New software claims to increase IQ

[Rotem Segev]

Same old trick, using N-back as the basis for the software, making a
nice package around it and selling it as a magic pill.

http://www.iqmindware.com/i3/

They even claim to increase all areas of intelligence... (See
"Scientific Basis")

Thoughts?

[UOchris1]

Looks like the researchers from the original DNB study are trying to
capitalize on their results; I'm pretty sure that i3 mindware is just
another rendition of DNB with some added brain puzzles that can be
found in any IQ puzzle book.  My advice is to stick with brainworkshop
and donate if you feel grateful for saving $60.  Overall I am a little
disappointed that the site would be so misleading and one would hope
that they would at least honor the money back guarantee if you were
dissatisfied.  Although I can't blame the researchers for trying to
make some money, to me it represents a conflict of interest and I
would have preferred that they stayed out of the market.

Further, id wager that their IQ problems along the way train you for
the second IQ test given to you at the end of your training regimen to
demonstrate your IQ gains.

[Zaraki]

"The scientists most central to the the historical development of this
training method are shown below. Note that these individuals were not
involved in the development of our brain training software, and this
is not an endorsement of our products on their part."

Seems you misunderstood their reason for showing the pictures.

Nothing worth noting on that site really. Just a lot of pictures to
convince the uninitiated and weak of will.

[shitsumeisha]

> Thoughts?

"[...] you can increase working memory capacity by over 65% over just
19 days of training. This improvement then results in a remarkable 40%
plus gain in intelligence as measured by a version of the time limited
Raven’s Advanced Progressive Matrices IQ test – one of the most valid
and highly regarded IQ tests for culture fair intelligence." (http://
www.iqmindware.com/i3/scientific-basis/)

40% of intelligence or IQ? I don't understand this.

From http://www.iqmindware.com/i3/highest-iq/what-is-an-iq-score/ :
IQ Range % (presumably of population)
<70 2%
70-79 8%
80-89 23%
90-109 50%
110-115 25%
115-129 15% (oh, overlapping)
130+ 2%

A Population larger than itself. Range 110-115 (6 points) with 25%
larger than range 80-90. Range 90-109 (20 points) only at 50%, compare
to range 110-115, which should be "smaller" per point. Maybe the
ranges should be 110-119 and 120-129. These numbers are nonsense for
me.

"The scientists most central to the the historical development of this
training method are shown below. Note that these individuals were not
involved in the development of our brain training software, and this

is not an endorsement of our products on their part." (http://
www.iqmindware.com/i3/credentials/)

At least they didn't review the website ;-)

Of course this doesn't necessarily mean that the software is bad. I
didn't look at much else, though some links under "Train IQ" (looks
more like "Test IQ") might be interessting. Oh well ...

[UOchris1]

More of an oversight than a misunderstanding. I read the intro of
Smith, the PhD developer and took the rest for the same. Apologies
for not being as thorough as I would on some things.

As far as the rest of my post, I obviously still think the software is
little more than DNB.

[Annie Anderson]

I've been using this software and it's designed as a clear departure from dual n-back training - as they say -

• Questionable IQ gains. With dual n-back training alone, the research is mixed as to whether an IQ gain automatically follows (although the other important cognitive benefits are not in question). A number of recent studies have challenged the Jaeggi study, showing that with certain populations (e.g. college students) n-back training does not result in an improved IQ – such as this one.  i3 Mindware′s ‘interference control‘ as well as its capacity-strategy brain training method, isolates precisely what results in IQ gains in n-back training to ensure an IQ increase.

The recent science behind the link between interference control and working memory  is discussed in some depth in their website. i3 bases its 2G training on this science. The link for this paper is here. 

I have seen big improvements in the tests using interference control training compared to standard dual n-back.

i3 has teamed up with Quantified Mind to do cognitive profiling before and after training - including IQ tests - so you've got valid metrics for what happens as a result of your training. Yonni Donner obviously trusts this company.

Annie

[Heinstar]

Doesn't Brain Workshop already have "interference control"?

And "capacity-strategy brain training method" sounds similar to LSAT test. I was considering practicing for LSAT to improve my reasoning ability but the study says that it took 3 months of intense study to see the cognitive changes in the brain. So it requires a lot of dedication and time. I might as well further my career then to spend so much time and energy solving logic questions.

[Annie Anderson]

It's not at all clear what interference control is in Brain Workshop - e..g whether or not it is the same algorithm used in the literature on interference etc. And if it does implement interference how many people using Brain Workshop know this - out of 20 or so options - is critical?  I'd rather focus on what works - as i3 does. 

Also - Brain Workshop doesn't have valid measures of pre and post-training in IQ, memory, attention, learning, etc - but i3 does. I'd not like to waste my time on somthing that didn't show me objective benefits that I could measure.

The problems and tutorials that go with the training help with IQ tests specifically. Certainly learning how to solve certain types of problems with some strategies helps in abilities on those problems - you don't need 3 months training. That's why we have math teachers, etc. Critical thinking/problem skills can also be taught effectively - there are studies showing this I have read.

I'm following the IQ Mindware stuff (i3) because it looks like there are a lot of innovations coming out based on the newsletters. An emotional n-back is coming out soon, as well as a decision-making app based on signal detection theory. None of this is out in the public domain. And the 2G training in i3 is definitely unique. 

[Pontus Granström]

I've bought it already. Did my first session today.

--
You received this message because you are subscribed to the Google Groups "Dual N-Back, Brain Training & Intelligence" group.
To unsubscribe from this group and stop receiving emails from it, send an email to brain-trainin...@googlegroups.com.
To post to this group, send email to brain-t...@googlegroups.com.
Visit this group at http://groups.google.com/group/brain-training.
For more options, visit https://groups.google.com/d/optout.

[Green]

As I recall, Persson 2008, initially reported that DNB+interference training improved executive functions:

http://www.ncbi.nlm.nih.gov/pubmed/18947353

But the study was retracted:

http://pss.sagepub.com/content/19/9/881.short

"This article has been retracted by both of the authors. During further extension of this work, it was discovered that a mistake had been made in the programming of the working memory training tasks used in this study."

Since that study was retracted, I have not seen any positive findings on DNB+interference training. 

[jagunit]

What did you think of it? Do you think it's worth purchasing when one already is using free resources like Brain Workshop, Cambridge Brain Sciences, etc.?

[Pontus Granström]

Well it's very similar to dual-n-back.

[Umut]

Well Pontus can you be more specific,  is it worth to spend over free Alternatives?

26 Nis 2014 19:50 tarihinde "Pontus Granström" <lepo...@gmail.com> yazdı:

[Annie Anderson]

Hi,

Thanks for the article on Interference control Green. What this tells me is the following:

1. The idea was obviously scientifically motivated enough to do this kind of study in the first place. 

2. What they were trying to show wasn't overall performance on IQ or  - but 'the ability to resolve interference' in different working memory, semantic memory, and episodic memory tasks. Resolving interference is calculated by subtracting RTs in low interference conditions from RTs in high interference conditions. i3 Mindware looks at 'executive functions' (not types of memory) as well as IQ, and is interested in performance gains - not this measure.

3. This study - which seems to me badly designed - only ran for 8 days. Dual n-back studies I've looked at don't show any results for at least 15 days of training.

Umut - if you play around with 2 and 3 settings on interference and rhythm you can see it's quite different from dual n-back - although I find it difficult to pin-point how the interference happens every time.

Annie

[Green]

Annie,

          The issue is that IQ Mindware claims that DNB+interference enhances the benefits of DNB training is not validated by any scientific studies that I know of. So, I went to the IQ Mindware website, to see if they had some studies on interferrence training, and instead I found something that gave me pause.

           The IQ Mindware website says:
         

Some studies such as this one and this one provide evidence for IQ gains from n-back training, while others such as this one and this one do not. Why the inconsistency?....An explanation for the inconsistency is that traditional dual n-back training does not incorporate  interference control in their n-back algorithm..... i3 Mindware is the only brain training app that gives users interference control training. (bold added for emphasis) 

            (qted at http://www.i3mindware.com/2g-nback/interference-control-training )

             But the studies they cite do not back up this bold claim. The links point to Jaeggi 2008, and Stephenson 2013 as studies that show successful transfer from DNB to IQ. First, the Stephenson 2013 abstract says nothing about interferrence control, instead they say "The primary goal of our study was to determine whether a visuospatial component is required in the training program for participants to experience gains in tests of fluid intelligence." So, they claim the key to success is visualspatial training, not interference as IQ Mindware says.

            As for the other study, Jaeggi 2008, this is the original DNB study! The Jaeggi 2008 study used the "traditional dual N-back training" that IQ Mindware's website condemns as ineffective. If this study is appropriately considered a success (and see Gwern's FAQ for reasons to think it wasn't), then the study's success has nothing to do with a newer N-back technique.

          So my issue is this: I know of no studies that suggest DNB+interference transfers to Gf, and based on the IQ Mindware website, it looks like they don't either!

 

 

 

     

 

[Pontus Granström]

I thinks its worth in the sense that a new training variant gives me motivation to train a bit longer and with better quality.

[Annie Anderson]

Hi Green,

If you look closely, the references you point to here are intended to show that while some studies (like the original Jaeggi study) show IQ gains with NB training (dual or spatial), others do not. This just shows the inconsistency of results (and remember studies are still coming out showing IQ gains from n-back training). What they say about interference control is:

Studies by Burgess, Gray, and fellow grad student Tod Braver (article 1, article 2) provide brain imaging evidence of a large overlap of gF and WM span brain mechanisms when there is need for interference control on a task – but not otherwise. They found that brain regions common to fluid intelligence and working memory became more active when there was more need to filter out distractions (‘lures’). The brain areas underlying both interference control, working memory capacity and fluid intelligence were the lateral prefrontal cortex (lPFC) and parietal cortex, shown in the fMRI images below which show activation in both the left and right hemispheres of the brain.

So the argument is I suppose that since interference control is the link between intelligence and working memory capacity, TRAINING working memory will improve intelligence. This isn't proven by a study it's true, but it's scientfically plausible and based on these studies. When people first started training working memory to look for effects on IQ this was based on what was known about the link between working memory and IQ. This just takes this reasoning one step further.

They don't have direct proof with a DNB + interference study, but if they did everyone would be offering this training! What they do have is good reason to believe this - and presumably they've got a lot of feedback from users on the effectiveness of this. And they guarantee your money back if you don't get a >10 point gain on independent IQ tests (e.g. JCT-i). So they are at least sticking their neck out on their claim.

Annie.

[Pontus Granström]

Read my years old post about lure trials and the correlation with rapm

--

[Heinstar]

I see, I didn’t know that i3mindware uses different type of interference. I thought that all interference was the same; I guess I need to learn more about how interference works.

I've tried various interference settings in BWS and so far, I've found that;
- With zero interference(aka Jaeggi mode), DnB started to feel predictable and boring quickly.
- With some interference(between 0.125 & 0.25), DnB was interesting and kept me more engaged and on my toes.
- With high interference(0.50 & higher), DnB was counterproductive because the blue squares frequently only stayed in 2 positions multiple times which made the match making too difficult and confusing especially in higher n-back.

I've tried i3mindware and the highest interference level on their software doesn’t seem to have this problem of blue squares only staying in 2 positions multiple times. So I guess if you can’t be bothered to tweak the correct interference setting yourself on BWS, i3mindware might be a good option. But one of the big down side with i3mindware is that there’s no manual mode. So I’m forced to always start from 1-back every time I start training which I find annoying. 

[Annie Anderson]

That's really interesting - thanks. I just went straight to setting 3 after about 2 sessions in i3. I agree with the manual setting idea - that would be a good feature.

What I like about i3 and IQ Mindware is they are making a big effort to record pre and post-training scores on different cognitive dimensions (using independent tests  unlike Lumosity) and  using this data are building in what works based on this.  This is why they teamed up Quantified Mind and why they offer the guarantees. The trouble I have with BWS - aside from the interface - is that it generates countless training options, but it's difficult to know what to focus on. It's like the idea here is - try everything possible that might work. In practical terms there are a lot of people who can't see the wood from the trees with this approach.

There's an emotional n-back coming out any day which I'm really interested in.

Annie

[jotaro]

i second to that, emotional n back seems the only interesting thing regarding n back for me now, and the only reason i might try n back again.

i wonder why no one tries to do a free version of it.

it is less popular for sure but it holds potential.

--

[Green]

   The Burgess, Gray, and Braver studies don't provide good evidence for DNB+interference because they're just correlational studies and not intervention studies. Just because IQ correlates with interference control does not mean improving interference control will improve IQ. That's just as fallacious as reasoning that, since IQ correlates with yearly income then you'll lose your intelligence if you quit your job and spend a year camping. Fallacious reasoning.

   I am sympathetic to the idea that DNB contributes to executive functioning though. I would be surprised if prolonged, varied, and intensive DNB training, like the kind many people on this forum often engage in, yields NO BENEFIT in terms of cognitive functions. So, if IQ Mindware motivates you to try harder, then you probably get some value out of it. It's just that brainworkshop does the interference training for free. 

  One thing I agree with: when I was doing DNB intensively on Jaeggi protocol, I remember that around N=6 it started to get easier to stay at on level, even if I couldn't stay focused. So, some kind of shaking up of the protocol seems called for.

  I'm also curious about emotional n-back. I'd pay some amount of money just to play around with it.

 

[Annie Anderson]

Hi - Just got a chance to look at this. I disagree that it's fallacious in this way, because there's a good mechanistic reason to believe that interference control is the 'missing link' between IQ and working memory - as explained in the Braver, etc, studies..Aside from that they seem to be getting good evidence based on using the self-quantification and independent IQ test measures. That's the philosophy behind IQ Mindware. It's not just motivation but also the fact that you can take a bunch of pre-training cognitive tests and compare them to post-training tests - with minimal practice effects (which Yoni Donner has helped minimize) to worry about. I like the fact that the whole thing is based on self-quantification. Basically the company specialize in cognitive testing as well as brain training. 

Anyway, their EQPro app for the emotional dual n-back has been released and I think you might like it:) So far it seems good IMO. Also you've got the 'manual' n-back option that someone else mentioned was helpful. (eqmindware.com)

Annie

[jotaro]

oh wow it almost seems as you are either a huge fan or have hidden motives,

--

[Annie Anderson]

:D I'm not an affiliate or anything like that:) I did end up thinking that another vendor was a waste of a lot of my time and money and I've committed a lot of time to this one.

[Annie Anderson]

For a negative, I do feel like they could improve on the instructions. 

[Annie Anderson]

The other vendor I've heard has scientific credibility (going beyond just the n-back) is this one.-Merzenich's Posit Science. I'm considering it next.

Annie

[Green]

My negative comments about IQ Mindware notwithstanding, I was really interested in the emotional n-back training. So, when I learned that they had released "EQPro" I paid 29.99 for a single-user license and downloaded the software. Here is a mini-review:

Installation on my ancient Dell laptop running Windows XP went smoothly. Initializing a new profile was a little hard to understand as there were some options that were not labeled clearly. However, I was able to get it figured out quickly by guess-and-checking. After I got the software up and running, I noticed a couple things that I did not like:

First, after playing with the software for 20 rounds it became obvious that the verbal channel just did not work. Whether I ignored verbal stimuli or not, my score never for the verbal channel went above 70% (on N=1). I have done dual N-back before and reached N=8 during regular training, and I can't believe that I was unable to get a single verbal match. I even tried ignoring the visual stimuli and JUST focused on words, but I could not get over 70%. Sometimes verbal stimuli did not appear THE ENTIRE ROUND, so there was NO WAY for me to fail to get a 100%, but the software would still report my score at 70%. I did a whole 20 rounds so that they would have some good data before I wrote them about the bug.

Second, the 'anxiety' program is just a dual-n back task using visually-displayed words in place of the auditory channel (words like 'cancer' or 'rape' are interspersed with emotionally neutral words like 'tape' and 'trap'). Angry or horrified looking faces are used to mark locations for the spatial channel. But none of this seems like it could plausibly build emotional awareness or emotional self-control: The emotional content of the words and images does not play a role in the memory component. You are supposed to match words and match locations, but you never match faces or match the emotional valence of words or faces. Since words appear in the center of the screen, you can't even look at the frightened and angry faces except through your peripheral vision. Admittedly, some of the faces were not only horrified, but genuinely HORRIFYING, and I was a little creeped out just seeing them in my peripheal vision. But I adjusted pretty fast since there are less than 20 faces total (I would guess closer to 10). Also, I admit that words like 'rape' and 'death' make me a bit jumpy even just when displayed on a black laptop screen without any other verbal context, but the only plausible benefit this could have for me is to make me less sensitive to those particular words.

In conclusion, the software neither worked properly (unless I'm doing something grossly wrong) nor seemed like a promising way to train emotional regulation. I was hoping for an N-back task that required me to match the emotional valence of words or images (for example: 'death' and 'kill' might be a match because they both have negative emotional valence).  If I wanted desensitization to upsetting words and images I would just turn on netflix and watch some bad horror films.

I wrote the company about the first problem, but not the second. I will update about whether, and how quickly, they get around to my problem. I'm not sure whether I should keep the software if they fix the first problem but not the second  because I don't see much potential benefit right now.

[jotaro]

thank you green.

i guessed that this would happen because in the hype page

it  seems they were indicating they will use the faces as interference

and not only that it seemed they will have few pics that u wil adjast quickly.

so i dont know hope you can continue to provide info about your experience

--

You received this message because you are subscribed to the Google Groups "Dual N-Back, Brain Training & Intelligence" group.

[Annie Anderson]

Could be because of the old computer? (ancient Dell laptop running Windows XP) I didn't have any of these problems so I'm sure that it's not the software.

Sorry you didn't feel you got anything out of it. It's just an implementation of the emotional dual n-back in the literature. I don't think they claim anything more than that.

Anyway, I'll be signing out now:) Hope it works out for you.

[Green]

Update: They refunded my money quickly (within 12 hours of my initial contact). That's professional conduct, which bodes well for the company, in my mind. The emotional N-back software package I purchased depends on Adobe Air, which, Dr. Smith indicated, does not support SOMETHING about my computer - it could be windows XP or it could be that the Dell laptop was built in 2002. 

Anyone interested in taking the N-back concept and turning it into a marketable product should look at the little iphone app called "dynamic duo." That's an example of a creative reinvention of the N-back task. It's much more "gamified" and less "clinicized" than IQ Mindware in the way it's marketed, and it doesn't keep track of progress, but the task design is inventive without departing from the CONCEPT of dual N-back:  it uses a variety of different stimuli and presents stimuli as a continuously moving stream of colored designs. Jaeggify Dynamic Duo a little bit, and I think you'll have something worth selling.

[whoisbambam]

wow.

poor customer service IMO

[Pontus Granström]

eqmindware seems promising.

[Claude Kaminski]

I have created a facial expression n-back (3 pictures simultaneously and asynchronous) and some other variations of the task for my own use.  I can send you the facial expression n-back task as an .exe(flash projector), I just need your email address. However it is not polished and I have no access to the various databases, so you may have to add your own images.

[jotaro]

send me craz...@gmail.com

--

[jotaro]

also if so how can you adjust pics quicly so they will fit the image folder for the software use? so we can do this with brain workshop.

[ApOkwARG]

I am really interested in trying this Dynamic Duo, but was unable to locate it on the app store. Would love to try it! Could you post a link to their website?  

[Green]

Hi. I'm so sorry - I've been calling it by the wrong name - it's called "Double Dynamo."