Nobody's experience reports I've collected in my FAQ have reported
feeling *less* creative; and there's one mention specifically of
performing better on creative tasks.
(There's now a HTML version of my FAQ at
http://code.haskell.org/~gwern/static/N-back%20FAQ.html if anyone
hasn't read it yet. :)
--
gwern
That was probably only when they were meditating. If you couldn't weed
out irrelevant stimuli, you wouldn't be able to function as a human
being.
I was stuck at dual-6-back in default mode for about 4 or 5 weeks. I
figured that was it for me, but yesterday I made it to dual-7-back! Today,
I easily did it again and just missed making it to 8-back on a couple of
trials! I guess it's possible to continue to improve even after being stuck
at a certain level for an extended period of time (or at least this seems to
be true in my case).
Rick
I assure you that writers, at least, by and large are not brain-damaged with no working memory to inhibit the awful godly flow of images and stimuli. No, they (the great ones anyway) have incredible, unbelievable will and focus when it comes to their work.
What's the worry here? If you're worried about sacrificing your creativity by working hard on something mental and training your focus, then perhaps you shouldn't be involved in a creative profession in the first place, as this sort of thinking with lead you precisely nowhere.
Instead I recommend spiking your dual-n-back sessions with low to moderate doses of LSD, which is a real hoot for lowering latent inhibition when you need somma dat mojo.
I take you don't drink caffeine, do you?
First, those 5 kids didn't party hard because they took ADHD drugs for
performance. That's a cum hoc ergo proctor hoc logical fallacy,
believing correlation equals causation. This is like saying "I know
five baseball players who chew tobacco, so I'd never play baseball."
They have nothing to do with each other.
I'd bet those kids did much
better in school than they would have w/o the ADHD drugs and to
suggest otherwise makes you very ignorant of tons of data proving ADHD
drugs do improve grades.
I'm not advocating a deluge of nootropics, mind-altering drugs, and
what have you. I'm merely pointing out the flawed logic behind this
sort of thinking. These things cannot be written off and dismissed so
easily. And keep in mind many people are not going to buy into this
thinking and so avoid taking advantage of possible loopholes in their
neurochemistry.
Similarly:
If many of the foods we eat are cognition-enhancing drugs/supplements/
etc., then they are also external chemicals.
If external chemicals rob me of achieving a greater sense of
satisfaction and experience a greater quality of life precisely
because I wouldn't deserve all the rewards I had received by means of
them, then I shouldn't take them.
Many of the foods we eat are cognition-enhancing drugs/etc.;
therefore, I shouldn't take them.
(Ancillary: If I don't eat many kinds of foods, then my health and
well-being are compromised.
I don't eat many kinds of foods; therefore, my health and well-being
are compromised.)
So I've been looking at the uploaded latent inhibition, and so far I'm
skeptical that any of them are useful for training.
For example, take the Kaufman study. The test of LI was
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
iEYEAREKAAYFAkpsZ5QACgkQvpDo5Pfl1oL/WwCgikgmUhCmlunENB1c3eclfpKn
82wAoIfyUoAP5QjxW31z7g/YOm73uiNe
=0w8k
-----END PGP SIGNATURE-----
> "In the preexposed phase, participants were presented with
30 nonsense syllables, repeated 5 times with white noise bursts
superimposed randomly 31 times over the course of the recording.
Participants were instructed to determine how many times they
heard the third nonsense syllable (bim). Therefore, during this first
phase, the white noise bursts were irrelevant to the task.
> During the second task, the test phase, the same recording of the
syllables was replayed, but participants also watched yellow disks
appear one by one in rows on the computer monitor. This time, the
white noise bursts were relevant to the task: Each yellow disk
appeared before the white noise bursts. Participants were in-
structed to try to discover the auditory stimulus that caused the
yellow disks to appear (the correct answer is the white noise
bursts) and to write down their answer and raise their hand when
they thought that they had figured out the rule."
The second half sounds a *little* like games like Zendo*, but those
are known as inductive logic games, and I don't see how any
computer-generated secret rule would be nonlogical - so that doesn't
seem much like it'd help latent inhibition.
I also found this:
http://www.mshri.on.ca/roder/behavioural/protocols.html but their LI
test protocol seems infeasible to computerize & use on humans. :)
This paper http://homepages.gold.ac.uk/aphome/belfastli.pdf on testing
LI describes 8 or 9 tests of LI.
1. nonsense syllables/white-noise; seems to be the same as the first paper
2. variant of #1
3. Eye saccading tests; this seems infeasible as well, as none of us
have eye-trackers hooked up to our computers, and wouldn't be a good
training tool either
4.1. spatial working memory. Doesn't seem to have anything to do with
LI besides the generic WM/LI connexions?
4.2. Tower of London task. Ditto.
4.3. This one seems promising, as it looks eminently computerizable:
> "Attentional set shifting (Owen et al., 1991): Subjects are required to select one of two stimulus dimensions, when only one is correct. Initially, subjects are required to attend to different examples within the same dimension (intradimensional shift, IDS) and then to switch attention to the previously irrelevant dimension (extra-dimensional shift, EDS). At each stage, continuation to the next stage depends on six successive correct responses. The test is terminated automatically by the computer if the criterion is not met after 50 trials."
The citation is: Owen A M, Roberts A C, Polkey C E, Sahakian B J,
Robbins T W (1991) Extra-dimensional versus intra-dimensional shift
set shifting performance following frontal lobe excisions, temporal
lobe excisions or amygdalo-hippocampectomy in man. _Neuropsychologia_
29: 993–1006
Unfortunately, my library's copy of _Neuropsychologia_ only goes back
to '95; can anyone get this, or should I put in an ILL?
5. "Subjects were asked to generate as many words as possible
beginning with the letters F, A and S for 60 s each. They were
instructed not to produce numbers, proper nouns or the same word
with a different suffix."
This is a word generation task; it's perfectly computerizable - I
remember playing something like it at luminosity.com, although it
would let you have different suffixes. Although I wonder if it would
generalize, or would simply train verbal fluency?
6. inventory of 'akathasia'. Not a task.
7. inventory of 'subjective mood and sedation'. Ditto.
So of the tests I've found, only 1 looks like it might be an actual
computerizable test which isn't domain-specific.
* https://secure.wikimedia.org/wikipedia/en/wiki/Zendo_%28game%29
--
gwern