Thanks for sharing this. I am curious to see the research paper which explains data and methodology in detail.
<Comment deleted by user>
Thank you for the post. It's great to see research being done in such areas. The test results seemingly suggest that calculation is primarily habit-driven, and brief explicit instructions have little impact on changing a player's deeply ingrained and automatic calculation processes.
Did you forget to link to the paper itself? Thanks for bringing attention to this effort.
@DailyInsanity said in #4:
> Thank you for the post. It's great to see research being done in such areas. The test results seemingly suggest that calculation is primarily habit-driven, and brief explicit instructions have little impact on changing a player's deeply ingrained and automatic calculation processes.
The online quiz test link is also proviing the results? And the population that took the test? Perhaps the range of expertise might be too close to show discernable patterns. I think, exagerating the dependant variables input output response range, might help seeing more factors.. If not enough range, they the signal we might seek might be buried in the noise.. If not enough spread of experience level, a dominant factor might be the only thing visible, and the actual learning dynamic signal might not be on par. so it might seem negligible effect on whatever the experimental measure protocol was using. I am speaking without knowing, the detail.
Currently I understand that only a categorical dependent variable was being controlled, the question with some controlled choice, and one less constrained similar question. How the more limited choice was designed might be something to look at.
Also: was there some characterization of the differences across the 207 individuals, the span of their different experience levels (the usual elo for lack of other things). The range of such dependent variable distribution might be interesting to consider, before interpreting too fast, or too generally.
The paper might have its own discussion. I like the discussion section of scientific papers. So we can modulate the strenght of concluding new hypotheses.
When probing a black box, and not knowing much, one might want to cast a wider net in the experimental design about the input data set. I think the link was forgotten. It was announced. perhaps in one of the links not directly about the paper. I did not look.
> Thank you for the post. It's great to see research being done in such areas. The test results seemingly suggest that calculation is primarily habit-driven, and brief explicit instructions have little impact on changing a player's deeply ingrained and automatic calculation processes.
The online quiz test link is also proviing the results? And the population that took the test? Perhaps the range of expertise might be too close to show discernable patterns. I think, exagerating the dependant variables input output response range, might help seeing more factors.. If not enough range, they the signal we might seek might be buried in the noise.. If not enough spread of experience level, a dominant factor might be the only thing visible, and the actual learning dynamic signal might not be on par. so it might seem negligible effect on whatever the experimental measure protocol was using. I am speaking without knowing, the detail.
Currently I understand that only a categorical dependent variable was being controlled, the question with some controlled choice, and one less constrained similar question. How the more limited choice was designed might be something to look at.
Also: was there some characterization of the differences across the 207 individuals, the span of their different experience levels (the usual elo for lack of other things). The range of such dependent variable distribution might be interesting to consider, before interpreting too fast, or too generally.
The paper might have its own discussion. I like the discussion section of scientific papers. So we can modulate the strenght of concluding new hypotheses.
When probing a black box, and not knowing much, one might want to cast a wider net in the experimental design about the input data set. I think the link was forgotten. It was announced. perhaps in one of the links not directly about the paper. I did not look.
As I say in my posts - it's not enough to know the technique; you need to practice it to move through the four levels of competence (unconscious incompetence, conscious incompetence, conscious competence, and unconscious competence).
wow, a lot of comments, the time that i spent doing this blog was worth it :)
We're still missing the source article whereabouts, though. Or am I missing something from the blog? I tried all the links, and one got be onto a book pdf offer, that did not seem to be what the blog was referring to.
I hope this is not a buy the book thing. I guess as long as there might be dangling profit, the relation between words and the scientific method effort, might be a bit warped. If Chessable is funding research at that level, it might do so, in the full spirit of scientific method, since it already has the infrastructure for maximal sharing. Should I keep worrying, in that direction? The doubt of having only been given what we might want to hear, and then get something that might be more of the same? I hope not. I have lots of imagination, so, I also have many doubts in many directions. A balance of sorts.
If you are transmitting marketting material, with enough catch words that seem to make sense, and do include at least what anyone ones own bias (in unknown it might be all we got..) pet hypothesis, that might be like SEO strategy applied to our chess mind search engine patterns. Which is basically content farming. Should I keep going in this skeptical direction. There is a lot of authority stuff in the blog on top, of the keywords I just referred to, as more content promise.
hmm.. I have other things to read.. already.. doubt cloud... here.
I hope this is not a buy the book thing. I guess as long as there might be dangling profit, the relation between words and the scientific method effort, might be a bit warped. If Chessable is funding research at that level, it might do so, in the full spirit of scientific method, since it already has the infrastructure for maximal sharing. Should I keep worrying, in that direction? The doubt of having only been given what we might want to hear, and then get something that might be more of the same? I hope not. I have lots of imagination, so, I also have many doubts in many directions. A balance of sorts.
If you are transmitting marketting material, with enough catch words that seem to make sense, and do include at least what anyone ones own bias (in unknown it might be all we got..) pet hypothesis, that might be like SEO strategy applied to our chess mind search engine patterns. Which is basically content farming. Should I keep going in this skeptical direction. There is a lot of authority stuff in the blog on top, of the keywords I just referred to, as more content promise.
hmm.. I have other things to read.. already.. doubt cloud... here.
Top demais parabéns