Stephen Downes

Knowledge, Learning, Community

Aug 25, 2017

This Proceedings Article published as Becoming Connected in Konferencja Pokazać – Przekazać, Centrum Nauki Kopernik Aug 25, 2017. (Authored for proceedings - publication not verified) [Link] [Info] [List all Publications]

Introduction

What is science? What is learning? These are questions that form the core of whatever it is we know about and can discover about ourselves and the world. They define not only who we think we are, but also the limits (and possibilities) of what we could be.

This article considers these questions from the perspective of our understanding that we are linked to everyone else and everything else. It describes a world of knowledge and learning not in terms of facts and theories and arguments, but in the way we shape these connections in ourselves and our society by working with, and within, the environment as a whole.

What we find is not only a new understanding of knowledge and learning, but also a new understanding of ourselves and our place in the world.

What is Science?

Science is about the world. That, at least, is the traditional conception of science. The objective of scientific research is to find laws of nature and universal truths. It begins with observation and experience and describes the world, objects in the world and the properties of those objects. Then, using the analytical principles of logic and mathematics, researchers formulate general principles. These principles which begin as hypotheses, the hypotheses are used to make predictions, and if the predictions prove to be reliable (based again on observation and experience) then the principles may be elevated to the status of 'scientific theories' and 'laws of nature'.

The traditional conception has been successfully challenged by scientists and philosophers. The roles of empirical observation and abstract theory were not what we believed them to be. In particular, two dogmas proved to be false. First, the idea that all statements in science are based solely on experience and observation is unsustainable. Science, in some important ways, goes beyond experience. And second, the distinction between observation and theory itself cannot be maintained. All statements of observation contain some theory, leading some to criticize the traditional conception for depending on "theory-laden data".

In response to this challenge researchers and philosophers advanced the idea of 'science as construction'. On this model, the core activity of science is to build a model or theory, and this model or theory is evaluated based on whether it is useful, coherent and productive. The history of science was conceived as a series of 'paradigms'. Each paradigm would define, in its own way, the objects and properties it describes. Theories would be tested, as a whole, against experience. And they could be regarded as different perspectives, or different "lenses", through which to view the world.

For many people, however, this depiction of science as construction is unsatisfying. Science derives its authority not from the creativity of its constructions but from its basis in observation. For an individual scientist, the foundation of science is experience, immersion and practice. Becoming a scientist is essentially the act of seeing the world like other scientists, by participating in the scientific community, and learning what questions are important, what counts as evidence, and how to evaluate evidence. To learn a science is, in an important way, to become a scientist. And whatever scientists do: that's science.

What is Connectivism?

To understand how science works as a community, and to understand how scientists become scientists, we need to take a detour through a learning theory called connectivism.

According to connectivism, 'knowledge' is the set of connections between entities in a network, and 'learning' is the creation and development of those connections over time. Connectivism is based in part on a branch of computer science called connectionism, which uses artificial neural networks to generate artificial intelligence. And it is based on social network theory, which describes how ideas are created and spread through a society. The principle is that networks can contain knowledge, and that those networks by growing and shaping themselves in response to input – that is, to experience.

Traditional conceptions of science depict knowledge as a set of statements. Theories are thus thought of as models constructed from statements, where these expressions in mathematics or scientific language represent states of affairs in the world. In a network, however, what constitutes knowledge is a pattern of connectivity. There are no sentences or theories, only the connections. In a person, these connections are between neurons in the brain, connected by axons in a neural network. In a society, these connections are between the people and things that interact with each other through community, commerce and culture.

Though it draws from other sciences, connectivism was developed in the context of educational theory. Over the last ten years, it has been used as a model to inform the development of online learning environments, and in particular massive open online courses (MOOCs). The difference between traditional and connectivist conceptions of science and education can be seen in the design of online courses. The traditional course looks like a book, organized into chapters, and read in order from beginning to end. The connectivist course is designed as a network, where instead of reading participants interact with each other. Connectivist students don't focus on learning content, they focus on working with it, shaping and reshaping it, in an environment of research and inquiry.

Connectivist networks are self-organizing. They don't need to be taught or instructed in order to learn. This is a phenomenon we can observe in many places in nature. But not all networks are self-organizing. It depends on the structure or properties of the network in question. We can identify four key principles these networks – let's call them 'learning networks' – must embody. First, entities within them must be autonomous, defining its own activities in its own way. Second, and related, the entities in a network must be diverse. Each entity can be said to represent a distinct point of view or a distinct piece of evidence. Third, the network must be open. It needs interaction from outside the network and it needs to allow communication between entities in the network. Finally, it needs to be interactive. No entity is in charge. Content can originate from anywhere within the network.

The knowledge in a network is not contained in the signals sent from one entity to the other. Nor is the knowledge contained in sentences or models or theories constructed from those symbols. Instead, the knowledge is said to emerge from the network. It is an emergent property of the network. The knowledge is found in the pattern of connectivity that results over time as the entities in the network communicate with each other and form and shape connections with each other.

What is Learning?

We have reached a point of decision: whether to think of learning as based in construction, or to think of learning as based in experience. If construction, then we view learning as something we make using language, logic and mathematics, possibly according to a plan, such as a representation or a model. Alternatively, if experience, then we view learning as something we become as a result of these experiences, as through immersion in an environment we are shaped by our senses, observations and practices.

We can distinguish between these two alternatives by distinguishing between what might be called personalized learning, and compared to what may be called personal learning.

Personalized learning is defined by the content we want to learn. Our full acquisition of this content defines the ideal state we want to achieve. This idea state creates a set of requirements we must fulfill. Our satisfaction of these requirements is measured by grades and testing where we are evaluated by the teacher or instructional system. The gaps that remain in our attainment become the new content we need to learn, thus achieving the 'personalization' of personalized learning.

By contract, personal learning is defined by what we want to do, where the completion of the task represents a desired state. We attempt to achieve the desired state through a set of actions, or practice, in which we interact with the environment we want to change. This interaction represents an opportunity for growth or change or development. The role of the teacher or instructional system is to support our practice as we iteratively develop expertise in the domain or discipline.

Connectivist learning is personal learning. Instead of manipulating statements using logic and mathematics, connectivist networks learn through processes of association. That is, they learn through the physical processes that cause one node in a network to be connected to another node, and the processes that make these connections stronger or weaker. These principles are described in a large body of literature called 'learning theory'. Some of these principles are simple: Hebbian associationism, for example, suggests that similarity causes entities to link to each other. The principle of back propagation describes how feedback in artificial neural networks can be used to modify connections.

We can see the difference between personalized and personal learning through the differing descriptions of their outcomes. In personalized learning, which is based in content, the objective is to remember some model, representation or theory. By contrast, in personal learning, which is based in practice, the objective is to recognize and respond to relevant states of affairs or phenomenon.

Implications for Practice

Learning requires immersion in, and interaction with, authentic environments. To learn is to attempt to achieve an outcome (to solve a problem, achieve a goal, complete a project) through an iterative series of attempts to achieve that outcome, supported as needed by models and demonstrations of successful practice. The outcome is achieved by the way the network reshapes itself in response to the environment.

Success in learning therefore requires successful networks. These are networks that can adjust to the environment, but which are not fooled by temporary changes or deceptions. The networks that function best as networks are those composed of autonomous and diverse individuals, where the network is open to change and input, and where the network responds as a whole through the interaction between these individuals with each other in response to that input. It is this interaction that produces emergent properties in the network.

For students, the learning process is akin to functioning as a node in that network. It is based on interactions with other entities in the network, including people, ideas, objects and events. There are four major stages, or four major types of activity: aggregation, in which we open ourselves and actively seek out new ideas and experiences; remix, in which we filter and curate the most significant of these and combine them in various ways; repurpose, in which we shape or form or meld these to practical effect or purpose; and feed forward, in which we share our creative output with others.

There is nothing new or special about this process; it is a physical process we see replicated in throughout the world, in human bodies, in chemical reactions, in environments and ecosystems. And it is also the process that describes, in very general terms, such cognitive phenomena as creativity, imagination and invention. We stand on the shoulders of others, as it is aid, and see their world through new eyes.

What is it like, to learn in this way? We might think of it as like a new literacy, a way of reading the world, seeing the patterns, understanding what they mean, putting them in context, using them to achieve our goals, predicting consequences and observing change. These are the new critical literacies, and they include: recognizing patterns, similarities and syntax; relating these to our values, objectives and goals; combining then to see how they stand in relation to each other in an environment; using them in a productive and effective way; extending them though analysis and explanation, and comprehending them through interactions with the world.

These are practices that, for teachers and students, take us beyond the content-focused domain of the traditional classroom, and teach us through engagement with the world and with each other. We regard knowledge and learning from the perspective of what we can do, not what we can know. We see ourselves and independent and autonomous individuals, drawing our own conclusions and thinking for ourselves, but also bound to, and inextricably linked with, the rest of society, and the world as a whole.

 

More Reading

What is Science?

W.V.O. Quine, 1951, Two Dogmas of Empiricism. http://www.ditext.com/quine/quine.html

Norwood Russell Hanson, 1958 , Patterns of Discovery. https://www.amazon.ca/Patterns-Discovery-Inquiry-Conceptual-Foundations/dp/0521092612

Thomas Kuhn, 1970, The Structure of Scientific Revolutions. http://projektintegracija.pravo.hr/_download/repository/Kuhn_Structure_of_Scientific_Revolutions.pdf

What is Connectivism?

Albert-laszlo Barabasi, 2002, Linked: The New Science of Networks. https://www.amazon.com/Linked-Science-Networks-Albert-laszlo-Barabasi/dp/0738206679

George Siemens, 2004, Connectivism: A Learning Theory for the Digital Age. http://www.elearnspace.org/Articles/connectivism.htm

Stephen Downes, 2005. An Introduction to Connective Knowledge. http://www.downes.ca/post/33034

What is Learning?

David Jonassen, 1997, Instructional Design Models for Well-Structured and Ill-Structured Problem-Solving Learning Outcomes. https://link.springer.com/article/10.1007%2FBF02299613

2005ff. Constructivist Foundations. http://www.univie.ac.at/constructivism/journal/1/1

Will Richardson and Rob Mancabelli, 2011, Personal Learning Networks: Using the Power of Connections to Transform Education. https://www.amazon.ca/Personal-Learning-Networks-Connections-Transform/dp/193554327X

Implications for Practice

Seymour Papert, 1993. Mindstorms: Children, Computers and Powerful Ideas. https://www.amazon.ca/Mindstorms-Children-Computers-Powerful-Ideas/dp/0465046746

Paulo Freire & Donaldo Macedo, 1987, Literacy: Reading the Word and the World. https://www.amazon.ca/Literacy-Reading-World-Donaldo-Macedo/dp/0897891260

Ivan Illich, 2000, Deschooling Society. https://www.amazon.ca/Deschooling-Society-Open-Forum-Illich/dp/0714508799

Mentions

- Two Dogmas of Empiricism, Jun 20, 2019
, - Patterns of Discovery: An Inquiry into the Conceptual Foundations of Science: Hanson, Norwood Russell: 9780521092616: Books - Amazon.ca, Mar 06, 2024
, - , Jun 20, 2019
, - Linked: The New Science Of Networks Science Of Networks: Albert-laszlo Barabasi, Jennifer Frangos: 9780738206677: Amazon.com: Books, Jun 20, 2019
, - , Jun 20, 2019
, - Stephen's Web ~ An Introduction to Connective Knowledge ~ Stephen Downes, Jun 20, 2019
, - Instructional design models for well-structured and III-structured problem-solving learning outcomes | SpringerLink, Jun 20, 2019
, - Constructivist Foundations, Jun 20, 2019
, - Personal Learning Networks: Using the Power of Connections to Transform Education: Will Richardson, Rob Mancabelli: 9781935543275: Books - Amazon.ca, Jun 20, 2019
, - Mindstorms: Children, Computers, And Powerful Ideas: Amazon.ca: Seymour A. Papert: Books, Jun 20, 2019
, - Literacy: Reading the Word and the World: Donaldo Macedo: 9780897891264: Books - Amazon.ca, Jun 20, 2019
, - Deschooling Society: Ivan Illich: 8601300388793: Books - Amazon.ca, Jun 20, 2019
, - , Jun 20, 2019



Stephen Downes Stephen Downes, Casselman, Canada
stephen@downes.ca

Copyright 2024
Last Updated: Mar 29, 2024 04:21 a.m.

Canadian Flag Creative Commons License.

Force:yes