Pattern Recognition: At The Heart Of Entrepreneurship, Learning, And Puzzles

Posted on March 13, 2012


The main thesis behind Thinking Fast And Slow is that we utilize two systems of thinking. The first, System 1, requires very little energy and allows us to survive in a very complex world. At the heart of System 1 is pattern recognition that allows us to to instantly recognize a face, detect danger (a ball flying at our face), etc, in a way that happens without us consciously thinking. Pattern recognition is one of our greatest skills and it requires very little energy to do. It is also one of the things that we are much better at doing than computers. Some of the most difficult algorithms to write involve facial recognition and natural language recognition, activities that we do without expending any effort at all.

Pattern recognition is fundamental to creativity and empathy, two skills that are essential to entrepreneurship. Every successful entrepreneur should really be called an expert pattern recognizer. They recognized a pattern within their environment that could be affected by a product or service. They knew if they built that product or service hundreds if not millions of people would want to use it. In designing the product or service they were able to visualize the usage patterns so that they could build the most useful, attractive, and intuitive product possible.They were able to look at the world around them and recognize further patterns that informed their decisions on pricing and distribution so that the product or service made sense to their customers within the larger scheme of things.

Pattern recognition leads to theories or strategies that the entrepreneur can pursue. Not all of them will be successful. Pattern recognition at it’s best is an imperfect system frequently leading to failure, but the failure is beneficial in that it informs the pattern even more. Pattern recognition allows us to act on imperfect information and change course, iterating frequently, as we learn more from our successful and unsuccessful strategies, eventually leading to more nuanced and accurate strategies.

Pattern recognition is also essential to learning (or at least it should be). Organic learning that babies engage in is completely based on pattern recognition. They don’t understand the nuances of the language they are learning, they don’t understand that some words are nouns and some words are verbs, they simply understand that if they say one thing it will illicit a certain response. They learned this through trial and error based on the patterns they see around them during their daily lives. Much like the entrepreneur, babies are constantly evaluating the world for patterns that allow them to make sense of all of the complexity. As more and more patterns are established they are able to build on them to form more nuanced strategies until they are eventually able to understand that they are speaking words that can be mixed and matched to provide even greater communicative possibilities.

More learning should take this form. It is extremely enjoyable and challenging to try and figure out patterns that can inform strategies that can lead to greater understanding and capability. As we try and figure out the best ways to teach new material we should be trying to create environments that allow a new learner to easily recognize patterns, create strategies on their own, execute a given strategy and then iterate based on the learning that has taken place.

Unfortunately most formal education takes the form of a cookbook. You follow a set of instructions that lead to the right answer or you simply memorize the answer itself. There’s no pattern recognition involved, no opportunity to develop a strategy by yourself, no room for trial and error or iterations that lead to incrementally better understanding.

This is why I think we should strive to transform the way we teach in to puzzles. Instead of providing the most direct path to knowledge (linear learning), we should provide an environment that facilitates pattern recognition. To me, puzzles are the most self-contained representation of this strategy. When you approach a puzzle of any sort you are looking for patterns that will allow you to develop strategies for solving the puzzle quickly. With the jigsaw puzzle you might recognize that there are fewer edge pieces than center pieces and that each edge has an additional piece of information (the edge) that reduces the number of places where that piece can go. That’s a pattern that leads to the strategy of starting with edge pieces and building out from there.

The more we can figure out how to set up simple, self-contained puzzles and puzzle-like activities and environments, that are filled with recognizable patterns that can inform strategies for gaining a better understanding of new material, the more likely we are to engage new learners in an intrinsically motivating activity that will lead to a greater understanding of new material with very little outside help.

My efforts with Peanutty are one example of this. I’m trying to create an environment where people can learn to program through pattern recognition. They can start with simple patterns (if I change this number then something happens) and eventually progress to larger patterns (if I use a “for” loop I can make something happen over and over again, changing it slightly every time), which eventually leads to a very practical understanding of how to program that allows them to build anything from a Peanutty level to a website to a game to a search engine. The whole idea is to create an environment where a new learner wants to engage and, as they engage, they begin to recognize more and more patterns inherent to the environment that inform how they should engage with the environment in more complex ways (strategies), eventually leading to a very nuanced understanding of new material (e.g. programming).

I think we can create puzzles and puzzle-like activities and environments for every subject, but I think it will be very challenging. Already schools like Monstessori use puzzles to teach a wide variety of subjects and the students engage with them for hours at a time, gaining a very practical understanding of complex material.

Some subjects like learning a foreign language are already puzzles if treated properly (babies looking for patterns in the language they hear around them and what those language patterns enable them to do), but instead we teach them through rote memorization and understanding the grammatical structures. I think that’s why it’s so much easier to go to a country where the language is spoken than it is to learn the language in the classroom.

So I’m dedicating a lot of my time to figuring out how to create puzzles that engage our pattern recognition abilities with the goal of learning new material. Hopefully these efforts will eventually be used in The Puzzle School, but I’m still a ways away from that.