4 ms·
There was a fantastic Hacker News discussion about Sold a Story a few months ago: https://news.ycombinator.com/item?id=35599181 https://news.ycombinator.com/ite
by AlanYx 3y ago
There was a fantastic Hacker News discussion about Sold a Story a few months ago: https://news.ycombinator.com/item?id=35599181 https://news.ycombinator.com/item?id=35599181
Apparently the same sort of analysis with math isn't quite as straightforward. There is some evidence that pure constructivism is not successful with struggling math students. But unlike reading instruction where skilled readers never use the strategies taught in things like three-cueing (for "decoding", anyway), skilled mathematicians do approach novel problems in an exploratory manner.
- taeric 3y agoOddly, I'd expect skilled readers do use cueing. But, I'd expect that understanding and interpreting the different cues around requires experience. I also expect that it is never done "blind" on the word. For example, if you see the word "sake", cues let you know if it is one or two syllables. But the cues alone are almost certainly not usable for determining what word would appear somewhere. (edit: LLMs and an interpretation of them being only the connection of cues not withstanding.)