A few thoughts on learning management systems, and on integrated learning environments and their implementation

Why do we build digital learning systems to mimic classrooms?

It is understandable that, when we teach in person, we have to occupy and make different uses of the same or similar environments like classrooms, labs, workshops, lecture theatres, and offices. There are huge financial, physical, and organizational constraints on making the environment fit the task, so it would be madness to build a whole new classroom every time we wished to run a different class.

Online, we could build anything we like

But why do we do the same when we teach online? There are countless tools available and, if none are suitable, it is not too hard to build them or modify them to suit our needs. Once they are built, moving between them just takes a tap of a screen or the click of a mouse. Heck, you can even occupy several of them at once if you have a decent monitor or more than one device.

So why don’t we do this?

Here are a few of the more obvious reasons that using the perfect app for the context of study rarely happens:

  • Teachers’ lack of knowledge of the options (it takes time and effort to discover what’s available).
  • Teachers’ lack of skill in using them (most interesting tools have a learning curve, and that gets steeper in inverse proportion to the softness and diversity of the toolset, so most teachers don’t even know how to make the most of what they already have).
  • Lack of time and/or money for development (a real-life application is what it contains, not just the shell that contains it, and it is not always as easy to take existing stuff and put it in a new tool as it might be in a physical space).
  • Costs and difficulties in management (each tool adds costs in managing faults, configuration, accounting for use, performance, and security).
  • Cognitive load involved for learners in adapting to the metaphors, signposts, and methods needed to use the tool itself.

All of these are a direct consequence of the very diversity that would make us want to use different apps in the first place. This is a classic Faustian bargain in which the technology does what we want, and in the process creates new problems to solve.  Every virtual system invents at least some of the dynamics of how people and things interact with it and within it. In effect, every app has its own physics. That makes them harder to find out about, harder to learn, harder to develop, costlier to manage, and more difficult to navigate than the static, fixed facilities found in particular physical locations. They are all different, there are few if any universals, and any universal today may become a conditional tomorrow. Gravity doesn’t necessarily work the same way in virtual systems.

image of a pile of containersAnd so we get learning management systems

The learning management system (LMS) kind of deals with all of these problems: poorly, harmfully, boringly, and painfully, but it does deal with them. Currently, most of the teaching at Athabasca University is through the open source Moodle LMS, lightly modified by us because our needs are not quite like others (self-pacing and all that). But Moodle is not special: in terms of what it does and how it does it, it is not significantly different from any other mainstream LMS – Blackboard, Brightspace, Canvas, Sakai, whatever.

Almost every LMS essentially automates the functions, though not exactly the form, of traditional classrooms. In other parts of the world people prefer to use the term ‘managed learning environment’ (MLE) for such things, and it is the most dominant representative of a larger category of systems usually described as virtual learning environments (VLEs) that also includes things like MOOs (multi-user dungeons, object oriented), immersive learning environments, and simpler web-based teaching systems that replicate aspects of classrooms such as Google Classroom or Microsoft’s gnarly bundle of hastily repurposed rubbish for teaching that I’m not sure even has a name yet. Notice the spatial metaphors in many of these names.

Little boxes made of ticky tacky

The people who originally designed LMSs back in the 90s (I did so myself) based their designs on the functions and entities found in a traditional university because that was their context, and that was where they had to fit. Metaphorically, an LMS or MLE is a big university building with rather uniform classrooms, with perhaps a yard where you can camp out with a few other systems (plugins, LTI hooks, etc) that conform to its requirements and that are allowed in to classrooms when invited, and a few doors and gateways (mainly hyperlinks) linking it circuitously or in jury-rigged fashion to other similarly weakly connected buildings (e.g. places to register, places to seek support, places to talk to an advisor, places to complain, places to find books, and so on). It doesn’t have metaphorical corridors, halls, common rooms, canteens, yards, libraries or any of the other things that normally make up a physical university. You rarely get to even be aware of other classrooms beyond those you are in. Some people (me in a past life) might give classrooms cute names like ‘the learning cafe’ but it’s still just another classroom. You teleport from one classroom to the next because what happens in corridors (really a big lot of incredibly important pedagogically useful stuff, as it happens) is not perceived by the designers as a useful classroom function to be automated or perhaps, more charitably, they just couldn’t figure out how to automate that.

Reified roles

It’s a very controlled environment where everyone has a programmatically enforced role (mostly reflecting traditional educational roles), that may vary according to the room, but that are far less fluid than those in physical spaces. There are strong hierarchies, and limited opportunities for moving between them. Some of those hierarchies are new: the system administrator, for instance, has way more power than anyone in a physical university to determine how learning happens, like an architect with the power to move walls, change the decor, add extensions, and so on, at will. The programmers of the system are almost god-like in their command of its physics. But the ways that they give teachers (or learning designers, or administrators) control, as designers, directors, and regulators of the classroom, are perhaps the most pernicious. In a classroom a teacher may lead (and, by default, usually does). In an LMS, a teacher (or someone playing that role) must lead. The teacher sees things that students cannot, and controls things that the students may not. A teacher configures the space, and determines with some precision how it will be used. With a lot of effort and risk, it can be made to behave differently, but it almost never is.

Functions are everything

An LMS is typically built along functional lines, and those functions are mostly based on loose, superficial observations of what teachers and students seem to do in physical classrooms. The metaphorical classrooms are weird, because they are structured by teaching (seldom learning) function rather than along pedagogical lines: for instance, if you want to talk with someone, you normally need to go to a separate enclosed area inside the classroom or leave a note on the teacher’s desk. Same if you want to take a test, or share your work with others. Another function, another space. Some have many little rooms for different things. Lectures are either literally that (video recordings) or (more usefully, from a learning perspective), text and images to be read on screen, based on the assumption that the only function of lectures is information transmission (it is so very, very much not – that’s its least useful and least effective role). There’s seldom a chance to put even put up your hand to question something. Notices can usually only be pinned on the wall by teachers. Classroom timetables are embodied in software because of course you need a rigid and unforgiving timetable in a medium that sells itself on enabling learning anywhere, any time. Some, including Moodle, will allow you to break up the content differently, but it’s still another timetable; just a timetable without dates. It’s still the teacher who sets the order, pacing and content.

Robot overlords

It’s a high-tech classroom. There are often robots there that are programmed to make you behave in ways determined by those higher in the hierarchy (sometimes teachers, sometimes administrators, sometimes the programmers of the software). For instance, they might act as gatekeepers that prevent you from moving on to the next section before completing the current one, or they might prevent you submitting work before or after a specified date. They might mark your work. There are surveillance cameras everywhere, recording your every move, often only accessible to those with more powerful roles (though sometimes a robot or two might give you a filtered view of it).

Beginnings and ends

You can’t usually go back and visit when your course is over because someone decided it would be a good idea to set opening and closing enrolment dates and assumed that, when they were done, the learning was done (which of course it never is – it keeps on evolving long after explicit teaching and testing occurred). Again, it’s because physical classes are scheduled and terms come to an end because they must be, not because it makes pedagogical sense. And, like almost everything, you can override this default, but hardly anyone ever does, because it brings back those Faustian bargains, especially in manageability.

Dull caricatures of physical spaces

Basically, the LMS is an automated set of metaphorical classrooms that hardens many of the undesirable by-products of educational systems in software in brain-dead ways that have little to do with how best to teach, and that stretch the spatial metaphors that inform it beyond breaking point. Each bit of automation and each navigational decision hardens pedagogical choices. For all the cozy metaphors, programmers invent rather than replicate physics, in the process warping reality in ways that do no good and much harm. Classrooms solved problems of physics for in-person teaching and form part of a much larger structure that has evolved to teach reasonably well (including corridors, common rooms, canteens, and libraries, as it happens). Their more visible functions are only a part of that and, arguably, not the main part. There is much pedagogy embedded in the ways that physical universities, whether by accident or design, have evolved over centuries to support learning in every quadrangle and nook of a coffee shop. LMSs just focus on a limited subset of teaching roles, and empower the teacher in ways that caricature their already excessive dominance in the classroom (which only occurred because it had to, thanks to physics and the constraints it imposed).

LMSs are crap, but they contain recognizable semblances of their physical counterparts and just enough configurability and flexibility to more or less work as teaching tools, a bit, for everyone, almost no matter what their level of digital proficiency might be. They more or less solve the Faustian bargains listed earlier, but they do so by stifling what we wanted and should have been able to do in the first place with online tools, in the process creating new and quite horrific problems, as well as demolishing most of what makes physical universities work in the first place. It never has been true that virtual learning environments are learning environments – they are only ever parts of them – and there are places to escape from them, such as the Landing, other virtual systems, or even just plain old email, but then all those Faustian bargains come back to haunt us again. There has to be a better way.

Beyond the LMS

Cognisant of the issues, Athabasca University is now some way down the path to developing its own distinctive solutions to these problems, in a multi-year multi-million-dollar initiative known as (following the spatial metaphor) the Integrated Learning Environment (ILE). The ILE is not an application. It is an umbrella term for a lot of different, usually independent systems working together as one. Though some of the most interesting opportunities are still only loosely imagined, perhaps because they cause problems that are fiendishly hard to solve (e.g. how can we integrate systems that we build ourselves without creating risks for the rest of the ILE, and what happens when they need to be maintained?) a lot of progress is being made on the non-teaching foundations on which the rest depends (student admin systems, support tools, procedures, etc), as well as on the most visible and perhaps the biggest of its parts, BrightSpace, a proprietary commercial LMS that is meant to replace Moodle, for no obvious pedagogical or technical reasons (it’s no better). It might make economic sense. I don’t know, but I do know that open source software typically costs a fair bit to own, albeit because of the things that make it a much better idea (freedom, flexibility, ownership, etc). There is probably a fair bit of time and money being spent with Desire2Learn (makers of Brightspace) on the things that we spent a fair bit of time and money on many years ago to make Moodle a bit less classroom-like. The choice no doubt has something to do with how reliably and easily it can be made to work with some of the other proprietary commercial systems that someone has decided will make up the ILE. It bothers me greatly that we are not trying hard to choose open source solutions, for reasons that will become clearer in the rest of this post. However, (pedagogically speaking) all the mainstream LMSs are much of a muchness, making the same mistakes as one another in very similar ways, so it probably won’t wreck too much of what we already do within Moodle. But, on its own, it won’t move us much further forward and we could do it better. That’s what the ILE is supposed to do – to make the LMS just a part of a much larger teaching environment, intimately connected with the rest of what the university does for or with students, and extensible with new and better ways of learning, teaching, and assessing learning.

picture of lego bricksLego bricks make poor metaphors

When we were first imagining the ILE, though the approach was admirably participative, engaging much of the university community, I was very worried by the things we were encouraged to focus on. It was all about the functionality, the usability, the design, the tools, the pedagogies, the business systems that supported them. Those things matter, for sure, and should be not be ignored, but they should and will change and grow all the time: in fact, part of the point of building this thing is to do just that. Using the city metaphor, pretty much all that we (collectively) considered were the spaces (the rooms, mainly), and the stuff that goes on inside them, much like LMS designers thought of universities as just collections of classrooms in which teaching functions were performed. Space and stuff are, not uncoincidentally, exactly what Stewart Brand identified long ago as inevitably being the fastest-changing, most volatile parts of any town or city (after site, structure, skin, and services). I’ve written a fair bit on the universality of this principle across all systems. It’s a solid structural principle that applies as much to ecosystems and educational systems as to cities. As Brand observes himself, drawing from O’Neill et al (1986), the larger, slower-changing elements of any system affect the smaller, faster-changing more than vice versa. This is for much the same reasons that path dependencies set in. It’s about the prior providing the context for what follows. Flexible things have to fit into the gaps left by less flexible, older, pre-existing things. In physical spaces, of course these tend to be bigger and/or slower, but the same is true in virtual spaces, where size seldom matters that much, but hardness (inflexibility, brittleness) really does. Though lip service was paid to the word ‘integrated’ in our discussions,  I had the strong feeling that the kind of integration we had in mind was that of a Lego set. In fact, I think we were aiming to find a ‘Lego Athabasca University’ set, with assembly instructions and a picture on the box. The vendors who came to talk with us made much of how effectively they could do that, rather than how effectively they could make it possible for others to do that.

Metaphors matter. Lego bricks have to fit together tightly, in pre-specified ways, especially if you are following a plan. If you want to move them around, you have to dismantle a bit of the structure to fit them in. It’s difficult to integrate things that are not bricks, or that are made by different toy companies to work in different ways. At best you get what Brand calls ‘magazine architecture’, or ‘no road’ architecture, beautiful, fit for purpose, intricate and solid, but slow to learn. Lego is not a terrible way to build, compared with buying everything pre-assembled, but it could be improved.

Signals and boundaries

Drawing inspiration from John Holland’s brilliant last work, Signals & Boundaries, I tried to make the case that, instead, we should be focusing on the boundaries (the interfaces between the buildings and the rest of the city), and the signals that pass between them (the people, the messages, etc, the forms they take and how they move around). In Brand’s terms, I wanted us to be thinking about skin and services, and perhaps even structure, though site – Athabasca University – was a given. Though a few people nodded in agreement, I think it mainly fell on deaf ears. We wanted oven-ready solutions, not the infrastructure to enable those solutions. Though the city metaphor works well, because we are talking about human constructions, others would result in similar ways of thinking: cells in bodies, organisms in ecosystems, brains, termite mounds, and so on. All are organized by boundaries (at many levels of hierarchy) and the signals that pass between them.

The Lego set metaphor – whether deliberately or not – seems to have prevailed for now. A lot of old buildings are being slated for demolition and a lot of new virtual buildings are now being erected as part of this development, many of them chosen not because of problems with existing buildings but so that they can more easily connect together and live in the same cloud. This will very likely work, for now, but it is not cheap and it is not flexible, especially given the fact that most of it is not open so, like a rental property, we are not allowed to fix things, add utilities, change the walls, etc, and we are wholly dependent on the landlords being nice to us and each other (knowing that some – ahem, Microsoft – have a long history of abusing their tenants). Those buildings will age. We will find them cramped. Some will age faster than others, and will have to be modified to keep up, perhaps at high cost. Companies renting them might go out of business or change their terms so we might have to demolish the buildings and rent/make new ones. We will be annoyed at how they do things, usually without asking us. We will hate the landlords who dictate what we can do and how we can do it, and who will keep upping the rent while not doing what we ask. We will want more, and the only way to get it will be to build extensions, buy new brick sets, if it is not enough to pay someone to remodel the interiors (and it won’t be). Of course, because most of the big structural elements will not be open source, we will not be able to do that ourselves.

What the ILE really should be

The ILE is, I think, poorly named, because it should not be an environment at all. Following the building metaphor, the ILE is (or should be) more like the system that connects a lot of buildings, bringing them together into a coherent, safe, livable community. It’s infrastructure and services; it is the roads, the traffic signals, the doors, the sidewalks, the water pipes, the waste pipes, the electricity, the network cables; it is the services – fire, police, schools, traffic control, etc; it is all the many rules, standards, norms and regulations that make them work together to help make an environment in which people can live, work, play, and grow. It’s part of the environment – the part that makes it work – but it is not the environment itself. The environment itself is Athabasca University, not just the tools, processes, and systems that support its functions. That includes, most importantly, the people who are part of the university, or who are visitors to it, who are not just users of the environment or dwellers in its walls, but who are or should be the most significant and visible parts of it, just as trees are part of the environment of forests, not users of the forest. Those people live in physical as well as other virtual environments (social media, Word documents, websites, etc) that the ILE can connect together too, to make them a part of it, so the spatial metaphor gets weird at this point. The ILE makes environmental boundaries fuzzy, permeable, and shifting. It’s not an ILE, it’s an ILI – an integrated learning infrastructure.

If we focused on the connections and interfaces, and on how information and processes need to pass across them, and if we thought hard about the nature of those signals, then we could build a system that is resilient, that adapts, that lasts, that grows, that evolves, with parts that we can seamless replace or improve because the interfaces – the building facades, the mains pipes, the junction boxes, etc – will mostly stay the same, evolving slowly as they should. This is about strategy, not planning,  a way of thinking about systems rather than a sequence of things to do.

Some of the key people involved in the process realize this. They are talking about standards, protocols, and projects to build interfaces between systems, and imagining future needs, though they are inevitably distracted by the process of renting Lego bricks, so I am not sure how much they will be able to stay focused on that. I hope they prevail over those who think they are building a set of classrooms and tightly connected admin offices out of self-contained interlocking bricks because our future depends on getting it right. We are aiming to grow. It just takes one critical piece in the Lego building to fail to support that, and the rest falls apart like a… well, like a pile of bricks.

References

Brand, S. (1997). How buildings learn. Phoenix Illustrated. https://www.penguinrandomhouse.ca/books/320919/how-buildings-learn-by-stewart-brand/9780140139969

Holland, J. H. (2012). Signals and Boundaries: Building Blocks for Complex Adaptive Systems. MIT Press.  https://mitpress.mit.edu/books/signals-and-boundaries

O’Neill, R.V., DeAngelis, D.L, Waide, J. B., & Allen, T. F. H. (1986). A Hierarchical Concept of Ecosystems. Princeton University Press. http://www.gbv.de/dms/bs/toc/025157787.pdf

Postman, N. (1998). Five things we need to know about technological change. Denver, Colorado, 28.  https://student.cs.uwaterloo.ca/~cs492/papers/neil-postman–five-things.html

Skills lost due to COVID-19 school closures will hit economic output for generations (hmmm)

Snippet from OECD report on covid-19 and education This CBC report is one of many dozens of articles in the world’s press highlighting one rather small but startling assertion in a recent OECD report on the effects of Covid-19 on education – that the ‘lost’ third of a year of schooling in many countries will lead to an overall lasting drop in GDP of 1.5% across the world. Though it contains many more fascinating and useful insights that are far more significant and helpful, the report itself does make this assertion quite early on and repeats it for good measure, so it is not surprising that journalists have jumped on it. It is important to observe, though, that the reasoning behind it is based on a model developed by Hanushek and Woessman over several years, and an unpublished article by the authors that tries to explain variations in global productivity according to amount and  – far more importantly – the quality of education: that long-run productivity is a direct consequence of the cognitive skills (or knowledge capital) of a nation, that can be mapped directly to how well and how much the population is educated.

As an educator I find this model, at a glance, to be reassuring and confirmatory because it suggests that we do actually have a positive effect on our students. However, there may be a few grounds on which it might be challenged (disclaimer: this is speculation). The first and most obvious is that correlation does not equal causation. The fact that countries that do invest in improving education consistently see productivity gains to match in years to come is interesting, but it raises the question of what led to that investment in the first place and whether that might be the ultimate cause, not the education itself.  A country that has invested in increasing the quality of education would, normally, be doing so as a result of values and circumstances that may lead to other consequences and/or be enabled by other things (such as rising prosperity, competition from elsewhere, a shift to more liberal values, and so on).  The second objection might be that, sure, increased quality of education does lead to greater productivity, but that it is not the educational process that is causing it, as such. Perhaps, for instance, an increased focus on attainment raises aspirations. A further objection might be that the definition of ‘quality’ does not measure what they think it measures. A brief skim of the model used suggests that it makes extensive use of scores from the likes of TIMSS, PIRLS and PISA, standardized test approaches used to compare educational ‘effectiveness’ in different regions that embody quite a lot of biases, are often manipulated at a governmental level, and that, as I have mentioned once or twice before, are extremely dubious indicators of learning: in fact, even when they are not manipulated, they may indicate willingness to comply with the demands of the powerful more than learning (does that improve GDP? Probably).  Another objection might be that absence of time spent in school does not equate to absence of education. Indeed, Hanushek and Woessman’s central thesis is that it is not the amount but the quality of schooling that matters, so it seems bizarre that they might fall back on quantifying learning by time spent in school. We know for sure that, though students may not have been conforming to curricula at the rate desired by schools and colleges, they have not stopped learning. In fact, in many ways and in many places, there are grounds to believe that there have been positive learning benefits: better family learning, more autonomy, more thoughtful pedagogies, more intentional learning community forming, and so on.  Out of this may spring a renewed focus on how people learn and how best to support them, rather than maintaining a system that evolved in mediaeval times to support very different learning needs, and that is so solidly packed with counter technologies and so embedded in so many other systems that have nothing to do with learning that we have lost sight of the ones that actually matter. If education improves as a result, then (if it is true that better and more education improves the bottom line) we may even see gains in GDP. I expect that there are other reasons for doubt: I have only skimmed the surface of the possible concerns.

I may be wrong to be sceptical –  in fairness, I have not read the many papers and books produced by Hanushek and Woessman on the subject, I am not an economist, nor do I have sufficient expertise (or interest) to analyze the regression model that they use. Perhaps they have fully addressed such concerns in that unpublished paper and the simplistic cause-effect prediction distorts their claims. But, knowing a little about complex adaptive systems, my main objection is that this is an entirely new context to which models that have worked before may no longer apply and that, even if they do, there are countless other factors that will affect the outcome in both positive and negative ways, so this is not so much a prediction as an observation about one small part of a small part of a much bigger emergent change that is quite unpredictable. I am extremely cautious at the best of times whenever I see people attempting to find simple causal linear relationships of this nature, especially when they are so precisely quantified, especially when past indicators are applied to something wholly novel that we have never seen before with such widespread effects, especially given the complex relationships at every level, from individual to national.  I’m glad they are telling the story – it is an interesting one that no doubt contains grains of important truths – but it is just an informative story, not predictive science.  The OECD has a bit of track record on this kind of misinterpretation, especially in education. This is the same organization that (laughably, if it weren’t so influential) claimed that educational technology in the classroom is bad for learning. There’s not a problem with the data collection or analysis, as such. The problem is with the predictions and recommendations drawn from it.

Beyond methodological worries, though, and even if their predictions about GDP are correct (I am pretty sure they are not – there are too many other factors at play, including huge ones like the destruction of the environment that makes the odd 1.5% seem like a drop in the barrel) then it might be a good thing. It might be that we are moving – rather reluctantly – into a world in which GDP serves as an even less effective measure of success than it already is. There are already plentiful reasons to find it wanting, from its poor consideration of ecological consequences to its wilful blindness to (and causal effect upon) inequalities, to its simple inadequacy to capture the complexity and richness of human culture and wealth. I am a huge fan of the state of Bhutan’s rejection of the GDP, that it has replaced with the GNH happiness index. The GNH makes far more sense, and is what has led Bhutan to be one of the only countries in the world to be carbon positive, as well as being (arguably but provably) one of the happiest countries in the world. What would you rather have, money (at least for a few, probably not you), or happiness and a sustainable future? For Bhutan, education is not for economic prosperity: it is about improving happiness, which includes good governance, sustainability, and preservation of (but not ossification of) culture.

Many educators – and I am very definitely one of them – share Bhutan’s perspective on education. I think that my customer is not the student, or a government, or companies, but society as a whole, and that education makes (or should make) for happier, safer, more inventive, more tolerant, more stable, more adaptive societies, as well as many other good things. It supports dynamic meta-stability and thus the evolution of culture. It is very easy to lose sight of that goal when we have to account to companies, governments, other institutions, and to so many more deeply entangled sets of people with very different agendas and values, not to mention our inevitable focus on the hard methods and tools of whatever it is that we are teaching, as well as the norms and regulations of wherever we teach it. But we should not ever forget why we are here. It is to make the world a better place, not just for our students but for everyone. Why else would we bother?

Originally posted at: https://landing.athabascau.ca/bookmarks/view/6578662/skills-lost-due-to-covid-19-school-closures-will-hit-economic-output-for-generations-hmmm

Black holes are simpler than forests and science has its limits

Mandelbrot set (Wikipedia, https://en.wikipedia.org/wiki/Mandelbrot_set)Martin Rees (UK Astronomer Royal) takes on complexity and emergence. This is essentially a primer on why complex systems – as he says, accounting for 99% of what’s interesting about the world – are not susceptible to reductionist science despite being, at some level, reducible to physics. As he rightly puts it, “reductionism is true in a sense. But it’s seldom true in a useful sense.” Rees’s explanations are a bit clumsy in places – for instance, he confuses ‘complicated’ with ‘complex’ once or twice, which is a rooky mistake, and his example of the Mandelbrot Set as ‘incomprehensible’ is not convincing and rather misses the point about why emergent systems cannot be usefully explained by reductionism (it’s about different kinds of causality, not about complicated patterns) – but he generally provides a good introduction to the issues.

These are well-trodden themes that most complexity theorists have addressed in far more depth and detail, and that usually appear in the first chapter of any introductory book in the field, but it is good to see someone who, from his job title, might seem to be an archetypal reductive scientist (he’s an astrophysicist) challenging some of the basic tenets of his discipline.

Perhaps my favourite works on the subject are John Holland’s Signals and Boundaries, which is a brilliant, if incomplete, attempt to develop a rigorous theory to explain and describe complex adaptive systems, and Stuart Kauffman’s flawed but stunning Reinventing the Sacred, which (with very patchy success) attempts to bridge science and religious belief but that, in the process, brilliantly and repeatedly proves, from many different angles, the impossibility of reductive science explaining or predicting more than an infinitesimal fraction of what actually matters in the universe. Both books are very heavy reading, but very rewarding.

Address of the bookmark: https://aeon.co/ideas/black-holes-are-simpler-than-forests-and-science-has-its-limits

Originally posted at: https://landing.athabascau.ca/bookmarks/view/2874665/black-holes-are-simpler-than-forests-and-science-has-its-limits

Cocktails and educational research

A lot of progress has been made in medicine in recent years through the application of cocktails of drugs. Those used to combat AIDS are perhaps the most well-known, but there are many other applications of the technique to everything from lung cancer to Hodgkin’s lymphoma. The logic is simple. Different drugs attack different vulnerabilities in the pathogens etc they seek to kill. Though evolution means that some bacteria, viruses or cancers are likely to be adapted to escape one attack, the more different attacks you make, the less likely it will be that any will survive.

Simulated learningUnfortunately, combinatorial complexity means this is not a simply a question of throwing a bunch of the best drugs of each type together and gaining their benefits additively. I have recently been reading John H. Miller’s ‘A crude look at the whole: the science of complex systems in business, life and society‘ which is, so far, excellent, and that addresses this and many other problems in complexity science. Miller uses the nice analogy of fashion to help explain the problem: if you simply choose the most fashionable belt, the trendiest shoes, the latest greatest shirt, the snappiest hat, etc, the chances of walking out with the most fashionable outfit by combining them together are virtually zero. In fact, there’s a very strong chance that you will wind up looking pretty awful. It is not easily susceptible to reductive science because the variables all affect one another deeply. If your shirt doesn’t go with your shoes, it doesn’t matter how good either are separately. The same is true of drugs. You can’t simply pick those that are best on their own without understanding how they all work together. Not only may they not additively combine, they may often have highly negative effects, or may prevent one another being effective, or may behave differently in a different sequence, or in different relative concentrations. To make matters worse, side effects multiply as well as therapeutic benefits so, at the very least, you want to aim for the smallest number of compounds in the cocktail that you can get away with. Even were the effects of combining drugs positive, it would be premature to believe that it is the best possible solution unless you have actually tried them all. And therein lies the rub, because there are really a great many ways to combine them.

Miller and colleagues have been using the ideas behind simulated annealing to create faster, better ways to discover working cocktails of drugs. They started with 19 drugs which, a small bit of math shows, could be combined in 2 to the power of 19 different ways – about half a million possible combinations (not counting sequencing or relative strength issues). As only 20 such combinations could be tested each week, the chances of finding an effective, let alone the best combination, were slim within any reasonable timeframe. Simplifying a bit, rather than attempting to cover the entire range of possibilities, their approach finds a local optimum within one locale by picking a point and iterating variations from there until the best combination is found for that patch of the fitness landscape. It then checks another locale and repeats the process, and iterates until they have covered a large enough portion of the fitness landscape to be confident of having found at least a good solution: they have at least several peaks to compare. This also lets them follow up on hunches and to use educated guesses to speed up the search. It seems pretty effective, at least when compared with alternatives that attempt a theory-driven intentional design (too many non-independent variables), and is certainly vastly superior to methodically trying every alternative, inasmuch as it is actually possible to do this within acceptable timescales.

The central trick is to deliberately go downhill on the fitness landscape, rather than following an uphill route of continuous improvement all the time, which may simply get you to the top of an anthill rather than the peak of Everest in the fitness landscape. Miller very effectively shows that this is the fundamental error committed by followers of the Six-Sigma approach to management, an iterative method of process improvement originally invented to reduce errors in the manufacturing process: it may work well in a manufacturing context with a small number of variables to play with in a fixed and well-known landscape, but it is much worse than useless when applied in a creative industry like, say, education, because the chances that we are climbing a mountain and not an anthill are slim to negligible. In fact, the same is true even in manufacturing: if you are just making something inherently weak as good as it can be, it is still weak. There are lessons here for those that work hard to make our educational systems work better. For instance, attempts to make examination processes more reliable are doomed to fail because it’s exams that are the problem, not the processes used to run them. As I finish this while listening to a talk on learning analytics, I see dozens of such examples: most of the analytics tools described are designed to make the various parts of the educational machine work ‘ better’, ie. (for the most part) to help ensure that students’ behaviour complies with teachers’ intent. Of course, the only reason such compliance was ever needed was for efficient use of teaching resources, not because it is good for learning. Anthills.

This way of thinking seems to me to have potentially interesting applications in educational research. We who work in the area are faced with an irreducibly large number of recombinable and mutually affective variables that make any ethical attempt to do experimental research on effectiveness (however we choose to measure that – so many anthills here) impossible. It doesn’t stop a lot of people doing it, and telling us about p-values that prove their point in more or less scupulous studies, but they are – not to put too fine a point on it – almost always completely pointless.  At best, they might be telling us something useful about a single, non-replicable anthill, from which we might draw a lesson or two for our own context. But even a single omitted word in a lecture, a small change in inflection, let alone an impossibly vast range of design, contextual, historical and human factors, can have a substantial effect on learning outcomes and effectiveness for any given individual at any given time. We are always dealing with a lot more than 2 to the power of 19 possible mutually interacting combinations in real educational contexts. For even the simplest of research designs in a realistic educational context, the number of possible combinations of relevant variables is more likely closer to 2 to the power of 100 (in base 10 that’s  1,267,650,600,228,229,401,496,703,205,376). To make matters worse, the effects we are looking for may sometimes not be apparent for decades (having recombined and interacted with countless others along the way) and, for anything beyond trivial reductive experiments that would tell us nothing really useful, could seldom be done at a rate of more than a handful per semester, let alone 20 per week. This is a very good reason to do a lot more qualitative research, seeking meanings, connections, values and stories rather than trying to prove our approaches using experimental results. Education is more comparable to psychology than medicine and suffers the same central problem, that the general does not transfer to the specific, as well as a whole bunch of related problems that Smedslund recently coherently summarized. The article is paywalled, but Smedlund’s abstract states his main points succinctly:

“The current empirical paradigm for psychological research is criticized because it ignores the irreversibility of psychological processes, the infinite number of influential factors, the pseudo-empirical nature of many hypotheses, and the methodological implications of social interactivity. An additional point is that the differences and correlations usually found are much too small to be useful in psychological practice and in daily life. Together, these criticisms imply that an objective, accumulative, empirical and theoretical science of psychology is an impossible project.”

You could simply substitute ‘education’ for ‘psychology’ in this, and it would read the same. But it gets worse, because education is as much about technology and design as it is about states of mind and behaviour, so it is orders of magnitude more complex than psychology. The potential for invention of new ways of teaching and new states of learning is essentially infinite. Reductive science thus has a very limited role in educational research, at least as it has hitherto been done.

But what if we took the lessons of simulated annealing to heart? I recently bookmarked an approach to more reliable research suggested by the Christensen Institute that might provide a relevant methodology. The idea behind this is (again, simplifying a bit) to do the experimental stuff, then to sweep the normal results to one side and concentrate on the outliers, performing iterations of conjectures and experiments on an ever more diverse and precise range of samples until a richer, fuller picture results. Although it would be painstaking and longwinded, it is a good idea. But one cycle of this is a bit like a single iteration of Miller’s simulated annealing approach, a means to reach the top of one peak in the fitness landscape, that may still be a low-lying peak. However if, having done that, we jumbled up the variables again and repeated it starting in a different place, we might stand a chance of climbing some higher anthills and, perhaps, over time we might even hit a mountain and begin to have something that looks like a true science of education, in which we might make some reasonable predictions that do not rely on vague generalizations. It would either take a terribly long time (which itself might preclude it because, by the time we had finished researching, the discipline will have moved somewhere else) or would hit some notable ethical boundaries (you can’t deliberately mis-teach someone), but it seems more plausible than most existing techniques, if a reductive science of education is what we seek.

To be frank, I am not convinced it is worth the trouble. It seems to me that education is far closer as a discipline to art and design than it is to psychology, let alone to physics. Sure, there is a lot of important and useful stuff to be learned about how we learn: no doubt about that at all, and a simulated annealing approach might speed up that kind of research. Painters need to know what paints do too. But from there to prescribing how we should therefore teach spans a big chasm that reductive science cannot, in principle or practice, cross. This doesn’t mean that we cannot know anything: it just means it’s a different kind of knowledge than reductive science can provide. We are dealing with emergent phenomena in complex systems that are ontologically and epistemologically different from the parts of which they consist. So, yes, knowledge of the parts is valuable, but we can no more predict how best to teach or learn from those parts than we can predict the shape and function of the heart from knowledge of cellular organelles in its constituent cells. But knowledge of the cocktails that result – that might be useful.