“If content is king, then context is god!.” – Gary Vaynerchuk

This statement is fairly common, but it may be hard to understand why. The origin of the statement comes from marketing. The old adage that marketing is content has been undermined by more tailored approaches, understanding the audience and providing specifically appropriate context. This takes a different twist for learning.

Among the findings in learning is that learning is better when done in context. You may have heard that you should study in the room you’re going to be tested in. In short, if you learn in a context like the one you have to perform in, your performance will be better. If you want people to be able to do new things, you have to put them in a situation to perform the new thing, and give them a chance to practice. That’s what a designed learning experience is. But there is a lot going on here.

Do Abstracts Work?

Abstract learning doesn’t transfer to other situations. You might think that learning in the abstract gives you the flexibility to transfer to other situations, but it doesn’t work that way. What you get is the ability to perform on abstract problems, but not to apply it in particular contexts. It just doesn’t get activated. You can’t get good performance from abstract problems. (See: too much of what happens in school.)

So, if you need transfer across contexts, you need to learn in a variety of contexts that span the space of potential application. You don’t need to learn in all the contexts, but in representative ones that cover the space. So, for instance, if we want someone to learn to negotiate, we might practice it in a compensation discussion, a vendor negotiation, and a project role. You thereby increase the likelihood that the learning will be accessed and used in an appropriate situation.

How Does the Brain Work With Context?

What underpins this is that our brains abstract across the contexts to decouple the things that can change from the things that define that this is a situation for the topic. So the negotiation basics of a goal, and an entity you have to convince to provide what you need to achieve your goal, don’t vary, but the goals and the entities and constraints will. There are elements that signal that this is a situation that calls for the particular ability being developed, and others that can change. If we choose the right contexts, the learner’s brain will learn to recognize the trigger conditions for the ability, and the elements that can change without affecting the requirement to execute.

This also includes situations that suggest how to adapt the skill to different situations where it’s still relevant. Selling cars is different than selling computers, but some elements stay the same and some differ. If you want generic sales skills, you have to train across products and services. If you want car-selling skills, you just have to train across different cars, perhaps minivans, sports cars, sport-utility vehicles, etc.

How Do We Choose Appropriate Context for Learning?

This means we have to carefully choose our learning contexts. Ideally, we have existing situations we can leverage. Or we can anticipate the appropriate situations. And, again, we have to choose ones that are representative and will support decoupling the necessary components and making a robustly transferable skill. (And contextualized practice is what makes a difference in ability to do, not the ability to recite information!)

One of the things we do in learning is create contexts to practice in. We can run role-plays, or scenarios, or even serious games and simulations. The point is to minimize the difference between the learning event and the performance environment. And, the more that is at risk for performance, the closer we need to go. So, in medicine and aviation for example, when lives are on the line, there are a lot of simulations and mentored practice.

And developing the context doesn’t have to be completely rigorous. While simulations and virtual worlds can create really deep immersion, the minimum necessary contextualization is often a better idea to both provide support in abstracting and transferring to other situations. (It also works to be more cost-effective too.) We know that extraneous content can interfere cognitively, so working on the elements that will convey a context and the triggers for the action are more important than a full rendition.

How Can We Leverage the Context of IRL?

One of the opportunities we are increasingly seeing, however, is not creating context, but turning real performance contexts into learning opportunities. Learning in the work is becoming possible. We can detect and understand where the learner is, and provide support. So, either in a particular physical location (say a library or an office), or at a particular place in a piece of software, we can have a learning challenge.

The advantages to contextualized learning are several. First, if this is the real context, we are minimizing transfer distance. We can mimic a real situation in a context it would actually emerge in. As a consequence, we also do not need to provide as much content to convey a particular situation. And we do this naturally in developing on the job training and mentoring. But we can take it further.

Increasingly, our software can be aware of our situation. In a software program, it can know where a learner is. If that’s coupled with what a learner knows, the learning can be personalized to the individual. Even more generically, however, we can provide a sample task to perform as a refresher.

Don’t Confuse Performance Support and Learning

Be aware, however, that there is a difference between performance support and learning here. We have systems that can know what you’re trying to do, context-sensitive performance support, that can provide hints and tips about what’s required in this situation. Similarly, we can just access a ‘how to’ video about this particular tasks. In both cases, learning isn’t the desired outcome, the goal is to get the job done. This is very valuable, but it’s not necessarily learning. A separate situation would be required to layer on some additional information about how and why this is the right solution before it could be considered learning.

This brings us back to marketing: the right content to the right person (and more: at the right time and right place on the right devices…). What is required are content models to get more granular, and content engineering to deliver it in systemic ways. And these capabilities are now available, and it’s time for learning to catch up with marketing and start treating content as a discipline.

Going forward, mobile is going to be supporting both contextual learning and performance support. Increasingly, the sensors provided with these devices can assist systems to detect a user’s situation in more than one way (e.g. not just location), and provide specific help because of where and when you are. And, coupled with an understanding of your learning goals, they may well be able to make contextualized learning available as well.

While content is critical to support learning even in contexts, being aware and leveraging context can mean new opportunities to improve the learning, and performance, outcomes.