and won’t accept from partners, and we analyze our interactions to assure ourselves we understand what is happening.

But much of this is illusory overconfidence and over-reach; we usually have far less control over and understanding of our relations than we think. Sure we can list features we like and dislike, all else equal. And we might be mostly correct about which way those features influence our attraction. Even so, we mostly just don’t know why we like some and dislike others. Sometimes we don’t even realize who it is we like and dislike.

If we calculate that it would be in our interest to like or dislike someone more, we have only a very limited ability to actually make ourselves do this. Even when we decide we’d be better off breaking it off a relation, we can find that quite hard to actually do so. More likely we’ll break something off and then make up reasons about why that was a good idea.

I’m not saying to never think about your relations; I’m saying such thinking is more useful when you are more realistic about your influence. Of course if others get wind of your realism they may respect you less, or think they can walk all over you. So in that way it might be in your interest to be somewhat deluded about your influence. And you won’t get to be a famous inspirational speaker on relationships by speaking honestly about them.  But be careful to not take your confident image too seriously.

The same principle also applies in futurism. It is tempting to think we can remake the universe to be anything we now collectively want, and so to spend great efforts wondering how exactly we would want the universe to be if we had our druthers. But if we are actually very constrained in our influence, most of this effort will be wasted. Oh it might be a helpful exercise in far-mode thinking, to affirm far values and assert confidence in our abilities.  But it might not do much for the future.

When our ability to influence the future is quite limited, then our first priority must be to make a best guess of what the future will actually be like, if we exert no influence. This best guess should not be a wishful assertion of our far values, it should be a near-real description of how we would actually bet, if the asset at risk in the bet wer something we really cared about strongly. And yes, that description may well be “cynical.”

With such a cynical would-bet best guess, one should then spend most of one’s efforts asking which small variations on this scenario one would most prefer, and what kinds of actions could most usefully and reliably move the future toward these preferred scenarios. (Econ marginal analysis can help here.)  And then one should start doing such things.  Yes this approach seems less noble, fun, and optimistic, and talking this way won’t make you an inspirational futurist, speaking at all the hip conferences. Even so, those small shifts are what would actually most help the future.

This paper uses … [a] measure of overconfidence, based on CEO stock-option exercise, to study the relationship between a CEO’s “revealed beliefs” about future performance and standard measures of corporate innovation. … The model predicts that overconfident CEOs, who underestimate the probability of failure, are more likely to pursue innovation, and that this effect is larger in more competitive industries. We test these predictions on a panel of large publicly traded firms for the years 1980 to 1994. We find a robust positive association between overconfidence and citation-weighted patent counts in both cross-sectional and fixed-effect models. This effect is larger in more competitive industries. Our results suggest that overconfident CEOs are more likely to take their firms in a new technological direction. (more)

Added 8June: Postrel on John Nye:

Professor Nye argued that the wins and the losses probably don’t cancel out. Even the biggest winners don’t make enough money personally to cover the losses of all the individuals who went into businesses that failed. … The lucky-fools theory suggests, then, that Victorian Britain’s economy was successful but stagnant not because investors were irrationally afraid of risks but because they were all too mature and calculating. They didn’t tolerate the foolish chances that a vital economy requires.

Paul Davies, chair of the group that decides what SETI scientists will do if evidence of aliens is ever found, thinks similarly about science news: until scientists can say something to the public with great (~99%) confidence, they should say nothing.  (Quotes below.)  You see, frequent public updates on science issues of great popular interest, like evidence of aliens or asteroids headed toward Earth, would result in reporters bothering scientists at work with “mayhem”, disrupting their “lines of communication,” and disturbing their “dispassionate analysis.”  The fact that most early low-probability signs would end up being false alarms is “damaging to the credibility of science.”  So until scientists can confidently say that an asteroid will hit us or that we see aliens, they should just whisper to each other.

In the extreme case of receiving an actual alien message directed at us, Davies prefers scientists to kept quiet for the many years it would likely take to decode it fully.  And he prefers aliens to not send us any useful tech info, as then we would fight over who could decode it first.  How disruptive!

One might justify this confidence-or-silence policy by arguing either that non-scientists are biased to overreact to low confidence news, or that reporters are biased to present low probability news as if it were high probability, and non-scientists gullibly believe them.  I have not seen any systematic evidence presented in support of these claims, however.

Within academia, the bias against non-top articles seems like signaling.  Since folks confident they are great would not admit they’d ever done work that could not meet the highest standards, medium journal publications reveal a lack of confidence.  Similarly, I suspect signaling is behind the confidence-or-silence policy.  Since it is harder to credibly say something with great confidence than with low confidence, saying something with low confidence sends a bad signal about your abilities.  Keeping info secret is also a status move; info gives control and control marks status.

Quotes from Eerie Silence:

The fact that it may take days to be sure that a [SETI] signal is not manmade raises a very serious problem for managing the post-detection agenda. … Any hint of a positive result from a SETI project could immediately trigger media frenzy, and events might soon spiral out of control. … In the case of the SOHO satellite detection, the press got hold of the story even before the identification was made.  Fortunately the reporter concerned acted responsibly and waited for more data before rushing into print.  But not all members of the media can be relied upon to be so restrained. …

Because SETI astronomers are professional scientists, rigorous checking is an essential part of their training, and they want to be sure of their ground before making a definitive statement. History has shown that when scientists run to the press with sensational claims that haven’t been properly checked, the outcome is very damaging to the credibility of science itself, not to mention the reputations of the scientists involved.  A salutary lesson in how not to handle the media comes from the now largely discredited claim for cold nuclear fusion … They held a hasty press conference, and the media understandably had a field day. … It took many months for laboratories around the world to test the claim, and find it wanting. …

The lesson from that debacle is that it is wise to exercise restraint when dealing with the media about discoveries that carry sweeping implications for society. In the case of SETI, … once word got out, mayhem could ensue.  The astronomers might show up for work only to find their observatory besieged by journalists, … hardly an environment conducive to dispassionate analysis.  Even normal modes of communication are likely to be disrupted as lines become jammed by callers eager to check the rumors. …

It is in the nature of this type of investigation that false alarms greatly outnumber the real thing, so the above scenario might be played out many times, with the hullabaloo eventually subsiding as the story evaporates. A close analogy is the all too frequent announcement that civilization is menaced by an oncoming asteroid or comet. … When a new asteroid or comet seems to be moving on an Earth-crossing trajectory, … carefully checking takes time.  In the early days following the discovery, the projected orbits are uncertain because of normal measurement errors.  After the object has been followed for several days or weeks, the errors shrink enough that the astronomers can then work out whether it will or won’t hit Earth.  The most sensible strategy is to wait until the orbit has been properly determined, and only then, if there is still a clear and present danger, ‘wake the President’.  But … more often than not, the press get wind that a new object has been found that might strike our planet on the next orbital pass. It makes a wonderful scare story: ‘Killer asteroid may wipe out life as we know it!’  Headlines like that attract a lot of readers. … The known uncertainty in the measurements lets astronomers work out the probability of a collision – typically it is about one in 10,00 when the object is first identified. …

I personally believe the public does have a right to know, even if the news is bad – as soon as the situation is properly understood. I have yet to meet with a SETI scientist who doesn’t agree with this basic principle. There is no ‘code of secrecy’ in SETI. … Suppose the … discovery holds up at, say 99 per cent confidence level. …  If an astronomer were to spot something weird, which on closer inspection bore all the hallmarks of artificiality, then I believe it should be announced just like any other major astronomical discovery. … If an alien civilization were to send us a customized message, … I personally feel that the implications of simply  receiving such a message would be so startling and so disruptive that, all though eventual disclosure is essential, every effort should be made to delay a public announcement until a thorough evaluation of the content had been conducted, and the full consequences of  releasing the news carefully assessed in light of the Taskgroup’s recommendations. …

Information about the astronomical coordinates of the transmitter should be restricted to the astronomers involved … Even governments … no doubt would also want to take charge.  In my view, however, the less government involvement at the evaluation stage, the better. … Decoding it could take a very long time, perhaps involving years of meticulous work before we had any idea of what we were dealing with. … Most worring of all would be [a message] that merely handed us on a plate a revolutionary item of technology, e.g. a new source of energy, or a technique for engineering designier life forms reliably.  The problem here is that the gorup that possessed the knowledge first would be in a position of incomparable power. … Outright warfare might follow the scamble to grab the information. One can only hope that the aliens would recongnize the dangers and refrain from handing out scientific secretes like sweets.  [pp172-184]