Rule 1: Boundaries matter more than definitions. We learn what something is by also seeing what it isn’t. Understanding concepts means discovering their boundaries; what counts as an instance and what doesn't. Strong students may eventually intuit boundaries on their own, but

Rule 2. Not all concepts are alike, the structure determines how you teach them. Concepts fall into different types. Some are about categories and need positive and negative examples (mammal vs. fish). Some are about processes and require step-by-step transformations (long

Rule 3. Complex skills need task analysis. Break them down first. Learning fails when teachers assume students know what hasn’t been actually taught. Many so-called “ability gaps” are really design failures. For example, many essays fail because the component parts like

Rule 4. Don’t reinvent examples. Recycle them to teach similar concepts. Designing a clean, faultless sequence is hard work. Once you have one, it’s a resource, not a one-off. Related concepts can be taught by reusing the same examples with new signals or questions. Students

Rule 5. Every example needs a signal. Examples don’t speak for themselves. A shape on the board could be “triangle,” “pointy,” “blue,” or “large.” Without a signal, students often must guess. The signal, whether verbal, visual, or gestural, acts as a cognitive spotlight,

6. Variety is your friend, but don’t randomise. This I think, is where interleaving has become a lethal mutation. Variety in examples is important but often produces exactly the opposite of its intended effect. Random assortments of instances teach students whatever pattern the

7. If your examples allow two rules, some students will learn the wrong one. Here lies the heart of what Engelmann/Carnine call "faultless communication"; the principle that every unnecessary variable in your examples creates a potential trap for struggling learners. When

8. Teach the minimum set that makes the maximum rule. Flooding students with examples may feel effective for experts but can confuse the pattern for novices. More is not better, better is better. The art of good instructional design is finding the fewest examples that force the

9. Relying on correction to bail you out could mean the initial design was flawed. If a healthcare system is spending vast amounts of money treating dietary illnesses, there may be something wrong with people’s diets. Constant error correction suggests that the instructional

Rule 10: Most adaptive learning is backwards. A lot of what is called adaptive or "personalised learning" typically involves making tasks easier when students struggle, but this approach often masks the real problem. Reducing challenge doesn't build capability; it circumvents

Read full post here and subscribe for regular updates 👇🧵 https://carlhendrick.substack....

@C_Hendrick I commented on this yesterday and got 613 views and zero likes and comments. Can you help me out with why this got ignored? I want to know how to avoid my students ignoring me.

'Theory of Instruction' is a book that every educator should read. Link below ⬇️

A good question to ask when considering engaging an education consultant is: Why are they doing this and not teaching or running a school? There are some very good reasons for doing consultancy. Who knows, I may do it one day. However, it’s worth asking the question.

Dr. Nidhi Sachdeva this week on Better Teaching: Only Stuff That Works. Is it possible to spread the Science of Learning message to all teachers? @nsachdeva2019 https://share.transistor.fm/s/...

TBH if this means Brain Gym, Learning Styles, Reading Recovery etc are inspection red flags, then I'm behind this. Somewhere in the accountability system there needs to be some lines drawn that say, 'This is junk.' You can't just do *anything* and hope it works. Do we want to