# Retention, not Attention

**Published by:** [vintro.eth](https://paragraph.com/@vintro-eth/)
**Published on:** 2021-12-11
**URL:** https://paragraph.com/@vintro-eth/retention-not-attention

## Content

Special thanks to Renee Maria Lee, Courtland Leer, sidcode, and dmarz for their thoughtful feedback. Attention is the sincerest form of generosity -- I’m so grateful for you all. We live in a world dominated by ad-driven media platforms. Given this unfortunate circumstance, I personally try to make the best of it by looking for ways to learn. However, the goal of these platforms is not my learning and growth, but rather my continued attention. So they flood me with information, and I recall very little of it. To remember anything on YouTube, for example, I have to put a lot of effort into thinking about what I'm viewing. I'll also need to go through videos again in the future to really hammer concepts home -- which is a lot of work! I don't do it very often, and I'm sure most other people don't either. This results in poor retention, which is crucial as individuals grow more aware of their information diets. We deserve better. Why don’t media platforms do more for us? Why should I have to figure out what key concepts to remember from the video? Isn't it possible to offload this responsibility to the creator? And who will check in on me to see if I'm still remembering those concepts? I’m lazy. I want to automate memorization! This has been a challenge because retention doesn’t work well for selling ads. Luckily, web3 provides us the opportunity to tackle this. We’ve never had something like blockchains that allow us to align community incentives around a token with financial rewards for positive behavior. Thus, instead of focusing on attention, consider what optimizing for retention might look like. Rather than attempting to hack dopamine receptors, we can build in mechanisms for effortless recall. Instead of creating media to keep viewers' eyes glued to a screen, we can create media that allows creators to explore learning and, more importantly, retention as an objective. Here’s how…Mnemonic MediaThe main inspiration for this article is an essay titled "How can we develop transformative tools for thought?" that I studied during my time in Kernel on an adventure to explore decentralized education solutions. The essay focuses on how we might develop new forms of media that aid in memory. The authors (Andy Matuschack and Michael Nielsen) ask excellent questions about this subject:How could you build a medium to better support a person’s memory of what they read? What interactions could easily and enjoyably help people consolidate memories? And, more broadly: is it possible to 2x what people remember? 10x? And would that make any long-term difference to their effectiveness?As an experiment, the authors created quantum.country, a web-based essay. Their goal was to provide a simple introduction to the complex concepts underlying quantum computing. How? They thoughtfully sprinkled flashcards throughout the paper. The only feedback they collect from users is whether or not they remembered the answer to the question. Seems simple right? It is – but there's more going on here... This technique is more officially referred to as "mnemonic media," an ode to the platform's ability to integrate a memorization mechanism. Not only are you gently reminded of the concepts you're reading about as you read, but if you sign up with your email address at the end of the essay, they will send you personalized reviews with questions at carefully spaced times in the future. The key to committing to long term memory is the spacing of these reviews. In 5 days, 2 weeks, 1 month, and 2 months, you'd get emails with questions to answer. The spaced repetitions are planned exponentially. It's a seemingly unimportant but remarkably efficient technique for hacking our memory systems for maximum retention. Just as you’re about to forget a concept, it’s resurfaced to jog your memory. How effective, though, is this spaced repetition? To summarize, the authors discovered that readers of quantum.country received a exponential return on their time investment. Readers retained each of the 112 questions for an average of 54 days. Let that sink in for a moment... People remembered the answers to 112 different questions about quantum computing for an average of 54 days per question. That's nearly two months! I don't know about you, but I'm lucky if I remember one or two things from an article I read two months ago. The benefits come in the form of delayed gratification, but I’d argue that’s a feature, not a bug.True learning requires a sincere desire to know; a determination to develop the discipline required to practice what you know every day; a passion that will not let you stop until you know; and a heart clear enough to know its own intention for pursuing knowledge at all. [1]Expanding to Mnemonic VideoAndy and Michael take the concept of mnemonic media a step further. The effectiveness of their essay's flashcard component was obvious, but text as a communication medium is limited in its ability to evoke emotion – which is critical for connecting with an audience. In this regard, video has a significant advantage. YouTube has been a fantastic playground for creators to experiment with emotional connection. Grant Sanderson from the channel 3blue1brown is one of the creators mentioned in the essay. He creates beautiful, world-class math videos. First and foremost, they're always interesting stories, with ups and downs and aha moments, and his enthusiasm for the subject is contagious. Most people leave with a greater fascination for the concepts, but they do not gain a deeper understanding of mathematics.To paraphrase Einstein, attaining a detailed understanding without forming an emotional connection is lame; while forming an emotional connection without detailed understanding has no enduring power.What if we could connect emotionally while also stimulating our memory? A positive emotional experience alone is not enough. We need to create a sense of importance for remembering the content. If we keep the high-affect narrative going, we can leverage the emotional connection with the audience to increase retention by embedding questions in the videos. The key here is to do so in a way that detracts as little as possible from the narrative. This combination of the emotional and intellectual is what makes mnemonic video so powerful. It enables the creation of performances of information that help people retain more of what they saw.The Role of Web3The lack of a business model is the primary reason transformative tools for thought aren't front and center in our fast-paced tech world. Tools for thought are public goods that do not fit neatly into the traditional world’s rent-seeking funding models. Fortunately for us, web3 provides a means of funding public goods. This concept is being actively developed with protocols such as FreeLearn, which aims to use DeFi as a yield-bearing back-end to reward teachers while only borrowing collateral from learners. Although it has not yet been deployed on mainnet, infrastructure like this is inevitable. Given this truth, it allows us to think much more freely about what new tools for thought might look like. Thus, mnemonic video as a tool for thought makes a lot of sense. Web3 is also suffering from a severe talent shortage. Many protocols need people to be educated in order to further support their ecosystems. We can use this type of media to assist more people in developing a strong foundational knowledge base in blockchain and web3. What better place to put mnemonic video to the test than web3 on-boarding?Success through GivingI indicated earlier that my ability to consume information is limited since I have to think about what I'm learning in order to retain it. This is why I am so enthusiastic about building a mnemonic video platform. By embedding questions inside a high-affect narrative, we basically automate the task of thinking about what you're watching. Attention is much more than a precious resource to be rationed for productivity. It is the definition of what it means to be truly present; the rarest and purest form of generosity; attentiveness in the present moment. This point of view is what enables us to value attention at a much higher level. In this light, it appears almost criminal that we freely give our attention away, just to be exploited for more. Why doesn't media give to us? Can't we engineer ways to add rather than extract? Finally, with web3 we have the ability to craft incentive structures that make tools for thought feasible. The days of free content and data monetization for ad impressions are numbered.The goal of web3 is to generate and distribute value widely and liberally. Give, give, give. Consume less of the world, produce more joy. That's the difference between web2 and web3. That's how we win. [2]We're going to build this video platform and get started making mnemonic videos. Please join our Discord if you want to take part in this media experiment. If you’re associated with a project/protocol and are interested in creating a series of mnemonic videos, we have a channel for that in the Discord! All are welcome – developers (front end and back end), teachers, instructional designers, video creators, actors, producers, directors, playwrights, artists, community builders, web3 degens – we need diversity in order to address the emotional and intellectual components of this project. We also have a Gitcoin Grant live for our decentralized schooling effort. Mnemonic Video falls under the umbrella of things that will be supported on this grant, so if you would like to fund the development of this platform, consider donating to our cause! We intend to create a minimum viable product (MVP) and seek additional funding. Let’s make better media – for everyone.References:[1]: https://kernel.community/en/blog/community/free-learn/ [2]: https://twitter.com/v_stickykeys/status/1468339623126614019

## Publication Information

- [vintro.eth](https://paragraph.com/@vintro-eth/): Publication homepage
- [All Posts](https://paragraph.com/@vintro-eth/): More posts from this publication
- [RSS Feed](https://api.paragraph.com/blogs/rss/@vintro-eth): Subscribe to updates