We’re now in an age where AI can answer almost any question instantly, altering not only the way we access information, but how we think and learn. Whether it’s a boon to our self-education or a detriment, could be up to how we use it.

There have been many studies on how people learn, and theories on the best methods for teaching. A largely unanimous opinion has formed in favour of learning-by-doing.

Direct experience, with full engagement in a task and quick feedback upon failures and successes, has shown us again and again how effective it is at remodelling and refining the mind.

This approach is often compared to rote learning, which heavily relies on repetition and memorization, but lacks any focus on deep understanding or critical thinking.

The question we face is where will AI fit into our learning processes—will it encourage rote learning? Or can it enhance the learning-by-doing approach?

Kolb’s Experiential Learning Cycle

One of the most prominent theories to build on the learn-by-doing approach is Kolb’s experiential learning cycle.

David A. Kolb is a psychologist and educational theorist who focuses on experiential learning (another way of saying ‘learning by doing’). He helped found and currently chairs Experience Based Learning Systems, and is a professor of organizational behavior at Case Western Reserve University in Cleveland, Ohio.

Experiential learning is a framework for learning by doing, describing how people learn through experience. Kolb describes it as a continuous four-part process: concrete experience, reflective observation, abstract conceptualization and active experimentation.

Concrete Experience: direct participation in an activity, it’s not enough to just read or watch. For example, conducting an experiment, riding a bike, or designing a website.
Reflective Observation: step back and reflect on the experience, connect feelings and ideas about what happened, think about what you learned, and ask questions—what worked and what didn’t?
Abstract Conceptualization: At this stage, you start looking for patterns, theories, and principles that explain your experience while also providing something you can test in the next stage. If the prior stage was where you thought about what worked, here you start asking why.
Active Experimentation: Test your new ideas out, get feedback and create the next experience that serves as the basis for the cycle to repeat.

This cycle keeps learning dynamic, immersive, and application-driven, making knowledge more useful beyond just recalling facts. The task will be to find an appropriate space for AI within this cycle.

The Contrast With Rote Learning

Despite the overwhelming effectiveness of learning by doing, you’ll still find many examples of rote learning throughout schools, courses, and people’s lay theories on how to learn.

Part of the reason is that rote learning is faster to implement and easier to test and measure. If you scroll through a multiple-choice quiz, you’re likely being tested on how well you’ve memorized facts. Conceptual knowledge is much more difficult to evaluate.

This is not inherently bad—there are often facts, terms, dates, names and so forth that we need to remember efficiently. Repetition can be helpful in this regard.

However, it’s only a superficial form of learning, you don’t gain a deep understanding of complex ideas and concepts, it doesn’t teach you how to solve thorny problems. It’s difficult to even apply it to the more physical and hands-on skills that Kolb’s experiential learning cycle excels at.

AI can influence both approaches to learning. It can be a quick and easy route to answers and information, but it can also foster understanding through a dynamic back-and-forth conversation.

Learning by Doing with AI

We’ve spent enough time with search engines and apps to understand the value of information on demand. AI language models like ChatGPT take this a step further, giving you detailed and tailored answers to various queries.

It’s essential to note that care still needs to be taken given the ability of AI to confabulate some facts—you should always check the sources. Yet there’s still great value in having a particularly well-informed chatbot to converse with on demand, if used effectively.

Let’s explore how we can apply this type of AI to the stages of Kolb’s experiential learning cycle:

Concrete Learning

Here you need direct experience with the activity, to get a feel for it. This might be a physical, hands-on activity like playing the guitar or learning to cook, but it can also be more conceptual or theoretical, such as learning a language or studying law or mathematics.

Whenever you start out learning something new, it’s helpful to know where to begin. AI can help in this regard, it can take into consideration what level you’re currently at, what prior knowledge you have, and often provide real-time support.

For example, if you’re learning a language AI can simulate a conversation, correcting your responses along the way. It could generate math problems or legal cases for you to analyse, or suggest simple guitar songs or recipes for you to try, along with some basic instructions.

This has to go beyond passive reading to avoid the rote learning approach. AI can set you a challenge, and support you with feedback, but it must be you that tackles it.

Reflective Observation

The next step is to review what happened, to think about what worked and what didn’t. You should take some time to think this through yourself, consider how it felt to go through the process and reflect on what you did well and what you struggled with.

Beyond that, AI can help in a few ways. In areas like language learning, math, and other conceptual areas where you were solving problems in an interaction with the AI, it will be able to point out the errors and areas requiring improvement.

Activities like cooking or guitar playing will require you to provide some feedback to AI—tell it what you found difficult, or describe the result, and get AI to consider what the causes and issues were.

AI can help identify the problem areas, collect them over the course of an interaction, and summarise them to give you a detailed breakdown of how you performed. This gives you greater information to draw conclusions from and reflect on.

Abstract Conceptualisation

Here you’re looking for patterns and theories that will provide the predictions you experiment with in the next phase. Most domains of learning have overarching concepts that explain more than any isolated facts can—one of the reasons rote learning can only take you so far.

When you’re just starting out, chances are you won’t know many of the founding ideas and principles, but AI can help with that. Whereas in the previous stage AI could help by identifying errors, it’s also possible for it to consider the broader picture.

You might get the relevant rules of grammar in another language, music theory or picking techniques for guitar, the theory behind math equations or concepts within the natural sciences, and chemical processes that explain why your bread didn’t rise.

Knowing the particular strengths and struggles you have should help AI point you towards the explanations and ideas that are going to be most helpful to you.

Active Experimentation

Now you’re back at the active part. Having worked your way through the reflection and analysis of your past performance, you should be ready to start experimenting and further refining your knowledge and abilities.

Here you might use AI to recreate tests and challenges, like in the first stage, but in a way that focuses on the areas where you need to improve.

If you struggled more with picking than strumming the guitar, the song selection will change to reflect that. If you had trouble with specific verb conjugations in Spanish, your next conversation should work on that.

Again, this section can be quite dynamic, the idea is to experiment with different approaches, and that is one area AI thrives in—ask it questions, discuss your goals and abilities, adjust how you want feedback provided or tests designed.

Modern Education

AI hints at a world in which we all have a personal tutor at our fingertips, ready to answer questions, engage in deep conversations, and help us solve difficult problems. While there are still some kinks to iron out (check the sources!), the potential is clear.

This is a powerful tool that can upgrade your self-education, but it requires knowing how to use it effectively. For years we’ve known that experiential learning is the way to go, but rote learning is simple and easy, so we turn to that almost by default, this is a mistake.

Getting answers online has never been easier, but learning and mastering new skills requires hard work. There is no easy solution. AI can help, but if you want to remodel your brain, you’ll need AI to challenge you.

Author
Recent Posts

Kyle Pearce

I'm a serial entrepreneur, autodidact and a passionate advocate for community-driven experiential learning.

I built DIY Genius to help entrepreneurial people find their sweet spot for deep focus and flow states where passion meets purpose and profitability.

I believe the future of education starts with helping people find their "zone of genius" so they can develop their uniquely human ability to create meaning, connection and community (that A.I. can't replicate).

Learning By Doing With Kolb’s Experiential Learning Cycle

Kolb’s Experiential Learning Cycle

The Contrast With Rote Learning