While an educator may claim a theory-free approach to their practice, this isn't truly possible. Every instructional strategy fundamentally embodies a theory of human learning. #LearningTheory #TeachingStrategies
#LearningTheory
Why YouTube is obsolete: From linear video content consumption to AI-mediated multimodal knowledge production
Does the educational purpose of video change with AI?
The purpose of video in education is undergoing a fundamental transformation in the age of artificial intelligence. This medium, long established in digital learning environments, is changing not just in how we consume it, but in its very role within the learning process.
Video has always been a problem in education
Video has always presented significant challenges in educational contexts. Its linear format makes it difficult to skim or scan content. Unlike text, which allows learners to quickly jump between sections, glance at headings, or scan for key information, video requires sequential consumption. This constraint has long been problematic for effective learning.
Furthermore, in many regions where our learners are based, internet access remains expensive, unreliable, or limited. Downloading or streaming video content can be prohibitively costly in terms of both data usage and time. The result is straightforward: few learners will watch educational videos, regardless of their potential value.
The bandwidth and attention divide
This reality creates a significant divide in educational access. While instructional designers and educators in high-resource settings continue to produce video-heavy content, learners in bandwidth-constrained environments have been systematically excluded from these resources. Even when videos are technically accessible, the time investment required to watch linear content often exceeds what busy professionals can allocate to learning activities.
Emergent AI platforms are scanning YouTube video transcripts to extract precisely what users need. This capability suggests a transformation for the role of video. YouTube and other video platforms are evolving into what might be called “interstitial processors”, mediating layers that support knowledge production and dissemination for subsequent extraction and analysis by both humans and machines.
A more inclusive workflow for knowledge extraction
This changing relationship with video content could enable more inclusive approaches to learning. When I discover a potentially valuable educational webinar, I now follow a structured approach to maximize efficiency and accessibility:
- Download the video file.
- Transcribe it using Whisper AI technology.
- Ask targeted questions to extract meaningful insights from the transcript.
- Request direct quotes as evidence of key points.
This method circumvents the traditional requirement to invest 60 minutes or more in viewing content that may ultimately offer limited value. More importantly, it transforms bandwidth-heavy video into lightweight text that can be accessed, searched, and processed even in low-connectivity environments.
I suspect that it is no accident that YouTube has recently placed additional restrictions on downloading videos from its platform.
Bridging the resource gap with AI
Current consumer-grade AI systems like Claude.ai have limitations: they cannot yet process full videos directly. For now, we are restricted to text-based interactions with video content, hence my transcription of downloaded content. However, this constraint will likely dissolve as AI capabilities continue to advance.
The immediate benefit is that this approach can help bridge the resource gap that has disadvantaged learners in bandwidth-constrained environments. By extracting the knowledge essence from videos, we could make educational content more accessible and equitable across diverse learning contexts.
The continuing value of educational video production
Despite these challenges, educational video production continues to be a relevant method for humans and machines that need a way to share what they know. Hence, what we are witnessing is not the diminishing relevance of educational video, but rather a transformation in how its knowledge value is extracted and utilized. The production of video content remains valuable. It is our methods of processing and consuming it that are evolving.
Aligning with effective networked learning theory
This shift aligns with contemporary understanding of effective learning. Research consistently demonstrates that passive consumption of information, whether through video or text, remains insufficient for meaningful learning. Genuine knowledge development emerges through active construction – the processes of questioning, connecting, applying, and adapting information within broader contexts.
The AI-enabled extraction of insights from video content represents a step toward more active engagement with educational materials – transforming passive viewing into targeted interaction with the specific knowledge elements most relevant to individual learning needs.
Knowledge networks trump media formats
Our experience with global learning networks demonstrates the importance of moving beyond media format limitations. When health professionals from diverse contexts share practices and adapt them to their specific environments, the medium of exchange becomes secondary to the knowledge being constructed.
AI tools that can extract and process information from videos help overcome the medium’s inherent limitations, turning static content into formats that can not only be read, viewed, or listened to – but that can also be remixed and fused with other sources. This approach allows learners to engage more directly with knowledge, freed from the constraints of linear consumption and bandwidth requirements.
Rethinking video as a dual-purpose knowledge production format
We are witnessing the development of new approaches to educational content where media exists simultaneously for direct human consumption and as structured data for AI processing. When the boundaries between content formats become increasingly permeable, with value residing not in the medium itself but in the knowledge that can be extracted and constructed from it.
Despite the consumption challenges, video remains an exceptional medium for content production that serves both humans and machines. For content creators, video offers unmatched richness in communicating complex ideas through visual demonstration, tone, and emotional connection.
What is emerging is not a devaluation of video creation but a transformation in how its knowledge is accessed. As AI tools evolve, video becomes increasingly valuable as a comprehensive knowledge repository where information is encoded in multiple dimensions – visual, auditory, and textual through transcripts.
This makes video uniquely positioned as a “dual-purpose” content format: rich and engaging for those who can consume it directly, while simultaneously serving as a structured data source from which AI can extract targeted insights.
In this paradigm, video production remains vital while consumption patterns evolve toward more efficient, personalized knowledge extraction.
The creator’s effort in producing quality video content now yields value across multiple consumption pathways rather than being limited to linear viewing
How to cite this article: Sadki, R. (2025). Why YouTube is obsolete: From linear video content consumption to AI-mediated multimodal knowledge production. Learning to make a difference. https://doi.org/10.59350/rfr2z-h4y93
References
Delello, J.A., Watters, J.B., Garcia-Lopez, A., 2024. Artificial Intelligence in Education: Transforming Learning and Teaching, in: Delello, J.A., McWhorter, R.R. (Eds.), Advances in Business Information Systems and Analytics. IGI Global, pp. 1–26. https://doi.org/10.4018/979-8-3693-3003-6.ch001
Guo, P.J., Kim, J., Rubin, R., 2014. How video production affects student engagement: An empirical study of MOOC videos, in: Proceedings of the First ACM Conference on Learning@ Scale Conference. ACM, pp. 41–50. https://doi.org/10.1145/2556325.2566239
Hansch, A., Hillers, L., McConachie, K., Newman, C., Schildhauer, T., Schmidt, P., 2015. Video and Online Learning: Critical Reflections and Findings from the Field. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.2577882
Kumar, L., Singh, D.K., Ansari, M.A., 2024. Role of Video Content Generation in Education Systems Using Generative AI:, in: Doshi, R., Dadhich, M., Poddar, S., Hiran, K.K. (Eds.), Advances in Educational Technologies and Instructional Design. IGI Global, pp. 341–355. https://doi.org/10.4018/979-8-3693-2440-0.ch019
Mayer, R.E., Fiorella, L., Stull, A., 2020. Five ways to increase the effectiveness of instructional video. Education Tech Research Dev 68, 837–852. https://doi.org/10.1007/s11423-020-09749-6
Netland, T., Von Dzengelevski, O., Tesch, K., Kwasnitschka, D., 2025. Comparing human-made and AI-generated teaching videos: An experimental study on learning effects. Computers & Education 224, 105164. https://doi.org/10.1016/j.compedu.2024.105164
Salomon, G., 1984. Television is “easy” and print is “tough”: The differential investment of mental effort in learning as a function of perceptions and attributions. Journal of Educational Psychology 76, 647–658. https://doi.org/10.1037/0022-0663.76.4.647
Sun, M., 2024. An Intelligent Retrieval Method for Audio and Video Content: Deep Learning Technology Based on Artificial Intelligence. IEEE Access 12, 123430–123446. https://doi.org/10.1109/ACCESS.2024.3450920
Image: The Geneva Learning Foundation Collection © 2025
#ArtificialIntelligence #knowledgeConstruction #knowledgeConsumption #knowledgeTheory #learningTheory #linearContent #video #YouTube
@h4ckernews
Wow! Are MIT press giving this PDF textbook away free?
#learningTheory #statistics #maths #math #mathematics #machineLearning #AI
Learning Theory from First Principles [pdf]
https://www.di.ens.fr/~fbach/ltfp_book.pdf
#HackerNews #LearningTheory #FirstPrinciples #Education #PDF #HackerNews
Unlocking the Mind: The Power of Evidence-Based Psychology
#Psychology #MentalHealth #EvidenceBased #Mindfulness #CognitivePsychology #Psychotherapy #Wellbeing #Neuroscience #LearningTheory #MindfulnessInterventions #HumanBehavior #MentalHealthAwareness #PsychologyFacts #GrowthMindset #Exploration
In a rural health center in Kenya, a community health worker develops an innovative approach to reaching families who have been hesitant about vaccination.
Meanwhile, in a Brazilian city, a nurse has gotten everyone involved – including families and communities – onboard to integrate information about HPV vaccination into cervical cancer screening.
These valuable insights might once have remained isolated, their potential impact limited to their immediate contexts.
But through Teach to Reach – a peer learning platform, network, and community hosted by The Geneva Learning Foundation – these experiences become part of a larger tapestry of knowledge that transforms how health workers learn and adapt their practices worldwide.
Since January 2021, the event series has grown to connect over 21,000 health professionals from more than 70 countries, reaching its tenth edition with 21,398 participants in June 2024.
Scale matters, but this level of engagement begs the question: how and why does it work?
The challenge in global health is not just about what people need to learn – it is about reimagining how learning happens and gets applied in complex, rapidly-changing environments to improve performance, improve health outcomes, and prepare the next generation of leaders.
Traditional approaches to professional development, built around expert-led training and top-down knowledge transfer, often fail to create lasting change.
They tend to ignore the rich knowledge that exists in practice – what we know when we are there every day, side-by-side with the community we serve – and the complex ways that learning actually occurs in professional networks and communities.
Teach to Reach is one component in The Geneva Learning Foundation’s emergent model for learning and change.
This article describes the pedagogical patterns that Teach to Reach brings to life.
A new vision for digital-first, networked professional learning
Teach to Reach represents a shift in how we think about professional learning in global health.
Its pedagogical pattern draws from three complementary theoretical frameworks that together create a more complete understanding of how professionals learn and how that learning translates into improved practice.
At its foundation lies Bill Cope’s and Mary Kalantzis’s New Learning framework, which recognizes that knowledge creation in the digital age requires new approaches to learning and assessment.
Teach to Reach then integrates insights from Watkins and Marsick’s research on the strong relationship between learning culture (a measure of the capacity for change) and performance and George Siemens’s learning theory of connectivism to create something syncretic: a learning approach that simultaneously builds individual capability, organizational capacity, and network strength.
Active knowledge making
The prevailing model of professional development often treats learners as empty vessels to be filled with expert knowledge.
Drawing from constructivist learning theory, it positions health workers as knowledge creators rather than passive recipients.
When a community health worker in Kenya shares how they’ve adapted vaccination strategies for remote communities, they are not just describing their work – they’re creating valuable knowledge that others can learn from and adapt.
The role of experts is even more significant in this model: experts become “Guides on the side”, listening to challenges and their contexts to identify what expert knowledge is most likely to be useful to a specific challenge and context.
(This is the oft-neglected “downstream” to the “upstream” work that goes into the creation of global guidelines.)
This principle manifests in how questions are framed.
Instead of asking “What should you do when faced with vaccine hesitancy?” Teach to Reach asks “Tell us about a time when you successfully addressed vaccine hesitancy in your community.” This subtle shift transforms the learning dynamic from theoretical to practical, from passive to active.
Collaborative intelligence
The concept of collaborative intelligence, inspired by social learning theory, recognizes that knowledge in complex fields like global health is distributed across many individuals and contexts.
No single expert or institution holds all the answers.
By creating structures for health workers to share and learn from each other’s experiences, Teach to Reach taps into what cognitive scientists call “distributed cognition” – the idea that knowledge and understanding emerge from networks of people rather than individual minds.
This plays out practically in how experiences are shared and synthesized.
When a nurse in Brazil shares their approach to integrating COVID-19 vaccination with routine immunization, their experience becomes part of a larger tapestry of knowledge that includes perspectives from diverse contexts and roles.
Metacognitive reflection
Metacognition – thinking about thinking – is crucial for professional development, yet it is often overlooked in traditional training.
Teach to Reach deliberately builds in opportunities for metacognitive reflection through its question design and response framework.
When participants share experiences, they are prompted not just to describe what happened, but to analyze why they made certain decisions and what they learned from the experience.
This reflective practice helps health workers develop deeper understanding of their own practice and decision-making processes.
It transforms individual experiences into learning opportunities that benefit both the sharer and the wider community.
Recursive feedback
Learning is not linear – it is a cyclical process of sharing, reflecting, applying, and refining.
Teach to Reach’s model of recursive feedback, inspired by systems thinking, creates multiple opportunities for participants to engage with and build upon each other’s experiences.
This goes beyond communities of practice, because the community component is part of a broader, dynamic and ongoing process.
Executing a complex pedagogical pattern
The pedagogical pattern of Teach to Reach come to life through a carefully designed implementation framework over a six-month period, before, during, and after the live event.
This extended timeframe is not arbitrary – it is based on research showing that sustained engagement over time leads to deeper learning and more lasting change than one-off learning events.
The core of the learning process is the Teach to Reach Questions – weekly prompts that guide participants through progressively more complex reflection and sharing.
These questions are crafted to elicit not just information, but insight and understanding.
They follow a deliberate sequence that moves from description to analysis to reflection to application, mirroring the natural cycle of experiential learning.
Communication as pedagogy
In Teach to Reach, communication is not just about delivering information – it is an integral part of the learning process.
The model uses what scholars call “pedagogical communication” – communication designed specifically to facilitate learning.
This manifests in several ways:
- Personal and warm tone that creates psychological safety for sharing
- Clear calls to action that guide participants through the learning process
- Multiple touchpoints that reinforce learning and maintain engagement
- Progressive engagement that builds complexity gradually
Learning culture and performance
Watkins and Marsick’s work helps us understand why Teach to Reach’s approach is so effective.
Learning culture – the set of organizational values, practices, and systems that support continuous learning – is crucial for translating individual insights into improved organizational performance.
Teach to Reach deliberately builds elements of strong learning cultures into its design.
Furthermore, the Geneva Learning Foundation’s research found that continuous learning is the weakest dimension of learning culture in immunization – and probably global health.
Hence, Teach to Reach itself provides a mechanism to strengthen specifically this dimension.
Take the simple act of asking questions about real work experiences.
This is not just about gathering information – it’s about creating what Watkins and Marsick call “inquiry and dialogue,” a fundamental dimension of learning organizations.
When health workers share their experiences, they are not just describing what happened.
They are engaging in a form of collaborative inquiry that helps everyone involved develop deeper understanding.
Networks of knowledge
George Siemens’s connectivism theory provides another crucial lens for understanding Teach to Reach’s effectiveness.
In today’s world, knowledge is not just what is in our heads – it is distributed across networks of people and resources.
Teach to Reach creates and strengthens these networks through its unique approach to asynchronous peer learning.
The process begins with carefully designed questions that prompt health workers to share specific experiences.
But it does not stop there.
These experiences become nodes in a growing network of knowledge, connected through themes, challenges, and solutions.
When a health worker in India reads about how a colleague in Nigeria addressed a particular challenge, they are not just learning about one solution – they are becoming part of a network that makes everyone’s practice stronger.
From theory to practice
What makes Teach to Reach particularly powerful is how it fuses multiple theories of learning into a practical model that works in real-world conditions.
The model recognizes that learning must be accessible to health workers dealing with limited connectivity, heavy workloads, and diverse linguistic and cultural contexts.
New Learning’s emphasis on multimodal meaning-making supports the use of multiple communication channels ensuring accessibility.
Learning culture principles guide the creation of supportive structures that make continuous learning possible even in challenging conditions.
Connectivist insights inform how knowledge is shared and distributed across the network.
Creating sustainable change
The real test of any learning approach is whether it creates sustainable change in practice.
By simultaneously building individual capability, organizational capacity, and network strength, it creates the conditions for continuous improvement and adaptation.
Health workers do not just learn new approaches – they develop the capacity to learn continuously from their own experience and the experiences of others.
Organizations do not just gain new knowledge – they develop stronger learning cultures that support ongoing innovation.
And the broader health system gains not just a collection of good practices, but a living network of practitioners who continue to learn and adapt together.
Looking forward
As global health challenges have become more complex, the need for more effective approaches to professional learning becomes more urgent.
Teach to Reach’s pedagogical model, grounded in complementary theoretical frameworks and proven in practice, offers valuable insights for anyone interested in creating impactful professional learning experiences.
The model suggests that effective professional learning in complex fields like global health requires more than just good content or engaging delivery.
It requires careful attention to how learning cultures are built, how networks are strengthened, and how individual learning connects to organizational and system performance.
Most importantly, it reminds us that the most powerful learning often happens not through traditional training but through thoughtfully structured opportunities for professionals to learn from and with each other.
In this way, Teach to Reach is a demonstration of what becomes possible when we reimagine how professional learning happens in service of better health outcomes worldwide.
Image: The Geneva Learning Foundation Collection © 2024
Share this:
https://redasadki.me/2024/10/30/what-is-the-pedagogy-of-teach-to-reach/
#continuousLearning #globalHealth #learningCulture #learningStrategy #learningTheory #pedagogicalPatterns #peerLearning #TeachToReach #TheGenevaLearningFoundation
Why the Godfather of A.I. Fears What He’s Built
https://www.newyorker.com/magazine/2023/11/20/geoffrey-hinton-profile-ai #AI #NeuralNetworks #Connectionism #LearningTheory #Backpropagation #Feelings
Now in stock! Demonstrating how #learning theories are applicable to a variety of real-world contexts, this book will help #library workers better understand how people learn so that they can improve support for #instruction on their campuses and in their communities. https://www.alastore.ala.org/learningtheory
I’ve just read this paper by Karen Becker and Adelle Bish in which they offer this framework of the #onboarding of new employees in #organizations from a #LearningTheory perspective. I find their arguments about the relevance of the concept of #unlearning in this context particularly suggestive.
https://www.sciencedirect.com/science/article/abs/pii/S105348221930110X
#Introdiction : I'm a mathematician moving towards data science topics, based in Chile. Some interests:
#geometry #datageometry #discretegeometry #materialscience
#hyperbolicneuralnetwork
#GraphNeuralNetwork #GNNs
#topology #TDA #metricspaces
#equivariantneuralnetwork #enns #convolutionalneuralnetwork #networkscience
#computationalchemistry #dft #pinns
#optimaltransport
#learningTheory #generalization
#expressivity
#marinescience #biodiversity #oceanscience
#phylogenetics
Tong Zhang has a book on the tools of #LearningTheory (theoretical analysis of why #MachineLearning algorithms work). If you want to learn about concentration inequalities, generalization error, regret bounds, and more, have a look.
Book freely downloadable from http://www.tongzhang-ml.org/lt-book.html
I've been geeking out lately re-reading some of my favorite Learning Theory stuff. Vygotsky, Cole, Wertsch, Bakhtin, Marx, Engestrom...
I gave a talk today to a group of faculty from a variety of disciplines. I knew it couldn't be theory-heavy, but I made sure to build it into the framing and also wrap back in the claims and conclusions.
We know a lot about teaching and learning, and our work should build on good Theory.
#LearningTheory #Cognition #Mediation #Research #Teaching #Learning
Learning Theory book by Tong Zhang http://tongzhang-ml.org/lt-book.html #machinelearning #learningtheory
A #thread of recent papers coauthored by Eörs Szathmáry on learning and evolution, covering Darwinian principles in the immune system, brain, Bayesian learning theory.
Mentioned in today's #KLIAustria #KLIColloquium talk.
#evolution #ecoevo #EvolutionaryBiology #EvolutionaryAlgorithm #Darwin #Darwinian #Learning #LearningTheory #Bayesian #neuroscience #immunology #PaperThread #
Going live at 9am PST / 17:00 UTC! Teach Jenn Tech and I talk #python, #LearningTheory, and general tech things.
I got a suggestion to use hashtags to help kickstart my network here. So here's a hashtag-dense list of my interests:
I'm a data-focused #machinelearning engineer at #Apple, working with #PyTorch, #Spark, and #Python which I write with a #Haskell accent. I used to work in #LearningTheory at the ANU.
I enjoy long distance #running and #trailrunning and recently finished my first #ultramarathon.
In my spare time I make electronic #modularsynth / #eurorack music and play #jazz #piano and #bass.
Client Info
Server: https://mastodon.social
Version: 2025.04
Repository: https://github.com/cyevgeniy/lmst