Unlearning 101

See the Whole Picture

A wonderful 11-minute video from Scientific American on neuroscience and magic:

Unlearning & the Rise of the Supercomputer

Marshall McCluhan once famously said, “First, we shape our tools and then our tools shape us.” As a professional futurist, I have been trying to explain how some of our tools are getting exponentially more powerful. In turn, these tools are accelerating the advances that will soon shape our future and, by virtue of their findings, necessitate unlearning. Nowhere is this more true than in the field of supercomputers. 

According to this ComputerWorld article, supercomputers with 100 million cores and capable of one quintillion calculations per second—or 1,000,000,000,000,000,000 for those of you counting at home—will arrive around 2018. This growing processing power is important because it will help innovative researchers, scientists and entrepreneurs design next-generation materials, drugs, bio-fuels, batteries and much, much more. 

 One area merits special attention and that is in the area of how supercomputers are yielding new insights about the brain. As scientists learn more about how the brain really operates, I am convinced that we will need to unlearn much of what we now think we "know" about the brain. 

To better understand supercomputers amazing power, I invite you to watch this two-minute video I put together last year. Although be forewarned, one quadrillion is already rather quaint:

Do We Need to Unlearn Multi-Tasking?

A series of newspapers and Internet sites are reporting on new research, profiled in the latest issue of the Proceedings of the National Academy of Sciences, which suggests that multi-taskers are worse at multi-tasking.

Before jumping on the bandwagon and decrying the scourge of multi-tasking, I would encourage you to read this thoughtful post from one of my favorite unlearners Cathy Davidson.

Nevertheless, I believe the study is noteworthy for one reason in particular: low-multitaskers think they are worse at multitasking and high multitasker think they are gifted at it.

This is a classic example of why so many people could benefit from unlearning -- overconfidence. In particular, often people either think they know more than they actually do or, alternatively, they believe they are better at a certain task than they actually are. (If you don't believe me, read this old post on over-confidence or, better yet, try this little quiz.)

If you are a multi-tasker, I am not suggesting that you unlearn the skill -- only how proficient you believe you are at the skill. (Preferably, I'm hoping you'll unlearn your over-confidence before you slam into the side of my car at some intersection because you were texting a friend while at the same time checking out the latest Tweet from Shaq.)

Related Post

Have Some Intellectual Humility

Did You Get an "F" in Unlearning 101?

Read the following sentence:

FINISHED FILES ARE THE RE-
SULT OF YEARS OF SCIENTIF-
IC STUDY COMBINED WITH
THE EXPERIENCE OF YEARS.

Now count aloud the F's in that sentence. Important, count them ONLY ONCE: do not go back and count them again.

 

To find out the correct answer and learn if you passed the test, click on this link and move your cursor over the blue box.

Train Your Brain to Unlearn

Quick. I want you to say aloud what color you see in every word in the chart to the right. Do not say the word you read. (For example, sometimes the word "white" will be in the color red so you would say "red.") The quiz, better known as the Stoop task, is a classic psychology experiment and is used in neuropsychological evaluations to measure mental vitality and flexibility.

It is best done with another person observing you because you won't always catch your own mistakes. The reason this simple exercise is so difficult is because reading a word is an automated task and takes relatively little mental effort. Naming the color of the word, however, requires concentrated and deliberate thought.

In many ways the act of learning is similar to reading words -- it can become an automated process requiring little effort. For example, think of reading an article in a scientific journal or an opinion piece in the commentary section of your favorite newspaper. In many cases all you are doing is either adding new knowledge to your existing base of knowledge or reconfirming an existing viewpoint you have previously learned. (This isn't an inherently bad thing but if your base of knowledge is becoming obsolete or if your viewpoint is based on biased information it isn't helpful.)

Unlearning, like the naming of the color of the word, is more difficult. It requires deliberate thought and forces you to ignore the "obvious" answer. This act of ignoring the "obvious" answer can often be  the first step down the difficult, albeit rewarding, path of unlearning.

Interested in other unlearning exercises? Check out these past posts:

To Unlearn: Try a Blind Taste Test

To Unlearn: Meet Me in St. Louis

Train Your Mind to See Different Views

To Unlearn: Try a Blind Taste Test

Yesterday, I encouraged people interested in unlearning to "trust but verify" their instincts. One area where this technique might save you some money is to have a friend or loved one give you a blind taste of your favorite (and perhaps expensive) wine or coffee along with that of a lesser brand.

Interestingly, a few years ago researchers at Stanford gave a group of wine experts a blind taste test of five different wines. The only information provided was the price of each wine -- ranging between $5 and $90. Not surprisingly, the most expensive wine was identified as the best. What is interesting, though, is that three of the wines (the $5, $45, $90 bottles) were the same wine.

In other words, it was not the inherent characteristic of the wine (i.e. a"charismatic Cabernet Sauvignon with a full-bodied flavor and a hint of cherry and oak") that the tasters were reacting to, it was the price.

Now consider some of your favorite wines, coffees and chocolates. Are you really reacting to the superior taste -- or the inflated price? If you're unsure, take a blind taste test.  The results may very well encourage you to unlearn some of your expensive habits.

Do We Need to Unlearn the "Nature vs Nuture" Debate?

I understand the appeal of saying something is either black or white (and thus denying all the various shades of grey in between). Numerous debates occur everyday. For example, some people argue that  homosexuality is genetic in nature. Others claim it is a nurtured condition.

In spite of the efforts of supporters on both sides of the debate, the issue is not so clear cut. Neither are the hundreds of other such "nature vs nuture" debates.

In a new study being released today in Child Development Perspectives a team of researchers is calling for society to say "neither" to the debate. In short, they are saying that the world is far more complicated than many of us would like to acknowledge. But rather than gravitate toward "intellectually lazy" answers which tend to support one point of view over the other, or favoring "one-liners over 10-minute explanations," the authors suggest we should remain open to how various pieces of knowledge fit together.

In essense, the article and the study are really just a call for common sense but because common sense is a commodity which is often in short supply today -- even in academic and scientific circles -- it never hurts to stop and ask ourselves whether there is a new approach.

To this end, I'm in favor of unlearning the "nature vs nuture" debate and staying open to the idea that the real answer may be "neither."

Take a Course in Unlearning

At the end of yesterday's post on learning to unlearning, I posed the following question: How does one learn a new gestalt? To begin a person must start by unlearning some things. But what things do we unlearn? For our purposes, a good place to start might be to imagine what a course on unlearning what might look like.

 

One place to start is to imagine where the course would take place. Initially it will be—and already is—online. In 2007 the Massachusetts Institute of Technology announced that it was putting all of its courses online for free—for anyone in the world to access.

 

And let’s remember, the online, virtual classroom of the future is only going to get better. The Internet of the future will be streaming incredible amounts of data-rich information anywhere in the world, students will be capable of wirelessly downloading the latest information from flexible electronic books that display both the written word and video files, and new software programs will be capable of translating text from Mandarin Chinese, French, or Farsi into English—and vice versa. (See "The Future of Reading.")

 

Another place a course on unlearning might gravitate toward is 3-D virtual-reality environments such as Linden Labs’ Second Life—a site where anyone can create a personal avatar of himself, meet other virtual avatars, and engage in online training sessions. As of this writing IBM, Dell, Intel, Circuit City, and Sears have all created a virtual presence in Second Life.

 

Interestingly, one of the initial motives of this move was not to create a stronger presence on the Internet (although that is certainly a factor), it was to achieve cost savings on employee education.

 

What is more interesting from the perspective of unlearning is how Second Life and other virtual-reality sites can be exploited to provide people with different perspectives. In a virtual environment, people can take on any appearance they want. While some people will undoubtedly use it for escapist fantasies, it could also be a powerful tool to help people unlearn certain habits. Imagine, for example, customer service representatives or managers being required to act as customers in one of their own online stores. The experience could provide a unique and refreshing perspective. (And, as I recently wrote, we could all benefit on occasion from unlearning everything from our perspective of color to that person across the street.)

 

Longer term, the classroom of unlearning will likely become even more immersive. Perhaps Second Life will morph into Third Life. Among the technologies this environment are likely to incorporate will be enhanced visual, auditory, voice and speech recognition, and haptic technologies. Doctors and service technicians could use these tools to practice operations and repairs in silico before being allowed to ply their trade in the real world.

 

These tools will also be a boon for learning, unlearning, and relearning. People are often classified into one of three broad categories of learning: visual, auditory, or kinesthetic. Visual people learn by seeing or reading something, auditory learners by hearing it, and kinesthetic learners learn by doing it with their hands and muscles. (It is not quite this simple. Many people use a combination of different techniques for different things, but in general, most people tend to favor one of the three methods over the other two.)

 

A course on unlearning could exploit these natural tendencies and help people absorb new ways of doing things. For instance, instead of just reading about how a new drug works on a cancer patient, doctors could watch how it interacts with and disables a cancer cell. Other businesses could use such immersive technology to gain a deeper appreciation of what an elderly person experiences and create products that better address their needs. (See "Unlearning Your Age.")

 

Many courses on unlearning won’t have a teacher. They will rather be open source in nature, and the content will not be provided by a single “expert” but rather it will be continually added to and improved upon by a vast collection of people. To this end, a relatively new wiki called Curriki has recently been created. Its goal is to support the development and free distribution of world-class educational material to anyone who needs it—anywhere in the world.

 

But far from being a shoddy collection of disjointed or inferior ideas, the result of these wikis will be vastly superior to anything a single expert could pull together. In the case of business wikis, they will contain advice and insights from employees, suppliers, and customers.

 

Among the adjustments this will require is that managers will need to unlearn their own reliance on experts. People will need to unlearn the idea that money and quality are synonymous. In the future, many of the best products will be the creation of open-source methods and wikis.

 

Another thing people will have to unlearn is that there isn’t always an answer. This is because so many fields are constantly evolving. An admission of one’s own ignorance may well be the first step most people will need to take upon entering the unlearning classroom of the future. Exponential executives may even have to go a step farther and accept that ignorance will be the largest element in their future educational needs.

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. 

Is It Time to Unlearn the Scientific Method?

Last summer, Chris Anderson, the editor of Wired, wrote an excellent article entitled "The End of Theory: The Data Deluge Makes Scientific Method Obsolete” in which he convincingly argued that massive amounts of data, in combination with sophisticated algorithms and super powerful computers, offer 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 am open to the idea that he could be—this will take science in a whole new direction. In short, instead of modeling and waiting to find out if hypotheses are valid, the scientific community can instead rely on intelligent algorithms to do the heavy lifting.

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.

The implication for a field such as biology which, as Anderson points out is actually becoming more difficult to model as we 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, we will be able to analyze data without allowing hypotheses (which are, perhaps, wrong) to cloud our view of what the data is really showing us.