Deep or Shallow?

Guest Contribution by René Coronado

To a large degree, the purpose of learning is less to purely gain knowledge for the sake of it, and more to gain knowledge in order to use that knowledge to do something.

I propose that there are two basic types of learning: shallow and deep. Both types are useful, and both have positives and negatives.

---

 

Shallow learning is learning that comes by reading or watching instructions and following those instruction to the letter. Examples would include getting driving directions from your home to some place in town you haven’t been to yet, watching a youtube video on how to create a Skrillex styled wobble using Massive step by step, or building an IKEA bookshelf.

To learn something in this way you simply:

• define the thing that you want to do
• seek instructions for accomplishing that thing
• follow the instructions
• achieve the desired result

Much of the learning that we do on the internet is shallow learning by this definition. Again, I’ll state that shallow learning is not inherently inferior to deep learning, its just different.

The primary advantages of shallow learning are speed and diversity of things accomplished. Because you are simply finding and executing instructions, you’re bypassing the hard work of determining the best methodology for achieving the thing that you’re trying to do. This has you actually achieving results quickly, and because you reach the point of accomplishment so fast you are encouraged to try new and different things soon thereafter.

Shallow learning allows a person to explore the surface of any given discipline and to do simple things within that discipline. This is incredibly useful both if one only needs to do simple things (I have all kinds of graphic design shallow learning in my background, but you’d never hire me for it) or if one is just probing around a discipline to see if they want to get even deeper into it.

Shallow learning also leads to deep learning if it is repeated often enough. Building one Skrillex wobble via youtube tutorial won’t get you further down the road than just the one wobble. Building 30 wobbles, however, will start to ingrain the common elements in your mind allow you to start improvising around the edges of the technique and to start going “off book”.

A big disadvantage of shallow learning is that shallow knowledge does not allow one to adapt and to overcome obstacles that may arise when doing a given task. If the author of the instructions has to fork off too many if->then situations to deal with obstacles, the utility of the instructions for the reader decreases rapidly.

A second disadvantage of shallow learning is the inability of the person using shallow learning to create complex results.

The final and biggest disadvantage of shallow learning is that you cannot create new things this way. By definition shallow learning requires instructions, and as such that means someone has to have done it before you in order to create those instructions. The shallow waters are not the place where innovation happens.

---

 

Deep learning is learning that comes from doing a thing, evaluating the result, and then adjusting and doing that things again – improving the process repeatedly. Deep learning does not require instruction (though instruction certainly can help) but it does require lots and lots of time and patience.

To learn something in this way you must:

• define what it is that you want to do
• do it
• evaluate the outcome
• revise your approach and try again
• repeat until…
• you achieve the desired result

Examples of deep learning would include learning a language, learning a musical instrument, learning to paint, and learning a martial art.

The process of deep learning is the process of building a skill set that is efficient, flexible, and instinctive. Deep learning is iterative, with the student repeating the basic building blocks of the skill until they are ingrained, and then putting the building blocks together into a fluid and adaptable whole, then adding even more, like branches to a tree. This kind of learning often involves learning a language common to the discipline, practicing and critiquing one’s work, and employing creativity and ingenuity to overcome obstacles and problems.

The depth of learning one can pursue is almost limitless, with mastery of any one discipline coming many years after beginning the journey. Even mastery is not an endpoint for some.

The person who works to learn something deeply will have a very flexible and adaptable skill set that enables them to overcome complex obstacles. This is because a person with deep knowledge not only knows what they are doing, but also why they are doing it – giving them the ability to add to or subtract from the steps taken in doing the task. Deeper knowledge enables one to tackle greater complexity.

Deep knowledge also allows people to create new things, and also great things within their discipline.

The the biggest disadvantage to this type of learning is the time investment required. Because deep learning can take years, you can’t really have deep knowledge of dozens and dozens of subjects. You can’t do deep learning on the internet. The range of subjects is limited because of the time required for achieving deep knowledge.

Additionally, deep knowledge requires an increasing level of mental discipline and toughness as you go deeper and deeper and the problems become more complex and frustrating.

---

 

So why spell all of this out? Mainly to help clarify which methodology you should use when pursuing a given audio task.

To cast wide nets and do things quickly use the internet, ask your peers, and get instruction.

To do new or difficult things, lock yourself away, bear down, experiment, evaluate, and repeat.

The world is a vast ocean of knowledge. Wade around in the shallow waters until you find the things that compel you, then be prepared to go it alone and take a deep dive.

René Coronado is a partner at Dallas Audio Post and also publishes the Tonebenders Podcast. Find him @rene_coronado on twitter.