3 ms·
Incidentally, if this is intended as some sort of response to gwern's other post [1] as if he's never heard of dual n-back... please do read that FAQ. All 124 p
by jerf 13y ago
Incidentally, if this is intended as some sort of response to gwern's other post [1] as if he's never heard of dual n-back... please do read that FAQ. All 124 pages of it, according to my Firefox print preview.
https://news.ycombinator.com/item?id=6521885 https://news.ycombinator.com/item?id=6521885
- gwern 13y agoOr if you're lazy, just check the latest version of my n-back meta-analysis (http://www.gwern.net/DNB%20meta-analysis#control-groups http://www.gwern.net/DNB%20meta-analysis#control-groups): looks like the current point estimate for n-back's increase of IQ is 0.15 (so, 2 IQ points), and is not statistically-significant (p=0.14). A few of the researchers in the area have encouraged me to clean this up for publication, which would be nice. (Not that the rest of OP is much better than the n-back material. 'I worked with autistic kids and I can cherrypick a few examples of improvement!' As if he had never heard of regression to the mean or selection effects.)