“Technology is not neutral. We’re inside of what we make, and it’s inside of us. We’re living in a world of connections—and it matters which ones get made and unmade”.
—Donna Haraway
Turning the Tide from Disruption to Educational Potential
Itwould be a bleak reality if technology offered only challenges and disruptions to education that somehow need to be accounted for. Yet more numerous are the voices that stress the potentials for improving—or rewiring—education in the 21st century, by the use of technology.
But there is little reason to believe that such improvements to education will happen merely as the result of the introduction of new technologies. Rather, they are likely to occur chiefly through synergies between the worlds of tech and education.
The rationale for such a strategic partnership is not difficult to see: tech industries are interested in tech-savvy workers and consumers; governments and supporters of education are interested in using tech to enrich learning in its widest sense.
Very likely, within decades, education can and will be entirely transformed by the use of new technologies. Early adopters will likely have an advantage over others. This path is, however, far from straightforward. It can be shaped in many different ways, and not only by the innovations and business interests of the tech and ed-tech (educational technology) industries, but just as much by how the education sector itself adapts, innovates, and builds alliances with tech. And while this is a dynamic that is hard to predict, it can certainly be shaped—or, at the very least, education can be designed more or less well to materialize technological potentials.
Educational design, on an institutional level, can thus affect the outcomes of technology. Some outcomes may be more desirable than others, or preferable in different ways, or have unwanted or unexpected side-effects. It is thus important to see what the main potentials are, so that they may be balanced and navigated. And, of course, they can be more or less well coordinated with the other seven pathways that are explored in this series of articles. It is time that this debate takes a front seat in educational design.
Big Data and AI to Solve the Problem of Scaling
Despite the best efforts over the last century, the educational systems are almost unanimously criticized by all observers as being too dry, too mechanical, too bland, too focused on quantitative results and measures (more on this last point later). They all say that children are playful and curious, but that schooling and education, at least partly, kill that spark. Univocally, commentators ask for a more humanized, sensitive, and person-centric form of education—but from thereon, unsurprisingly, the ideas and analyses begin to differ.
The question thus presents itself:
- If almost all commentators, educational science scholars, teachers, school principals, philosophers and psychologists of education agree that a more alive and engaging form of schooling and education are needed—why does the conventional schooling system persist?
This is an important question. Different answers to this question are possible, but here we would like to offer a simple, but strong explanation: Education faces a problem of scaling. It is no secret that individual students have different talents, interests, needs, and ways of learning. Yet, if you put children together in a class, and classes together in a school, and schools unified under one nationally defined curriculum, you are forced to design education for the average (or median) student. Individual teachers can make some adjustments to their different pupils, but most of them will after all attend the same lessons, use the same teaching materials, and do the same tests. Schooling systems can allow for some student autonomy, but even this is limited—not all classes can be electable, and too much autonomy without guidance can lose some of the scaffolding and support that conventional teaching offers.
It has even been compellingly argued by neuroscientist Erik Hoel, in his Substack article “Why we stopped making Einsteins”, that the geniuses of past generations all had one thing in common: growing up with personal tutoring. He lays out the unpopular argument that the life conditions of aristocracy, where scaling was simply not a problem, had a pedagogy so superior that it shifted the average learning results by two standard deviations. Said otherwise, 98% of pupils who receive personal tutoring have better results than pupils of conventional schooling—likely because they can use their time more efficiently and engage in an active dialogue of learning with a knowledgeable teacher. Teaching, in such settings, is much better tailored to the individual pupil.
In other words, whatever sensitivities, critiques, and skills may emerge through educational research, philosophy and practical experience, these all face the problem of scaling; when they are scaled up and used in a wider frame, they lose precision and context-sensitivity; they lose sight of the singular learning individual—and the relations of that unique person.
The very fact that education is designed for the many implies that it can never fully accommodate the uniqueness of all; because each person and their context is unique. This is one way of understanding why, again, in spite of an almost ubiquitous sense that education is too mechanical, education is and remains just that: too mechanical.
One bold idea that I have discussed with my interviewees is that AI and Big Data could be used in service of individualizing—and thereby sensitizing and humanizing—education. Imagine if a significant part of students’ progress would be stored (Big Data, this would result in enormous amounts of information) and if AI technology would then analyze these data and continuously improve algorithms that tailor unique learning programs for children, according to what is statistically most likely to yield results such as more time spent studying, better learning results, more flow states, lower stress levels and pressure, and greater general wellbeing and happiness. As Erik Hoel writes and references in his article:
Recent research has shown the two-sigma effect of tutoring using AI tutors compared to traditional online courses. Perhaps in the future once could imagine personalized AI governesses and AI tutors. But by then, will we even need human geniuses?
Through their engagement in studies, learners would hone their own algorithms, making them better and better at predicting favorable outcomes, while still feeding data into the overall system that would become better at predicting what pieces of information and what tasks could be offered to each person at each moment, adjusting the pace of learning, the preferred sources of information, and preferred learning styles.
This would open up for an educational system in which students branch off in a complex manner, while still being connected to a certain overall curriculum, each having their own version and appropriate user interface of that curriculum. Such a system could also match students for projects, optimizing for good relations and productive co-creative endeavors. The role of the teacher could, as many have envisioned in different forms, thus take on a more guiding and mentoring form. The teacher would certainly not be obsoleted—the AI could not fill the need for human connection, for many outdoor activities, and so on. Rather; the teacher could focus and specialize more on these competencies, while spending less time giving lessons to students who are alienated, bored, or overwhelmed. Teachers could likewise access data in the system and get a clearer overview of the real needs and wants of students.
If this sounds utopian, it’s probably because it is. Naturally, there can and will be problems and unexpected side-effects of such a daring endeavor: a dystopian future of education in which learners are isolated by their screens. But given that AI and Big Data are likely to transform society—and education with it—does it not make sense to think of how it can be made to humanize, rather than dehumanize and further mechanize, education? If intelligently and carefully applied, AI and Big Data could even reduce the sheer amount of screen-time that youths are exposed to—by effectively inspiring to taking part in tasks and projects that lie outside of the virtual realm. If that’s what it is optimized to be doing.
Perhaps, after all, the many hours spent online today may be a sign that children and youths are alienated in today’s schooling systems. Maybe, then, technology can be leveraged to win back some of the youth’s attention span, from entertainment, to a more entertaining and individualized education?
Modelling Reality
Another development of ed-tech that is being discussed has to do with making learning more experiential by means of offering tools in which different realities can be modelled in virtual worlds. In physics, many of the simpler formulas can easily fit into graphically attractive programs which would let students play around with input variables and try to see different outcomes. But it doesn’t have to stop there; chemistry, biology (particularly ecology and systems cycles)—and to some extent social science can all be modelled and played with in a similar manner, for instance, by modelling different ballot systems for democracy, or by modelling feedback loops in economic interactions, thus understanding how the basics of economic growth and trade function.
This could serve to let children pick up more intuitive understandings of what otherwise all-too-easily becomes just a dry formula that one forgets after taking the exam. Narratives are important to understanding—but so are playing and modelling for at least a modicum of lived experience and own creativity. Not only can this kind of ed-tech be used to enhance learning of important subjects; it can also train capacities of problem-solving, since there can be problems that need to be addressed through learning-by-doing in a multivariate (but predefined and limited) virtual setting.
Another aspect of such modelling-driven learning is that it offers a useful venue for introducing two important skills: computation and complexity thinking (or systems thinking, an integral part of ecological relatedness). Starting with computation, there is an inherent risk in societies so technologically advanced as today, that the majority of the population develop a great distance towards the technologies which nevertheless comprise so much of our everyday life experience. Hence, it makes sense that at least a simple literacy of computation and programming should be introduced—into an admittedly already too crowded curriculum. Now, if there is virtual modelling in subject after subject, this implies at least some very simple forms of programming, which then intuitively guides young minds to an understanding of the information technology around us.
When it comes to understanding complexity and systems—which we have already mentioned as an important aspect of developing environmental relatedness—the modelling systems can introduce, in subject after subject, the basics of such thinking: feedback loops, stable equilibria, sensitive initial conditions, the difference between non-linear and linear systems, closed and open systems, emergence of new properties, and so on. Such tools can then accessibly and intuitively be learned by large portions of the population, who are then guided to not only think in such terms within each subject matter, but use these same tools to make connections and innovations across the subjects.
In other words, used correctly, there is reason to believe that ed-tech can do more than increase the access to education (with impressive projects like Khan Academy, the many MOOCs and so on). Ed-tech, if allied with renewed practices of conventional education, can upgrade the cognitive functioning of whole populations, making each person more apt at grasping, sensing, and navigating in an increasingly complex world—all while humanizing education and making it suit each person better. The value gained by such advances can truly be immeasurable.
There are many more possibilities—the issue is to get practical and strategical about materializing them, and this must happen not only in the tech startup world, but in a broad alliance among key stakeholders taking a lead in the world of global education.
—
Hanzi Freinacht is a political philosopher, historian, and sociologist, author of ‘The Listening Society’, ‘Nordic Ideology’ and the upcoming books ‘The 6 Hidden Patterns of History’ and ‘Outcompeting Capitalism’. Much of his time is spent alone in the Swiss Alps. You can follow Hanzi on Facebook, Twitter, and Medium, and you can speed up the process of new metamodern content reaching the world by making a donation to Hanzi here.
