Mindsparke defense

[Gwern Branwen]

http://www.danielwillingham.com/1/post/2012/06/new-study-fluid-intelligence-not-trainable.html
http://www.danielwillingham.com/1/post/2012/10/working-memory-training-are-the-studies-accurate.html

--
gwern
http://www.gwern.net

[Payman Saghafi]

Good study and a good defense of the study by Tom Redick.

Daniel wrote the following:

"But given the previous uneven findings of the effects, this study represents another piece of the emerging picture: either fluid intelligence is trainable only in some specialized yet-to-be-defined circumstances, or it's not possible to make a substantial improvement in fluid intelligence through training at all."

Yes, it seems pretty clear that trying to get a group of random people to increase their fluid intelligence in a short period of time is kind of like
asking a group of random people to hit the gym and bulk up in a week.

Pay

[☉]

It is an interesting claim that one should devote 4-5 days per week
for training, but it doesn't seem all that reasonable to devote that
much time only to see transfer.

Perhaps studies should start investigating other variants of n-back,
like QNB. I for one think more complex (WM) tasks have a better chance
of leading to some sort of result for above-average subjects than DNB.

That is to say, if "fluid intelligence is trainable only in some
specialized yet-to-be-defined circumstances", then it would perhaps be
best to investigate the varieties of the task out there so as to
determine those circumstances. Admittedly, there are many out there,
so there's a lot of research that could be done.

argumzio