Wednesday, February 20, 2008

Genius

I just started reading Genius: The Life and Science of Richard Feynman By James Gleick.



This is just a fun book. It reminds me so much of when I was kid and didn't yet know that there were "limits" to what I could learn and discover.

I think I'm going to have new little wrist bands made that say, "WWFD?" Yep, that's right: "What Would Feynman Do?"

Tuesday, February 19, 2008

Five Models of Business

Some time ago, a friend of mine suggested that all businesses could be boiled down to 5 types associated with certain illicit activities:

  1. drugs - processing raw material into a refined, usable form

  2. pornography - content created once and distributed through low margin media

  3. prostitution - hourly labor

  4. pimping - providing access to aggregated valued services or products

  5. religion - selling promises about something the purveyor cannot actually deliver


It puts a new spin on business, doesn't it?

I wonder, though, if this "catalog" is complete. Are there other categories that could be added? What about moving things from one place to another? Waste disposal?

Monday, February 04, 2008

Definitions of words to be employed for the study of stupidity

There are a lot of books out there that help one attempt to be successful in a number of endeavors. But I think understanding how not to succeed may be as important as understanding how to succeed. Because stupidity sometimes seems to be a contributing factor to failure (although success sometimes seems to occur in spite of it), I propose that we invent a science of stupidity. To that end, I begin here a lexicon of useful terms. I will update this list from time to time as I learn more.
constigration: (constipation + confligration) failure to allow to pass through one's mind (recognize or admit) the contribution of one's own stupidity to the disastrous consequences that are unfolding around one in a consuming inferno.

epistemiology: (epistemology + epidemiology) the study of the population mechanisms by which stupidity propagates.

Godwin's inversion: The probability that once an analogous reference to Hitler or Nazis is made in an argument, the remaining comments by all contributors will be stupid.

jedi mind trick: the mistakenly optimistic belief that one's stupidity won't be observed in public.

ridiculocity: (ridiculous + velocity) the rate and direction at which stupidity propagates.

spidey sense: the mistakenly optimisitic belief that one's own level of good sense and special insight are always good enough.

If you blink, you just might get it

I entered the office today feeling rather blue, after all, rainy days and Mondays always get me down. But today was a little different. I was maybe feeling a tad bit more deflated than is typical for a rainy Monday morning.

"You look depressed," observed George P. Burdell, a man known for his instinctive awareness. "What's got you down?"

"Well, I just read a book review of by Malcolm Gladwell, and I think we may be out of business. I mean, who's going to call us experts when all people have to do is simply trust their immediate intuition, you know, blink and not think?"

"Hmmm. You may have misunderstood Gladwell's book. My observation is that a number of reviewers are doing this, and the one you read this morning may be among that group that has grasped incorrectly on the message of Blink."

Intrigued now that all may not be lost for us, I lifted my left eyebrow and cocked my head to Burdell. "Please continue," I implored with an unblinking stare.

"Well, let me begin with a comment that might be more like a footnote. If you really want to dig into the science of decision making as dealt with by Gladwell, I recommend the precursor work that Gladwell mentioned in the early pages (2007 paperback version, pp 8-11) of Blink, that is, the work by Gerd Gigerenzer published in his book, . Don't be deceived by the concept of 'simple.' It means neither 'simplistic' nor 'uneducated.' That is a key to understanding Blink."

George continued on for a while, and I studiously took notes. The following is what I took away.

Blink addresses two large themes: the power of making snap decisions, and the capability to achieve this power through "thin slicing." Gladwell weaves compelling anecdotes from such diverse sources as family counseling, art history, trauma care, food tasting, police work, and war to demonstrate how people often make accurate judgments and consequential decisions within a very short time frame through a mental process that appears to slice information from the environment into sparse, manageable chunks. If you have not yet read the book and have only read reviews, you are probably most familiar with this high level synopsis. Furthermore, you may have also received the implication from these reviews, as I seem to, that this snap decision making process is more powerful than thinking through problems systematically and critically. If this is the case, I think it is a mistake on the reviewers' part and fails to bring the reader to an appropriate understanding of Gladwell's message in Blink.

It is George P. Burdell's opinion that Blink DOES NOT actually promote the idea that snap decisions are always the best decisions. To this point, Gladwell spends a great amount of time discussing how snap decisions can go terribly awry. There exists, in fact, a large body of evidence that suggests that gut-level decision making (the unfortunately popular way of thinking about snap decisions) alone is fraught with failure and difficulties. A large part of this is due to the internal biases we develop over time, both cognitive and motivational. The representative work of Amos Tversky and Daniel Khaneman (Memorial Nobel Prize winning work) demonstrate this quite effectively. Other recent books, and (each by Nassim Taleb), eloquently discuss the effects of biases on investors who often make snap decisions as well as well thought out decisions. It is very difficult to overcome the effects of our biases, especially those related to the species of expert overconfidence. The worst part is that they are most pernicious when we think they are not present at all, that we have overcome them with sheer reason and special gifted insight. Gladwell actually reveals this point on pg 233: ” ‘When we make split-second decisions,’ Payne says, ‘We are really vulnerable to being guided by our stereotypes and prejudices, even ones we may not necessarily endorse or believe.’ ” And on pg 252: “…we are often careless with our powers of rapid cognition. We don’t know where our first impressions come from or precisely what they mean, so we don’t always appreciate their fragility. Taking our powers of rapid cognitions seriously means we have to acknowledge the subtle influences that can alter or undermine or bias the products of our unconscious.” Overcoming these influencing biases, though, requires a certain degree of informed effort.

“Thin slicing,” the idea that sparse information helps experts focus on what is important rather than become distracted by informational spam, is another idea promoted within Blink that I have heard mangled by readers. I recently heard one person insist that as little information as possible was always preferable. O course, one can see that if that idea is taken to the limit of zero information or what we might call “negative” information (i.e., lies), decision making becomes an act of random behavior. While this might work for the species, for individuals it can be disastrous. It is true (i.e., confirmed by repeated experiments and mathematics) that a little information is oftentimes much better than more information. For example, there is an idea in decision science called value of information (VOI) that deals, in part, with just this issue. VOI is the rational upper bound one should be willing to pay for additional information on an uncertainty that could cause one to experience regret on making a given decision. Unfortunately, most people do not know how to evaluate VOI, and so they spend increasing levels of resources trying to get more and more information on issues that would not have a great likelihood of causing them to change their decision anyway. The effect is to get mired into analysis paralysis such that no decisions get made in a timely and economically efficient manner. VOI tells a decision maker how to limit the search for information. The point is that more and more information is not the answer to achieve better outcomes, but neither is less and less information. Each extreme leads to diminishing returns. Rather, the point behind thin slicing is that there is an optimal amount of information required by the neural structures of an expert decision maker that regresses data from the environment into inferences on which the decision maker acts (Gerd Gigerenzer describes these simple efficient heuristics as fast and frugal). This thin slicing is not related to novices facing novel situations, but ones in which an “expert” has numerous of hours of experience.

And this gets us to the point of Blink. Blink is not about gut decision making by lay practitioners or novices. Rather it is about how properly unbiased experts have developed the ability to make judgments at a near instinctual level. Experts in this case are people who have spent countless hours studying, thinking, and practicing in their particular area of concern. A normal Joe off the street CANNOT use snap judgment to determine the longevity of a marriage after spending only a few minutes with a couple, nor can he determine the veracity of the claims of authenticity of ancient artifacts unless, perchance, he has been given the clues that experts have learned through intense experience. “Blink”-thinking doesn’t work for just anyone on a moment's notice. It does, however, apply to people who spend obsessive amounts of time immersing themselves in a field of concern, who, in crucial moments when they are called upon to use their expertise, respond instinctively without wasting time using their executive cognitive processes.

Unfortunately, many people who have reviewed Blink seem to miss this point. They focus instead on the idea that expertise doesn’t matter. Maybe realizing this, Gladwell wrote an Afterword to Blink (2007) this shortcoming. He sums it all up on pg 260: “...when it comes to fast-moving, high-stakes situations like battlefields (or emergency rooms, or auditions, or late-night shoot outs in the Bronx), …formal conventional analysis doesn’t help that much. Chancellorsville came down to some ineffable, magical decision-making ability that Lee possessed and Hooker did not. What was that magical thing?…It’s the kind of wisdom that someone acquires after a lifetime of learning and watching and doing.”

But he doesn’t throw the baby out with the bath water, either: “I think the task of figuring out how to combine the best of CONSCIOUS DELIBERATION (emphasis added) and instinctive judgment is one of the greatest challenges of our time.” (pg 269)

It seems, then, that Gladwell is pointing to the route for decision makers that all good artists know they must follow to get to Carnegie Hall: practice, practice, practice. ...And be humble enough to know that bias is a terrible and frequently uninvited companion on that route.