By Zoe Enser

In our lessons, “learning” doesn’t always occur. So, why are some lesson activities more effective than others in enabling deep understanding and long-term retention of knowledge? Zoe Enser has an answer…

A few years ago, I was passed a paper written by American researchers Logan Fiorella and Richard E. Mayer. They were exploring the benefits of something called Generative Learning as a way to promote deeper learning. This wasn’t something I was especially familiar with, and as a busy teacher, after a quick glance over it, I popped it in a draw to wait until I had more space to consider it.

However, it played on my mind. Often, we worked hard to encourage students to learn but to no avail. Maybe this could have promise. Some of it seemed intuitive and it would have been quite easy for someone such as myself, who had been teaching for coming on two decades, to dismiss it.

On rereading the paper though, a new perspective began to emerge, and I could see something which had real potential. There seemed to be some compelling evidence, and this might bridge the ‘teacher does / student does’ gap I had been seeing so much of.

So, what do we need to know about it and how might it have an impact in the classroom?

What is it?

First, let me say again, I was right, it was not new. In fact, when unpicking it I saw it had its roots very much in the work of Piaget on schema theory, and Bartlett’s constructionist approach to learning. The term itself was coined by Merlin Wittrock in the 1974 and many of the strategies (eight which research show are the most powerful) are things which people will have already used in the classroom in one form or another.

Select – Organise – Integrate

Generative Learning is really looking at how we make meaning, literally how we generate that understanding. It centres on Mayer’s SOI mental model and explores how we understand information and assimilate it into our existing schemas. This model begins with how we (S) select that information, how we take different pieces of information from a range of materials, be they texts or videos or teacher explanation, and decide which are the most salient points. It then moves into how we (O) organise that information in a coherent way and finally how that is (I) integrated alongside pre-existing information we have embedded into our schema. This is therefore reliant on the relationship between the long-term and working memory, requiring access to prior learning, in order to relate it to the new data being received.

Different strategies or approaches allow us to do this. For example, when we are asked to write a summary, we need to select the appropriate details, re-organise those into a coherent new form, verbal or written, and, once we have checked that is correct, we can store this in its new form. The same process is employed in the other strategies which are self-testing, self-explaining, teaching, enacting, drawing, mapping, and imagining. All these activities have the potential to engage learners in these deeper thinking processes and enable them to store information more effectively in their long-term memory. Deeper learning has taken place.

What does this mean for teachers?

These activities will already be familiar in many classrooms. Indeed, the summary, the self-testing and the mapping were staples of my classroom for many years. However, the significant difference is how we utilise these approaches to make the most of them. For example, a summary may well be used following a learning episode, to capture information in a book, but if it is poorly written, where the incorrect information was selected, or if key elements are missing, then it may well not be generating deeper learning or may even be embedding false learning. If students are asked to draw something and their focus is primarily on producing a beautiful piece of art, rather than the careful selection and organisation of that information, this will lead to an episodic memory, but one which won’t necessarily bring about the meaningful learning we want.

This returns us to clarity of purpose and how we hone these tools to ensure they are really working for us, and our students.

Modelling the processes we go through to produce a concept map, articulating the metacognitive steps how we go through in order to select and organise, will help to ensure that when students are generating meaning, they are approaching it in a generative way.

Selecting the right tool for the right job is also important and using “enacting” with a group of students who already hold expertise may be counterproductive. Instead something like enacting may well be more suitable for students at a time where they are finding it difficult to make an abstract concept concrete. Equally, if we are asking students to generate meaning when they have insufficient knowledge or lack access to high quality to resources to draw upon, including excellent teacher explanation, we may find that, whilst these activities are happening, they are not activating the SOI model.

Careful consideration of how, when, and why we are using these approaches, and making sure they are done well, is essential.

Why does it matter?

Teachers work really hard to help students understand the information they need. We model, question, check, guide, and support learning until the cows come home. However, unless that learning is embedded and students really understand what is going on, create meaning, then we can end up reteaching, finding lots of gaps and, quite understandably, feeling frustrated at their lack of progress. If we use generative strategies or rather, these strategies in a generative way, we begin to make the learning process much more visible. We can check that students are selecting the correct information, organising it in an appropriate way and integrating what we know needs to be retained ready for the next steps.

Importantly, we also begin to empower students. If we share with them how these strategies work and how to do them well, students have a ready-made set of approaches to use in independent work and revision. Our ultimate aim is to create effective and independent learners and Generative Learning can be a very effective way to do just that.


Bartlett, F. C. (1932) Remembering (Cambridge: Cambridge University Press)

Enser, Z. and Enser, M. (2020) Generative Learning in Action (Woodbridge: John Catt Educational Limited)

Fiorella, L. and Mayer, R.E. (2016) ‘Eight ways to Promote Generative Learning’ in Education Psychology Review, 28, 2016, pp.717-741.

Fiorella, L. and Mayer, R.E. (2015) Learning as a Generative Activity: Eight Strategies that Promote Understanding, (Cambridge: Cambridge University Press)

Piaget, J. (1926) The Language and Thought of the Child (London: Kegan Paul, Trench, Trubner, & Co)

Wittrock, M. C. (1974) ‘Learning as a Generative Process’, in Educational Psychologist II, (2), pp. 345-376

Write A Comment