FWIW, I told a good friend to get her dementia-diagnosed father on a schedule of daily Dual N-Back training using open source programs (e.g. Brainworkshop). She asked his neurologist how effective this would be and he basically said, "it's the only cognitive training that could do anything."
Not only is Dual N-Back a great mental workout, but it will also put you to sleep pretty quickly if you set the parameters such that the task is sufficiently taxing. Here's one that should make you tired within 15-20 mins. if you're just starting out:
(1) Set N-Back to 3 (for every category of items that you have to remember, you are trying to remember the repeats in a string that have two items in between)
(2) Set number of item categories to 3 (e.g. Shapes, Sounds, Spatial/Placement, and Colors)
(3) Up the baseline 'randomness' factor or level of pattern obscurity.
I cannot overstate how important (2) is. So many people in the N-Back community care about maxing out (1). IMHO, the real benefits come when you get (2) maxed out while keeping an N-Back of 3. THEN, work your way to up to higher levels of N-Back.
You have a one in a billion short-term memory capacity and don't feel sleepy yet? Okay, throw in some multiple mentalism exercises while literally in the middle of doing the N-Back training. Ex.: visually 'recite' the alphabet backwards in your head with the letters being in mirror image format (so from 'behind').
You'll find it impoasible to get even remotely close to 80% accuracy if you turn the various parameters up high enough, and you'll have an easy way of falling asleep.
That seems like quite a strong claim. This would need to be supported by extraordinary empirical evidence. It's rather difficult for us to independently verify your friend's neurologist's word. Do you perhaps have some scientific studies to hand which might demonstrate the efficacy of this treatment?
My claim is only that a single neurologist said if any cognitive training might help, then N-Back training would be it. I have no other evidence other than my friend's testimony. Say the neurologist did say this, what's the implication? That you have defeasible justification for trying N-Back training, on recently diagnosed dementia patients, if you've already made the decision that cognitive training is worth pursuing (along with other treatments).
Hearsay and anecdote are not useful for determining the empirical validity of a theory or therapy. I don't know you, and I have no idea if your reporting is accurate, or if so, who this neurologist might have been, or whether they are trustworthy, and even if they are, I don't have any information as to whether this specific opinion is sound. Without empirical evidence, there is no objective reason to believe any part of this claim.
You are right that hearsay and anecdote are not useful for determining the empirical validity of a theory or therapy. However, hearsay and anecdote have been useful a million times over in helping people solve problems or challenges they've faced at certain times and in certain places (whether fully or partially).
People you don't know will tell you things like, "Do crossword puzzles. That will keep you sharp." I'm someone you don't know, and I'm telling you, "If you've decided to do some cognitive training, you should give N-Back training a try in the manner I described."
Maybe it will help you sleep on a night where you have trouble sleeping. Maybe it helps your mental acuity. Maybe nothing. Maybe it will be a waste of 30 minutes for you.
I have no idea why you think that any of this might be persuasive. Do you really do what random people on the Internet tell you, with no evidence, just because they say it's a good idea? And were I to accept this proposal, there are no conditions for success or failure: a willing mind would make for any number of post hoc ergo propter hoc fallacies.
Empiricism is a much more reliable way to know things than hearsay: this is the general reason for the push towards evidence-based medicine. There is every reason to reject unproven therapies, and what you are offering can not be distinguished from fiction. If you want to make a convincing case for this, you have no alternative but empirical evidence.
Personally, if programming, playing an instrument, drawing, learning languages, and studying mathematics, physics, and history aren't enough to keep my mind sharp, I'm probably not going to worry too much about it.
"Hey Joe, this brand of probiotic worked well for me. You should give it a try."
After hearing that, should Joe then look for meta-analyses of the effectiveness of different probiotic brands, or does he have defeasible justification for trying it out? Maybe you'll say a search for meta-analyses is too stringent, but that you could at least look at anonymous reviews on Amazon. Fine. Then go look at collections of anonymous anecdotal reports in the N-Back community.
You keep trying to describe a form of rigid empiricism, as-if I don't actually know or care about epistemology. I did a concentration in philosophy of science in undergrad, and was an editor/reviewer for an entry on the philosophy of science in the Stanford Encyclopedia of Philosophy. I'm not unfamiliar with empiricism.
Joe should ignore unproven therapies, particularly when accompanied by grandiose claims, and especially when the proponent refuses point-blank to substantiate said claims. Probiotics are another good example of an unproven therapy, but with substantially more evidence against their efficacy. Joe should certainly read meta-analyses about such things, and avoid quackery -- and others should avoid spreading quackery.
Is there any scientific evidence behind "the real benefits come when you get (2) maxed out while keeping an N-Back of 3"? Sounds really curious to know.
Two different examples representing two different dimensions to this:
(1) Say you're doing a minute long session and trying to catch auditory letter repeats that are supposed to occur around 5-8 times. On more fully featured N-Back progams you can reduce that number, so that only 1-2 repeats (say) occur. This can make it more difficult for some people because they hear too many false positives (if that makes any sense).
(2) You can change up the default parameters so that repeats are more finely staggered. For instance, take the following two strings:
AHJADCBWPORP (repeats are A and P)
AHPACPWBORJD (repeats are A and P)
The default setting for some programs will just create the first type of string, which can make it more predictable, and thus less mentally taxing. You can tune the parameters so that you get more strings like the second.
Not only is Dual N-Back a great mental workout, but it will also put you to sleep pretty quickly if you set the parameters such that the task is sufficiently taxing. Here's one that should make you tired within 15-20 mins. if you're just starting out:
(1) Set N-Back to 3 (for every category of items that you have to remember, you are trying to remember the repeats in a string that have two items in between)
(2) Set number of item categories to 3 (e.g. Shapes, Sounds, Spatial/Placement, and Colors)
(3) Up the baseline 'randomness' factor or level of pattern obscurity.
I cannot overstate how important (2) is. So many people in the N-Back community care about maxing out (1). IMHO, the real benefits come when you get (2) maxed out while keeping an N-Back of 3. THEN, work your way to up to higher levels of N-Back.
You have a one in a billion short-term memory capacity and don't feel sleepy yet? Okay, throw in some multiple mentalism exercises while literally in the middle of doing the N-Back training. Ex.: visually 'recite' the alphabet backwards in your head with the letters being in mirror image format (so from 'behind').
You'll find it impoasible to get even remotely close to 80% accuracy if you turn the various parameters up high enough, and you'll have an easy way of falling asleep.