Learning Ecosystems: #WorkingOutLoud on Learning Science

This post is part of #WorkingOutLoud on the Learning Science Guidebook, and considers the notion of the Learning Ecosystem. This section has taken us a couple of weeks to write, with two of our evening ‘workshops’ to kick it around. For me, this experience has been great, as we have really got into one of our aims – to challenge each other and ourselves – i have seen us building upon our respective language and ideas. So whilst this section may not yet be right, i do feel that we are finding our way.

What’s a ‘learning ecosystem’? 

Let’s consider that question from a learner’s perspective: imagine the most effective path to mastery of a new subject. It’s probably a mix of experiences, people, content, and technologies spread out over time.

That learning path might start with ‘how-to’ videos, then self-driven exploration through books, podcasts, an online course followed by in-person classes, then opportunities to put the knowledge into practice at work, some trial-and-error, and after a while, finding a mentor and also higher stakes moments to practise those skills.

Now consider that path multiplied across other subjects and skills, all intersecting across life.

It’s a meandering path, evolving nonlinearly from the simpler to the more nuanced, from the mechanical to the automatic, from the beginner to (for some) the expert. 

Everything, Everywhere: The Wild Forest

At the broadest level, ‘The Learning Ecosystem’ is the entire system of lifelong learning that someone (self) experiences or could experience.

It’s every learning activity from formal (like primary school or professional seminars) to informal (like on-the-job learning) and everything in between (including various forms of Social Learning).

The Learning Ecosystem includes every subject, every level of complexity, and every form of delivery across time.

This ecosystem is like a wild forest – often unconstrained and only partially visible, even to ourselves. We all learn (to varying degrees of effectiveness) within this cacophonous context, and individuals skilled in self-directed learning (also called ‘metacognition’) can traverse The Learning Ecosystem – the wild forest – fairly adequately. But many find it daunting or unnavigable on their own, and even the most metacognitive forest rangers benefit from some support. 

The wild learning ecosystem is highly complex and authentic but poorly managed.

Managed Experiences: Walled Gardens

We try to control our paths through the learning wilderness. Classically, that meant a mentor blazing a trail for a few apprentices. In more recent centuries, we’ve tried to cultivate order – building walled gardens in the forest. In other words, defining controlled learning experiences using methodological instructional processes and latterly the insights of scaled data and analytics, even towards automation..

These learning gardens let us draw boundaries. (‘This’ is a semester-long biology class for 8th grade students. ‘That’ is a day-long teamwork seminar for surgical teams.) The walls help us manage instructional activities, cull out the ‘weeds’ (such as the complexity of other subjects), and foster the growth of specific outcomes. Walled gardens also let us control who enters, charge admission, and verify the quality of ‘produce’ that people leave with (through degree or licences).

Through the formalisation of learning as an activity, we have learned to create structures of organisation and understanding.

But walled gardens have drawbacks. They’re highly curated, with a limited number of ‘flora’ (or areas of focus, bounded by time and complexity) in each.

While this makes them more manageable, it also means that they’re only optimised at local (episodic) versus ‘neighbourhood’ (or broader) levels.

We can make a garden look tidy, beautiful even, and yet that tidyness does not reflect the system at a biological level or true ecosystem level. We have created an understanding, or perspective, as a level of abstraction.

Each garden typically treats each visitor (or learner) as a blank slate – failing to account for the wealth of contextual information that could have been gleaned from that person’s history or related knowledge.

Many gardens have controlled admission gates, limited by time or resources, and the walls between gardens create artificial barriers that force individuals to make sense of the assembly (of the range of learning experiences) independently.

Similarly, less formal learning experiences are often ignored – relegated to the obscured wild outside of the wall, along with performance contexts too. 

Said another way, the walled-garden approach allows for focused control by reducing learning into controllable chunks. This creates a high level of manageability but at the cost of ecological complexity, authenticity, and broader (‘neighbourhood’) outcomes.

In practice, this means that each learning experience – as well as the aggregate continuum of learning – is less efficient and effective than it could be, and that some learners are unable to access the ‘right’ (for them) learning experiences, in the right ways, at the right times.

Harmonisation: The Coherent Learning Ecosystem

Over the last twenty years, there’s been growing interest in creating a ‘Coherent Learning Ecosystem’ that has both the boundless authenticity of the wild forest as well as the manageability of walled gardens.

We’ve added the term ‘coherent’ to distinguish this artificial ecosystem from the true, boundless reality of The Learning Ecosystem. 

When done effectively, a coherent learning ecosystem gives us an increased span of control, more opportunities for optimising outcomes, and greater flexibility in terms of access to and availability of learning. In theory, this lets us monitor, measure, personalise, and optimise learning at a grand scale.

One way to imagine it is like this: today we tend to boundary learning experiences as defined by structure, technology, ownership, sequence, scope. All taking place within a broader ecosystem. Through the potential of the emergent learning technologies, and with a more holistic mindset, we are able to bring the science of learning to bear, and reimagine our approach. To create broader, Coherent Learning Ecosystems that still give us the benefit of being quantifiable, mappable, manageable, and yet also provide greater insight and stretch, adaptation and personalisation.

Coherent learning ecosystems – these semi-natural habitats – try to make the invisible Learning Ecosystem more discernible to both individuals and their organisations.

These human-crafted Coherent Learning Ecosystems also try to create discernable connections among their parts (that is, among the typically separate garden plots), and ideally, to enable meaningful interactions among those components. 

In other words, a coherent learning ecosystem is an assembly of learning experiences that are visible and intentionally interact with one another across the traditional boundaries – across subject areas, time, platforms, and institutions – to improve learning outcomes.

An Organisational Terrarium is not a Learning Ecosystem

Some organisations have attempted to create a ‘walled’ learning ecosystem – a learning terrarium so to speak. The idea is to have a greater diversity and interconnectedness than a walled garden, but still with a high level of hierarchical control. In practice, this typically looks like a learning management system paired with a few other components like a mobile app, digital gradebook for in-person seminars, and something like a catalogue of skills and credentials. Sometimes these systems also include some personalisation based on performance, goals, or attributes like occupation or seniority.

Such systems have value, and most could arguably be called ‘learning experience platforms,’ but they don’t meet the spirit of a ‘Learning Ecosystem.’

Such hierarchical learning stacks are still walled, and although their acreage may be larger than a conventional walled garden, they’re still constrained in ways that limit their flexibility, authenticity, and strategic impact.

These ‘terrariums’ still only represent a small subset of the breadth of learning experiences. They’re poor reflections of the wider Learning Ecosystem or even the coherent learning ecosystem concept introduced in the prior section.

One way to consider it is this: if an etymologist wishes to understand the biosphere bubble of an oak tree, they may put a net under it and shake the tree, or smoke it, to catch all the bugs. These can be counted and categorised. Through observation (who flies in and out) and direct collection, we may come to understand that tree. But dig it up and put it into a glass box, with all those bugs, and the tree will die.

The biosphere bubble of the oak tree is held within a broader ecosystem, acting upon it and being acted upon. We may gain local insight and understanding, but the true ecosystem is broader and interdependent.

An Artificial Learning Ecosystem in Practice

Although it’s impossible to truly know (let alone control) the wild forest of learning (the full Learning Ecosystem – unquantifiably complex), there’s an attainable midpoint between it and a simple walled garden (the Terrarium – often a mix of legacy and owned infrastructure).

In prior work, Sae has called this concept the ‘Future Learning Ecosystem,’ but we could also call it the ‘Technology Learning Ecosystem’ or something like the ‘Coherent, Nonhierarchical Learning Ecosystem.’ (Clearly, we need some titular help – this is still work in progress!) 

The idea is to make visible and to connect learning across time (i.e., learning over a lifetime), place (e.g., institution, platform, various learning experiences), and granularity (where, granularity refers to the level of focus of a learning experience, from a single lesson to a career-enabling assembly of expertise). And, to the extent feasible, include a range of formal and informal learning experiences.

Although this system can’t fully reflect the breadth of the full Learning Ecosystem, it attempts to approximate that complexity by shifting from a hierarchical approach (where control resides within a single organisation) to a distributed model (where different organisations communicate and interact) – similar to how the internet works.

Control is passed from organisation (or learn/self) to organisation (or learner) over time, similar to how relay racers might pass a baton, although nonlinear. 

This concept is enabled through technology, data exchange, and learning engineering processes. Although such advanced technologies aren’t technically required. Alexander the Great, for instance, enjoyed a sort of ‘artificial learning ecosystem’ thanks to his famous cadre of tutors. But, since we’re not all as wealthy as the ancient king, the rest of us have to rely on AI and inter-organisational systems to fill the gap. 

Despite the technical intricacy of this idea, it isn’t some hypothetical dream. These capabilities – including data and standards, interoperable competencies, learning analytics, adaptive algorithms, and learning engineering principles – already exist.

Our collective challenge is not so much technical as it is social and human-centric: How do we shift mindsets to accept this nonhierarchical approach across organisations? How do we reliably expand craftsperson-like effects (such as top teachers) from local levels to larger scales? And how do we navigate this (soon to be) ‘new normal’ for learning outside of our old, walled-in methods?

What you need to know

  • Learning takes place at different levels of abstraction – from the full Learning Ecosystem (everything), to the Walled Gardens (structural systems) and Organisational Terrariums (owned and often technologically boundaried).
  • We propose that there is value in considering Coherent Learning Ecosystems (intra-organisational – across TIME – PLACE – and GRANULARITY).
  • This will require a fundamental revision of our notions of how learning is owned and structured.

This is already happening: both technology and our understanding from Learning Science can allow us to do this – but we will be held back by our legacy structures of ownership and understanding – unless we create space to experiment and explore.

What we can do about it

  • Find or create communities to explore this. We know what we used to have, but not what we will have in the future. Be part of the conversation.
  • Consider how our thinking about learning design has been constrained by that which is easy to own or control, to quantify – and whether that mindset is driven more by convenience and ease, than by evidence of how we truly learn. And how can you think differently about this? Part of the conversations may be to decide what we can stop doing.

This post is shared as part of #WorkingOutLoud as i write with Sae Schatz and Geoff Stead to pull together the Learning Science Guidebook. We are working at speed, collaborating in various spaces, and having some fun with our ideas. As such, it’s evolving thinking and first draft illustrations. Errors are my own.

About julianstodd

Author, Artist, Researcher, and Founder of Sea Salt Learning. My work explores the context of the Social Age and the intersection of formal and social systems.
This entry was posted in Learning and tagged , , , , , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.