Measuring IQ for a study.

[MugginBuns]

Hey there! I'm developing a program that's meant to increase fluid
intelligence by improving feature selection rather than working memory
(so that it can be used in conjunction with working memory training
and produce more results) and I was wondering how I should go about
measuring IQ before and after in test subjects.

My problem lies in determining what is reputable to use, and
preferably cheap. For example I could use
http://portal.wpspublish.com/portal/page?_pageid=53,69636&_dad=portal&_schema=PORTAL
which would be considerably expensive, or I could use http://www.iqtest.dk/main.swf
which would be free. An in between point would be
https://www.gigiassessment.com/shop/index.php

Does anyone have any suggestions? Or methods for deciphering what
would and would not be good to use?

[Samples]

It depends. What type of intelligence (e.g. verbal reasoning, non-
verbal reasoning etc) is your cognitive training targeted towards? If
you can manage to answer this, it'll be easier to workout what kind of
intelligence test may be used to help measure possible before/after
effects.

On Nov 26, 4:13 pm, MugginBuns <goodwillwebcam4s...@gmail.com> wrote:
> Hey there! I'm developing a program that's meant to increase fluid
> intelligence by improving feature selection rather than working memory
> (so that it can be used in conjunction with working memory training
> and produce more results) and I was wondering how I should go about
> measuring IQ before and after in test subjects.
>
> My problem lies in determining what is reputable to use, and

> preferably cheap. For example I could usehttp://portal.wpspublish.com/portal/page?_pageid=53,69636&_dad=portal...
> which would be considerably expensive, or I could usehttp://www.iqtest.dk/main.swf
> which would be free. An in between point would behttps://www.gigiassessment.com/shop/index.php

[Samples]

I'm actually really interested in this idea! However, I'm not to sure
how practical it's application would be. I'm only guessing, but do you
have some kind of randomized generator of stimuli for training to try
and shove practice effects out of the way? It's probably not good of
me to speculate on what you have done, however, because I don't know
about anything you have done so far.

It makes me think of training that is catered towards articulating a
certain type of thinking process, similarly, critical thinking and or
the scientific method (wouldn't mind practicing this at all).

Would be great to hear any other thoughts you have about it! However,
I understand if you are weary about this.

Regards,

Sameil Pleirz

On Nov 26, 4:13 pm, MugginBuns <goodwillwebcam4s...@gmail.com> wrote:

> Hey there! I'm developing a program that's meant to increase fluid
> intelligence by improving feature selection rather than working memory
> (so that it can be used in conjunction with working memory training
> and produce more results) and I was wondering how I should go about
> measuring IQ before and after in test subjects.
>
> My problem lies in determining what is reputable to use, and

> preferably cheap. For example I could usehttp://portal.wpspublish.com/portal/page?_pageid=53,69636&_dad=portal...
> which would be considerably expensive, or I could usehttp://www.iqtest.dk/main.swf
> which would be free. An in between point would behttps://www.gigiassessment.com/shop/index.php

[MugginBuns]

The training is geared towards developing ability for pattern
recognition. I've thought this would be best accomplished by having a
person compare and contrast sets of visual stimuli. As they progress
they're given larger and larger sets of stimulus to compare in the
same short amount of time. It's ultimately creating an intuitive
sensitivity to similarity and differences.

[Noah Dunn]

Hey,

This might work for you: http://www.cerebrals.org/wp/?page_id=44

It seems reputable. As that page says, there is a correlation between
this test and Sats/other intelligence tests. Plus it is free.

[MugginBuns]

Thanks a lot! I find this quite encouraging as it's close to my result
on another test, I think I'll be using this.

Also, I'd like to report a gain of 11 points in IQ from 126 to 137
over the last 3 months. :)

[Gwern Branwen]

On Fri, Nov 26, 2010 at 2:50 PM, MugginBuns
<goodwillw...@gmail.com> wrote:
> Thanks a lot! I find this quite encouraging as it's close to my result
> on another test, I think I'll be using this.
>
> Also, I'd like to report a gain of 11 points in IQ from 126 to 137
> over the last 3 months. :)

On what test, and corresponding with how many n-back levels improvement?

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

[Thomasthetankengine]

Interesting, but I think this is already a feature in BW.

Multi-stimulus n-back.

How would your idea differ from this?

[MugginBuns]

http://www.iqtest.dk/main.swf - 126 3 months ago
https://www.gigiassessment.com/shop/index.php - 126 3 months ago
-
http://www.iqtest.dk/main.swf - 140 2 weeks ago
http://www.cerebrals.org/wp/?page_id=44 - 137 yesterday
Multi stim n-back does not challenge the user with a bulk of
unnecessary information in a single frame. It uses its multi-
stimulation aspect to increase the amount of load on working memory. I
look to improve the process by which we determine what aspects of
separate sets of data are similar and different. In short my idea
presents the user with much more stimulus (at higher levels) all at
once and does not rely on working memory for success.

On Nov 26, 9:27 am, Noah Dunn <noahdunn...@gmail.com> wrote:

[dualnback]

Muggins,

Would you say your jump in IQ is due to n-back or to something else?

On Nov 27, 9:43 am, MugginBuns <goodwillwebcam4s...@gmail.com> wrote:
> http://www.iqtest.dk/main.swf - 126 3 months agohttps://www.gigiassessment.com/shop/index.php - 126 3 months ago
> -http://www.iqtest.dk/main.swf - 140 2 weeks agohttp://www.cerebrals.org/wp/?page_id=44- 137 yesterday

[MugginBuns]

I've also been doing aerobic exercise 7 days a week 1 hour each day
for the last 2 months.

On Nov 27, 9:42 am, dualnback <kripkef...@gmail.com> wrote:
> Muggins,
>
> Would you say your jump in IQ is due to n-back or to something else?
>
> On Nov 27, 9:43 am, MugginBuns <goodwillwebcam4s...@gmail.com> wrote:
>
>
>
>
>
>
>
> >http://www.iqtest.dk/main.swf - 126 3 months agohttps://www.gigiassessment.com/shop/index.php - 126 3 months ago

> > -http://www.iqtest.dk/main.swf - 140 2 weeks agohttp://www.cerebrals.org/wp/?page_id=44-137 yesterday

[dualnback]

Thanks

[milestones]

On Nov 26, 2:13 pm, MugginBuns <goodwillwebcam4s...@gmail.com> wrote:
> Hey there! I'm developing a program that's meant to increase fluid
> intelligence by improving feature selection rather than working memory
> (so that it can be used in conjunction with working memory training
> and produce more results) and I was wondering how I should go about
> measuring IQ before and after in test subjects.

I wonder, though, if by increasing feature selection ability one is
increasing fluid or crystalized ability? I imagine if one improves
pattern recognition ability in other areas of life that this will
probably transfer to performance on fluid IQ tests and vice versa. The
black box theory of G says that fluid intelligence is not about any
specific ability but rather a medley of process that enable a wide
variety of abilities in which pattern recognition and feature
selection appear as a major aspect of identifying those who are high
in G. But, pattern recognitiona ability is just a way of identifying
people, not G itself. There are many other ways of identifying people
without much in the way of pattern recognition -- vocabulary, general
information, reaction, and inspection time, etc. all of which can be
very g loaded. The ability to improve in pattern recognition, as with
one's ability to improve vocabulary et al, is surely highly correlated
with one's level of fluid intelligence but the question is, is it
fluid intelligence itself?