Unlearning 101

Don't Assume You More Than You Do

"... he has persuaded me that we can solve many of our problems as long as we don't assume that we know more than we do."

So concludes John Horgan in his insightful review of David Deutsch's new book, The Beginning of Infinity: Explanations That Will Transform the World.

I not only agree with the thinking behind quote, more importantly, I like that Mr. Horgan is willing to admit the author's book (and ideas) had changed his mind. (Apparently, Deutsch even criticized some of Horgan's own ideas in the book.)

As I've said before, I wish more people would expose themselves to unlearning and admit when they have been persuaded to see the world differently. Ironically, it would make the world a better place. 

Related Posts

Ignorance Isn't Bliss But It May Hold the Key to Wisdom

A Notebook of Things I Don't Know

Expose Yourself to Unlearning

Unlearning the Future

Carl Sagan once advised the public to “never underestimate the power of exponential.” It was sound advice. Unfortunately, as with so much other wise counsel, it is easy to dismiss.

You shouldn’t.

Consider just a few recent examples of exponential growth. In 1998, a small Silicon Valley start-up was conducting 25 million web-based searches a day. The company had developed an impressive technology but it hardly seemed transformative. Ten years of exponential growth later, though, and Google has changed everything from how people get their news and entertainment to how politics and health care are practiced.

Wikipedia offers yet another example of exponential growth. Who amongst us could have imagined in 2001 when the first 100 Wikipedia entries were placed on the Internet that within eight years the site--which relies on free labor no less--would become the most widely used information source in the world?

It would be easy to believe that Google and Wikipedia are unique and, as such, represent one-time anomalies. They are not.

Ironically, if you type the word “exponential” into Google, the first search item that comes up is an entry from Wikipedia. If one then clicks on the list of “exponential topics” at the bottom of the definition there is an odd looking word: zenzizenzizenzic.

The term is defined as a number to its eighth power. Thus, the zenzizenzizenzic of 2 is 2(8)--or 256.

Now, zenzizenzizenzic might seem like a rather innocent concept but recall that Google and Wikipedia both experienced zenzizenzizenzic-like growth and, in the process, transformed a number of industries; and, in the coming decade, so too will a number of other emerging technologies.

Consider, for example, that Intel Corporation recently announced it had successfully doubled--to 2 billion--the number of transistors it has crammed onto its next-generation computer chip.

The real world manifestation of Intel’s zenzizenzizenzic is that we can now purchase a computer over 200 times more powerful than the one we bought a 1997--at one-third the cost.

Moreover, because the semiconductor industry will likely continue its little trick of effectively doubling the number of transistors it can place on a computer chip every 18 months this implies that by 2020 the zenzizenzizenzic of today’s 2 billion transistor could morph into a 500 billion transistor computer chip. Try wrapping your mind around what such super-powerful computers might enable us to do and achieve.

Exponential growth is not simply limited to the computer and Internet industries. Data storage, bandwidth, gene sequencing and even the growth of synthetic diamonds are all expected to undergo zenzizenzizenzic progression in the coming decade.

In fact these fields are just the tip of the proverbial iceberg. According to Gartner Research, the field rapid prototype manufacturing (i.e. machines which can physically print objects) will grow more than 100-fold. And in the past year, researchers in Finland have replaced a patient’s lower jaw with synthetic material growth from stem cells; researchers at the University of Minnesota have successfully demonstrated that they could grow a beating rat heart from living cells; and self-driven robotic cars have successfully navigated through an urban environment. Progress in these fields will not stop.

It has been said that the future is here, it is just unevenly distributed. This is true and if one studies the amazing technological progress occurring in information technology, biotechnology, nanotechnology, neurotechnology and robotics and then applies the power of zenzizenzizenzic to these advances, it is easy to understand how it would behoove us all to reconsider Carl Sagan’s sage advice and begin preparing ourselves for a great deal of unlearning. 

Unlearning Disease

With all of the information now pulsing around us, it is easy to forget that "we don't know what we don't know." I constantly try to remind my clients in the health care sector of this simple fact. A case-in-point is the extraordinary strides being made in the field of genomics -- which, due to accelerating gene sequencing technologies, are occurring on a daily basis.

One the bigger implications of these cascading discoveries is that they will likely fundamentally alter how we classify disease in the future. As a little historical reference, it is useful to consider that in 1850 there were only 140 categories of disease. In 1993, the last time the World Health Organization undertook the classification of all diseases, there were some 12,000. The next classification isn't expected to occur until 2015, but it is not hard to imagine that the number of diseases -- and how they will be classified -- will increase even more.

For a great primer on this possibility, I'd encourage you to read this article, Redefining Disease, Genes and All. I especially liked this quote at the end of the article. One doctor, responding to the fact that as late as 1909 one of the leading causes of death was listed as "visitation of God," he went on to offer this warning: "Imagine how they are going to to be laughing at us," he said, "not 100 years from now, but even 50 or 20 years now."

As a result of accelerating technological change, it is certain that "we don't know what we don't know" about a wide variety of diseases. To stay open to the possibility that these diseases might soon be both classified and treated differently , it is helpful to keep in mind the possibility that if you don't change your views you will likely be widely mocked at some point in the future -- and it might even be sooner than you expect.

Pharmaceutical Industry Needs to Unlearn

In 2003, I wrote the book, ”The Next Big Thing is Really Small: How Nanotechnology Will Change the Future of Your Business.” In it, I outlined the many ways in which nanotechnology will impact the pharmaceutical industry in the near future. Recently, researchers at Children’s Hospital in Boston have uncovered a method of reviving a once-promising cancer drug, TNP-470—which was cancelled in the mid-1990’s due to concerns over toxicity.

The problem of toxicity is not a new one to the pharmaceutical industry. In fact, it is a big reason why so many early stage drugs fail to make it to market.

This vexing problem may soon be partially alleviated due to advances in nanotechnology. One advantage of reformulating drugs at the nanoscale is that they are water soluble and, as such, don’t require the solvents which so often cause nasty side-effects in patients.

Another application being employed from the nanotechnology bag of tricks is to use nanoscale platforms such as micelles and dendrimers to prevent drugs from crossing the blood-brain barrier. In this fashion, older drugs may be reformulated so as to prevent certain drugs from wreaking havoc on the human brain. (This was one of the problems with TNP-470).

The pharmaceutical industry has invested billions of dollars developing promising drugs. Unfortunately, most developmental drugs end being shelved when it is found they cause some unintended consequence. Nanotechnology holds the promise that a handful of older drugs may now be revisited. Some may even find their way to the commercial marketplace.

With a single drug’s ability to add billions of dollars to a pharmaceutical company’s bottom-line, it is time many of them begin turning their attention to nanotechnology. It’s a small technology, but it could pay big dividends. All the industry needs to do is unlearn the idea that just because something didn't work in the past, that doesn't mean that it is necessarily doomed to fail forever.

Related Post by Jack Uldrich

Does the Pharmaceutical Industry Also Need to Unlearn

Unlearning Models

Chris Anderson, the editor of Wired, has written an excellent article entitled "The End of Theory: The Data Deluge Makes Scientific Method Obsolete" in which he convincingly argues that massive amounts of data, in combination with sophisticated algorithms and super powerful computers, offers mankind a whole new way of understanding the world.

Anderson believes that our technological tools have now progressed to the point where the "old way" of doing science -- hypothesize, model and test -- is becoming obsolete. In its place, a new paradigm is now emerging whereby scientists, researchers and entrepreneurs simply allow statistical algorithms to find patterns where science cannot.

If Anderson is correct -- and I believe he very well could be -- this will take unlearning to a whole new level. In short, instead of having to do the unlearning ourselves (e.g. when our hypothesis doesn't work out or is proved wrong) scientists and researchers can instead rely on intelligent algorithms to do the heavy lifting.

The implication for a field such as biology which, as Anderson points out is actually becoming more difficult to model as learn more about it (due to our limited understanding of how genetics, microbes, personal behavior, the environment and a host of other factors work in partnership to determine a person's health), could be profound.

More specifically for the field of unlearning, this means we will be able to analyze data without allowing hypotheses (which may be wrong) to cloud our view of what the data is really showing us.

Before this vision can be achieved, however, it will require a great many brilliant scientists to unlearn the idea that their "model-based" method of trying to make sense of today's increasingly complex world is the only way to search for new meaning. Increasingly, we will be able to allow computers to crunch the data and extract the meaning for us.

Unlearn or Die Even More Unneccesarily

A few weeks back I had a posting  entitled "Change or Die ... Unnecessarily” in which I provided a few examples of how our inability to unlearn could, quite literally, cost some people their lives. Well, the other day, one of my favorite thinkers, Josh Wolfe, sent me a copy his weekly newsletter, The Forbes/Wolfe Weekly Insider. In it, he unveiled some startling research from a forthcoming book: apparently, six times as many people died in their cars from the 9/11 terrorist attacks as did the number of people who died in planes.

The authors don't state their research in quite this bold of a fashion; nevertheless this is one implication of their finding. On the face of it, the idea sounds absurd -- everyone knows only a few hundred people perished in the airplanes that crashed into the Twin Towers, the Pentagon, and in the rural field in Pennsylvania.

Essentially, it is the author's contention that because millions of people refused to fly after the 9/11 attacks -- and instead chose to drive (which is demonstrably more risky in statistical terms than flying) -- an estimated 1,595 additional people have since died as a result of the 9/11 attack than is generally recognized.

If these people could have unlearned that their fear of flying was an irrational act because it caused them to engage in an even riskier activity, it is likely they would be alive today.

On a somewhat related note, I recently received a copy of a powerpoint presentation from one of the world's leading epidemiologists, Michael Osterholm. In it, he coherently argues that the United States is woefully unprepared to deal with the very real possibility of a major pandemic.

Interestingly, in his slides, Osterholm makes the point that it possible that more people will die from non-influenza-related deaths than from the flu itself.

How so, you ask? Well, much of the pharmaceutical industry utilizes "just-in-time" inventory procedures. In the event of a pandemic; major shipping, trucking and distribution centers will be hampered and maybe even shut down entirely. As a result, hospitals -- which already have a dangerously low supply of many critical life-savings drugs -- will run out of important drugs in a matter of days.

The effect: people will die unnecessarily. That is, unless, we can unlearn the idea that "just-in-time" inventory for certain critical supplies is not necessarily a good idea.

Unlearn Genetic Nihilism

WebMD has interesting article suggesting our behavior can influence how are genes act. According to research by Dean Ornish people can change how certain genes perform by making modest adjustments to their life style. Ornish found that by engaging in exercise and eating a more low-fat diet "more than 500 genes changed the way they worked." More specifically, he found that some tumor-suppression genes became more active, while other genes with negative effects -- including some cancer-promoting genes -- were turned off when people engaged in a healthier lifestyle.

I especially liked this quote from Ornish, "People say, 'Oh, it is all in my genes, what can I do?' That's what I call genetic nihilism." Undoubtedly, many people do subscribe to the idea that our genes determine our fate. If, however, Ornish's findings prove to be true, it will dispel that idea and will require people to unlearn the idea that while "genes may be our predisposition, they are not are fate."

Get Prepared to Unlearn at Warp Speed

IBM recently announced that it has developed a new supercomputer capable of performing 1,000 trillion calculations per second. It is a little hard to wrap your brain around such mind-boggling numbers, but last year I wrote an article discussing an IBM supercomputer that was capable of performing 70 trillion calculations per second. In the piece, I noted that if you had to perform a similar number of calculations (and assuming you could work 365/7/24) it would take you roughly 60 million years to perform what that supercomputer could do in one second.

Well, with this latest advance, it would now take you -- theoretically speaking -- about 800 million years. Of course, such a comparison is slightly ridiculous. What is not ridiculous, however, is what these supercomputers are doing and learning. With their immense electronic brains they are now designing new materials and products; creating new drugs; and helping us better understand everything from the human body and brain to how mankind is impacting the environment.

One implication of these new findings is that all of us will need to unlearn at a faster rate.

Do Lawyers Prevent Unlearning?

The Mayo Clinic is now operating a pilot project to offer "virtual consults" to patients living far away. The idea that a doctor must be physically present when conferring with a patient is going to be difficult for many doctors and health care providers to "unlearn," but the practice can save time, money and, ultimately, even lives.

Unfortunately, lawyers could throw a serious wrench into this promising new practice. According to Thomas Malone, a lawyer who heads a national group of malpractice attorneys, the practice "creates a wide opportunity for malpractice." His main argument seems to be that virtual consults will result in doctors spending less time with the patient.

This is faulty thinking for a couple of reasons. First, if a virtual consult is the only way a patient might see a doctor, I'd argue that "something is better than nothing." Second, many diseases can just as easily be diagnosed virtually as they can be in person. Specifically, this is true of internal injuries. Finally, the lawyer is unlikely taking into consideration how advances in digital technology are continuing to improve the quality and speed with which doctors can view and receive information. In short, telemedicine technology is only going to get better with time.

It would be a pity if lawyers inability to unlearn thwarts or impedes this promising new practice. It would even more tragic (and ironic) if in their zealousness to "protect" patients they actually ended up costing some people their lives.