> and I said “but single-cell recording is pretty well played out, right?” and she said “yeah, but at least it actually works.” So she kept doing it, and fMRI turned out to be mostly a great way to fake yourself out, as she predicted. It is NUCLEAR and EXPENSIVE, though…
-
-
The Dead Salmon paper was not a wakeup at all to the large number of brain imagers who used correction for multiple comparisons, nor to the makers of most of the brain imaging analysis software who had such corrections built in.
-
That a subset of researchers chose to ignore such correction methods in the chase for "significance", and reviewers and editors allowed that to continue despite widespread guidelines about controlling alpha in brain imaging, was the problem.
- 1 more reply
New conversation -
-
-
Multi layer conv nets trained via back propagation can do some useful things. Things that people wanted done but that computers could not do, computers can now do (well enough for many purposes). This is good. (I am not disagreeing with the broader point you are making however.)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
But it can defeat DOTA 2 players!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Before it started delivering, and after an initial phase of enthusiasm, connectionist models went through a long and tough "winter" period when people became highly skeptic they would ever work. Sounds like a winter of sort may have started for fMRI.
End of conversation
New conversation -
-
-
Yeah, as long as our starting point is that mind is necessarily generated by physical arrangements of matter, our ending point will remain this very dead-end you write about. At best, we'll fiddle with the outsides to make cooler toys.
-
[T]he possibility of man's avoiding self-destruction depends on his realizing before it is too late that what he let loose over Hiroshima, after fiddling with its exterior for three centuries like a mechanical toy, was the forces of his own unconscious mind. -- Owen Barfield
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.