Frequently Asked Questions

Going above and beyond standard spaced repetition algorithms, we also collect qualitative feedback.

Example If you answer several questions on Cars correctly, but are bad at questions on Motorbikes, Revision.ai will reflect and choose questions to identify your misconceptions, like focusing a question on Wheels.

Once it identifies a possible misconception, it will create questions to overcome the mistake and represent this in the percentage completeness of your lecture!

You can see the systems "Third Brain" score-card and list of your strengths, mistakes and misconceptions in addition to normal progress.

This way, you can jump right into quizzes on your mistake topics, show them or discuss them with others, and reflect yourself on your learning. You can also add manual text input on your own thoughts on your areas of mistakes/reflections, to direct the AI your way too.

One of the brilliant ways Revision.ai helps you be a better student than otherwise is that it can see patterns in things you might not understand.

For example, imagine during a long lecture the following pictures are present, as in Open Stax Biology lectures:

These are regions of the brain, and the region at the back of the brain is important for vision.

If you struggle with answering questions about regions of the brain, but succeed at labeling the visual system, the "Third Brain" will create new text questions asking you to choose the right description for regions in the brain and it will provide you a short synopsis of the regions first to recap them.

So this way, even though the original lecture assumed you knew about Brain Regions - in your case an untrue assumption which held back deeper learning - now, with Revision.ai, you can accomplish that foundational step to build the groundwork for learning about specific regions, or reasoning about how regions interact!

This is the power of the both reflective, learning record and text based analysis that the Third Brain does, when constantly providing a fresh supply of New, Relevant, Answerable questions as you study.

By generating a comprehensive set of questions that cover each part of your slides and keeping a record of your knowledge, the system can constantly provide you fresh, relevant, exam-conditions-like questions as you go up into completing the whole lecture and completing it increasingly accurately as you study.

Until you have consistently covered 80% of the lecture with 80% accuracy, the system always adapts and highlights to show you the next exercise you should study - by doing this, you do not have to think about it or struggle with analysis paralysis - you can just jump into conquering a small part of your lecture one after another to succeed!

The Trophy system (provided only to skilled successful recalls of information from your slides) provides an accurate judgement of passing exams at US college level in our experience. Remember, the contributors to Revision.ai include 6 college graduates, with above average grades, so we know what you need to do to score well on exams!

It is our belief that if you have recently achieved trophies on each exercise in your lecture, within the last 2 weeks before the exam, you will go in as one of the strongest students and get the grade you need. What you need is the discipline and hard work on studying on the run up to that, in the regular months before hand - and we help take a big part of the early and intimidating work by generating all your flashcards/practise questions for you. It feels lighter, and more calm, to study with Revision.ai this way.

Our founder has a MSc Psychology degree with a focus on learning, as do two of our current/former interns.

We use our knowledge of the subject area to apply several research findings, as detailed in our articles and technically in terms of their implementation here.

Highlights of the research we have applied includes:

• The Testing Effect
• Spaced Repetition
• Summative Feedback research
• Misconception Refutation -
• Progress Gamification
• Mastery and indvidiualised feedback/planning - Blooms research
• Chi & Posners Theories of error types and learning modes

From there, you study exercises, one by one, gaining progress. The separation into exercises splits your lecture into relevant slides for that topic, and gives you a surmountable amount of questions for each exercise that fits your attention span. By taking them in order, you can familiarize yourself with the lecture, no matter how confusing some parts might be.

As you answer questions to make progress in each exercise, you can ask the AI to explain with context to your lecture:

Each time you study, the AI adapts the cards based on spaced repetition, and other factors. Each time you complete a session, you can view your strengths/weaknesses that session with a knowledge gap report:

Once you are confident, you can use the Mastery mode, where you view slides, and then need to write out what you remember for a blurt/free recall rating from the system.

This "Learning Loop" takes you from a few slides, to a few questions, then to recalling what you know. It's purpose-built to optimize your memory with a mix of easy and hard cognitive work for your brain - and to deliberately connect back to your original lecture with the slides relevant for the topic, while also ensuring that you are recalling information in exam-like conditions, the combination of which ensures you learn the right content (the lecture) in a comprehensive way, in the most effective method/fashion for the exam you need to take. Just stop and think for a second. How good does it feel to know you just accurately recalled most of the content of a topic in your lecture, and correctly answered short questions on it? Wouldn't that give you massive confidence for the exam and put you in flow state? This is the exact reason, as well as difficulty learning diagrams, that we built Revision.ai for Biology!

Then, you gather Trophies for each exercise by accurately free recalling what the lectures say and answering short questions, like in exams!

As you go, the system builds up a teacher-like picture of you, understanding the misconceptions and mistakes you are making, and any patterns you show. By looking at these, you can take rectifying learning, which quickly provides information on the misconception/issue, and then tests you on it.

Your progress is also represented for the whole lecture in the overview, so you can complete each lecture.

Misconceptions being resolved is arguably the most effective way to improve academics core in a subject according to Mueller (2008) and following studies.

It requires you to really intrinsically change your thinking patterns, reflect and overcome a hurdle mentally.

Our design around the Mastery mode learning loop going "Slides", "Questions", "Blurt" is purposeful. It enables you to tackle a topic comprehensively not just in terms of looking at all the relevant content, but testing your knowledge at different levels of difficulty to establish what you know, and in doing so, improving your knowledge of the topic.

See here is the thing. I actually quite like listening to lectures, especially the first time. They are interesting! The highlights are great, and memorable-by-default to me. But the bits inbetween. The foundations. The information on the methods used, related/foundational studies - or the diagrams - learning those parts helps your career, your future, your academic understanding and going on to do research. But it does not stick out, and will not stick in your brain, unless you study those parts purposefully.

Now you can try the traditional approach to make flashcards covering all parts and that's fine, but it is nowhere near as sophtiscated or effective as using the right learning techniques. It's also harder to not feel a daunting sense of fear that way - you can see the number you have studied correctly sure, but it's hard to study 300 cards. Every wrong one feels a bit worrying. The difference with our seperation of the topics into exercises and then taking Mastery mode for each of them is that you can consolidate your motivation, focus your attention, and show yourself as you apply effecitve learning methods that you can learn this topic, and can succeed. That's the feeling we want you to have and the rationale behind why the learning loop - informed by research - is built how it is.