Unlearning 101

New Voice Recognition Technology Mandates Unlearning

Late yesterday, Google released a very cool new mobile application which employs voice recognition technology. The question is not so much what the technology can do today, the question is what will the technology be able to do in the near future—and how might it require professionals in the education, health care, and a host of other businesses to unlearn?

I’d love to hear your thoughts on how this technology could unfold and how it might alter people’s behavior. Below are my initial thoughts from my other website: www.jumpthecurve.net:

Unlearning Bubbles

The Internet, the real estate market, fuel cell technology, nanotechnology and, now, "clean tech" have all experienced an investing "bubble." In retrospect, it is easy to point to signs that the market was inflating to dangerous levels. It is even easier to suggest that such bubbles are a bad thing -- precisely because so many investors got burned and lost a great a deal of money.

Bob Metcalfe, an early and successful Internet entrepreneur, see bubbles differently. He sees bubbles as a good thing. Here's his quote: "From Internet history, we know that bubbles are normal. Bubbles are an accelerator of technology progress. Bubbles go against the status quo. We should encourage bubbles."

I suspect that such statements are much easier to make when you're already a billionaire; as opposed, say, to someone to just lost a large amount of money in the stock market. Nevertheless, there is a certain logic to his argument and we may want not to dismiss all bubbles as cases of irrational investors getting sucked up by the winds of "irrational exuberance." Instead we want to view them as "technology accelerators." (This might not ease the financial pain of those who lost money in the bubble, but it might at least make them feel a little bit better about themselves.)

Unlearning Choice

It is easy to believe that having more choices will lead to a greater level of happiness. However, as this new study suggests, people may be more satisfied when selecting from a smaller set of options.

There are a couple of reasons for this somewhat counter-intuitive finding, but the most poignant seems to be that when people select from a wider array of choices they often experience a greater sense of regret (after their selection) because they aren't as confident that they made the optimal choice.

Trying to convince people that having fewer choices is a good idea will be difficult endeavor, but smart companies may be able to enhance their customers feeling of satisfaction by presenting information in new and innovative methods. For instance, think of on-line dating companies. Instead of presenting customers with an ever widening array of potential partners, they might instead employ sophisticated algorithms to narrow the list of prospective mates from which a person selects their choice.

Unlearning Learning

When contemplating the future, people need to keep a very open mind about what might be possible. Consider this article which describes how researchers at UC San Diego are developing facial recognition technology that can recognize if a person is having trouble understanding an educational lesson—say in mathematics or biology.

As the technology continues to improve, one possible implication is that smart devices and robots will become better and more effective teachers because they will be able to pace lesson plans to an individual student’s ability to comprehend the information which is being presented.

Longer term, it is possible that robots and other smart devices will become more effective teachers than even human teachers because the machines will understand each student’s learning idiosyncrasies and then present material in a manner which is optimized for that individual student’s learning style.

Now, I understand how discomforting the idea that a robot might be a better teacher than your old favorite third grade teacher, Mrs. Hubbard, ever was; but, as that wise American philosopher Yoggi Berra once said, “The future ain’t what it used to be.”

For a more contemporary look at how technology is transforming education, I’d invite you to read this article which explains how Nintendo DS is helping Japanese students learn the English language, or this article which explains how video games are actually helping students learn such topics as history and world affairs.

In each case, advances in technology may result in parents, teachers and administrators having to unlearn some long-held assumptions -- such as the idea that teachers are the only ones who can teach, or that video games impede students ability to learn.

P.S. For opposing view on why robots might never make good teachers, I'd invite you to read this review of a recent article which appeared in the New Yorker.

Unlearning Computer Programming

One of the reasons I have established this blog on unlearning on Typepad is because it allows me to maintain and modifiy the site easily without knowing a lick of computer programming or software development. And, trust me, I don't know jack about those fields.

Nevertheless, I recently stumbled upon this old article, The Joy of Unlearning, which was written by Andy Hunt -- who is knowledgable about software. The article encourages programmers to embrace unlearning. I especially liked this quote:

Old habits are hard to break, and even harder to notice. The first step to unlearning--and the hardest part--is to realize that you're using an outdated approach. The next hardest part is actually letting go. Mental models and patterns of thought are built and refined at great cost, and one doesn't discard them lightly. It takes awareness, conscious effort, and hard work.

It's great advice and it doesn't just apply to programmers. It applies to all of us.

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.