10 Responses to “Working-Memory Training Report – Shaun – Session 122”

1. Shaun Luttin says:

   May 23, 2009 at 9:31 pm

   Sat 23 May 2009

   In the training report (above) I committed to re-taking the TRI52 this weekend, and I just finished. Whew, it took four hours. But lunch with my mom broke that up for about two hours. So two 120 min blocks of matrix-type reasoning. Whew.

   My research question asked whether I would improve my gF, and by how much. My predicted stated that the TRI52 measurement would increase from 810 to 840. Well… I exceeded my expectations. My score on Monday 02 March 2009 reached 810 on the TRI52, and my score on Saturday 23 May 2009 reached 870!!! The TRI52 percentile went from 99.5th to 99.9th, or so.

   Here it is in a terse format:

   Mon02Mar2009
   TRI52, 810, ~99.5th, ~four hours of work
   BFPro, completed 79 sessions, nAve ~5.50

   Sat23May2009
   TRI52, 870, ~99.9th, ~four hours of work
   BFPro, completed 122 sessions, nAve ~6.90

   Okay… that’s all for now. Tomorrow I am going to take the gIQ again. That should take much less time than the TRI52 did!

   Shaun (happy with the improvements).

   PS… Mr. Devil, or perhaps his advocate Will, might like to speak of the statistical significance of this. The TRI52 mean is 518 and the standard deviation is 114.
2. Shaun says:

   May 24, 2009 at 4:58 pm

   Sun 24 May 2009

   Well, I committed to taking the gIQ test today, and I did it. My score came in at 144 this time. Now, I did not make predictions for the gIQ results. But, the result of the gIQ today is the same as the result was in Jan 2009. Both times the score came in at 144.

   So, I have improved my score on the TRI52, kept my score stable on the gIQ, and increased my mean n-back. As Nanz might say, onwards.

   Shaun
3. Will says:

   May 24, 2009 at 11:23 pm

   Great effort, Shaun. I’ll say this — it “appears” your fluid intelligence may have increased in the range of 6 to 8 points since starting the training but it really impossible to say for sure because your baseline is the WAIS and the only way to know is to take the WAIS again.

   My own guess is you are now up from 139 and into the mid to high 140′s, though that really cannot be posited with any real conviction since, as I opined about Erin’s 12 point increase on her blog, you cannot compare an internet test to the WAIS, even if the test in question is well constructed. In any event, good job on the TRI52. As far as that test goes, there’s no doubt imo that the 2nd result represents your ability level, whether or not the BFP training or taking more care/time with the items or a combination of the two was responsible for the increase.
4. Shaun says:

   May 25, 2009 at 12:30 am

   Thanks for the comments Will!! This seems to be your position:

   *we can tentatively claim an intelligence increase
   *it’s spurious to compare the WAIS with Web-tests
   *the latter TRI52 scores reflects my actual ability
   *this might be BFPro or more time/care with the items

   I’d like to add that the TRI52 seems to test performance IQ rather than verbal IQ. So, it’s more reasonable to compare the TRI52 results to the WAIS’s performance IQ score than it is to compare the TRI52 to the WAIS’s verbal IQ score. Back in Jul-2008 my performance score reached 134 whereas my verbal score reached 135. But as you said, it’s unwise to compare the WAIS to the TRI52, so why even bother, eh.

   Alright… enough contemplation of statistics. You may have been accurate Will, when you mentioned that people like me who do IQ stuff online might be more interested in statistics than in actual IQ gains!

   Warm regards,
   Shaun
5. Will says:

   May 25, 2009 at 1:23 am

   Shaun, right on, you do have my position summed up.

   “I’d like to add that the TRI52 seems to test performance IQ rather than verbal IQ. So, it’s more reasonable to compare the TRI52 results to the WAIS’s performance IQ score than it is to compare the TRI52 to the WAIS’s verbal IQ score. Back in Jul-2008 my performance score reached 134 whereas my verbal score reached 135. But as you said, it’s unwise to compare the WAIS to the TRI52, so why even bother, eh.”

   Well, no, actually I disagree there. A bother could be made, since Jouve (unlike most people who create online tests) has done some work to this end — so it’s worth the speculative bother. Otherwise, yeah, I would say screw it.

   http://www.cerebrals.org/tests/tri/pdf/TRI52%20CRV.pdf

   The fairly high correlation between performance IQ and the Tri52 (at .73) means you can use the Tri 52 as a reasonable proxy. I’d say there most likely would be an even higher correlation if you look at just the visual reasoning index score apart from processing speed on the performance scale of the WAIS 3. Basing this on the notion that there does seem to be testing of a similar ability (Raven’s type of perceptual reasoning).

   If this were the case it would put your TRI 52 score in a better proximity to your visual reasoning index score which seem to have been 140 (or thereabouts). If your WAIS 3 index score was 140, it seems you might have increased it somewhat to level I suggested — 6 or 8 points. And possibly even higher for your overall performance score, if you have increased processing speed (which is hard to test for as its special to a professional type of test like the WAIS).

   As for verbal IQ of the WAIS, it shows only a moderate correlation with the TRI 52 so I’d agree with you there.

   “You may have been accurate Will, when you mentioned that people like me who do IQ stuff online might be more interested in statistics than in actual IQ gains!”

   I’m tending to think given your overall WM has improved in the way that it obviously has, it is very likely that other aspects of cognition have as well…even areas that were extremely high prior to training. However, I am not qualified to say either way as I’m not a psychologist.
6. Shaun says:

   May 25, 2009 at 6:09 pm

   Hi there Will:

   “…it’s worth the speculative bother… …The fairly high correlation between performance IQ and the Tri52 (at .73) means you can use the Tri 52 as a reasonable proxy. I’d say there most likely would be an even higher correlation if you look at just the visual reasoning index score apart from processing speed on the performance scale of the WAIS 3.”

   So I decided to bother, and it took me about 20 minutes. In the interest of looking at the impact of BFPro, here it goes:

   On the one hand, my WAIS performance IQ is nearly equal to my WAIS Perceptual Organization Index – both are around 135. The latter leaves out speed in processing info and performing simple tasks related to that info. My sense is that the POI is equivalent to what you called the Visual Reasoning Index.

   PIQ = 134
   POI = 133

   On the other hand, my initial TRI52 score was 810 and my latter score 870. You mentioned that comparing this to my WAIS performance scores might actually be worth a bother. So I did a bit of math to come up with the following equation. The equation seems to convert a WAIS score to a TRI52 score.

   ***
   Conversion of WAIS Score to TRI52 Score:

   7.6(WAIS-PIQ) – 250 = TRI
   7.6(134) – 250 = 770

   Therefor my initial WAIS-PIQ came in equivalent to a TRI52 score of about 770. But this is wildly speculative, and you might want to check my math.

   ***
   By combining that conversion with the time-line of my BFPro training, it’s possible to get a speculative look at how my performance IQ has increased from BFPro.

   Jul2008, TRI52 ~770^
   *Then 79 sessions of BFPro

   Mar2009, TRI52 ~810
   *Then 43 sessions of BFPro

   May2009, TRI52 ~870
   *More session of BFPro to come

   ^This one represents the converted WAIS-PIQ score of 134.

   *****
   So there it is. Based on the result from the WAIS and the gIQ test, my full scale IQ may have increased from 139 to a stable 144. However, based on the results from the WAIS and the TRI52, my performance IQ may have increased from 135 to 147 (or from 770 to 870).

   *****
   Warm regards,
   And thank for exploring this with me, Will!!!
   Shaun
7. Shaun says:

   May 25, 2009 at 6:20 pm

   Oops…

   It should read as follows:

   7.6(WAIS-PIQ) – 242 = TRI

   Darn….
8. Will says:

   May 26, 2009 at 6:45 pm

   “PIQ = 134
   POI = 133″

   Hi Shaun, I liked your analysis. One thing that doesn’t make sense to me, though. Your performance IQ should be lower than your POI because it’s your PSI that brought down your performance score — This should mean your POI would therefore be higher than your overall performance IQ not a point lower as you indicated — so I find that a bit perplexing.

   As for as improvement claim, the main problem is that you’re comparing a test where ceiling effects may have hampered your pre test score to a post test that does not have ceiling bumping issues. So it’s not something that works as a claim and ergo the good devil chooses to pass on buying the 12 point increase. Nice try, though.

   Yet for marketing purposes (which I know is not your concern), you may as well round up to 15 or even 20 points because the law of marketing is to exaggerate claims whenever possible, even with flimsy data. — I’m not joking. It’s counterintuitive — but people need to see big numbers to take action, even if they are skeptical. This has to do with the fact that all of us are irrational in buying decisions and we need emotional triggers. This is a piece of wisdom sales/marketing people have always known, against
   what logic and reason would otherwise tell us about why people purchase things. Thus, the only sophistication we have as buyers is a knowledge of our irrationality (which isn’t how most people view themselves). I think you (and whoever else) will enjoy this video below which gives empirical evidence via neuroimaging how people respond to sales pitches. This is a sort of newfangled crossbred field called neuroeconomics.

   http://mitworld.mit.edu/video/598

   This MIT website is a great resource for fascinating lectures in various fields as well, especially in science fields.

   Forgetting the WAIS 3, your increase on the TRI52 pre and post test is compelling evidence for an increase fluid reasoning intelligence so devil will buy the 810 to 870 bump. The currency used for the purchase will be discussed at a later time
9. martin says:

   May 26, 2009 at 7:19 pm

   Will said: “It’s counterintuitive — but people need to see big numbers to take action, even if they are skeptical. This has to do with the fact that all of us are irrational in buying decisions and we need emotional triggers.”

   Hi Will, this reminds me of an NPR story about a man who had had brain surgery (due to a tumor if I recall correctly). He was an accountant, and everything with his recovery seemed great. He was able to process information and perform intellectual processes just as well as before. But getting back to his regular life proved problematic.

   It soon became clear that he could no longer make decisions. Even choosing whether to sign a piece of paper in blue or black ink became an impossible choice. This was coupled with an odd emotional disconnect.

   It transpired that the surgery had damaged his ability to process emotions. The NPR piece then went on to present the case that many if not all of our day-to-day decisions and choices are, at their core, emotional or subjective. We can process data and try to make our decisions as rational as possible, but there will always be too many parameters and possibilities with weightings that are indeterminate at best.

   Do we want granola or bran flakes for breakfast? We like them both. We can imagine enjoying each of them. One might be more nutritious. or perhaps new data suggests that the other is. One might be less fattening. But we could have a smaller portion. Would we be equally satisfied… Eventually we have to reach out and take one of the boxes and pour ourselves a bowlful. (Or mix them together!!)

   The emotional base for decision-making reflects on your comment, perhaps, in that an exaggerated marketing claim appeals to our subjective sense of wanting a “reason” to make the decision, something we can rationalize to ourselves and others.

   Martin
10. Shaun says:

    May 27, 2009 at 4:57 pm

    “Hi Shaun, I liked your analysis. One thing that doesn’t make sense to me, though. Your performance IQ should be lower than your POI because it’s your PSI that brought down your performance score — This should mean your POI would therefore be higher than your overall performance IQ not a point lower as you indicated — so I find that a bit perplexing.”

    Yeah dude, I find that perplexing too. Perhaps the psychologist that tested me is a quack! I double checked the numbers, and the POI is indeed lower than my PIQ.

    “As for as improvement claim, the main problem is that you’re comparing a test where ceiling effects may have hampered your pre test score to a post test that does not have ceiling bumping issues. So it’s not something that works as a claim and ergo the good devil chooses to pass on buying the 12 point increase. Nice try, though.”

    Yeah… that leads me to think about what the ceiling effect is for the performance score on the WAIS. It also leads me to reflect on what my score would have been if my working memory and my processing speed were at the same level then as the level they are at now. Hmm… suppose I won’t find out unless I take the WAIS again. Not likely given my frugal value system.

    “…the law of marketing is to exaggerate claims whenever possible…”

    This reminds me of a book I read: “Influence” by Cialdini. He write about what he calls the six weapons of influence. These include social proof, scarcity, commitment and consistency, liking, perceptual contrast, authority, and perhaps another that I’ve forgotten. Cialdini also argues for the ubiquity of reflexive rather than rational decision making during consumer behavior. Also, your comment reminds me of an article from UBC Reports (UBC is my Alma Matter). The article mentioned the emerging field of something like neuroeconomic ethics. This field looks at the ethics that will be important for protecting people’s autonomy from the sophisticated marketing tactics that emerge from neuroscience! Fascinating!

    “…your increase on the TRI52 pre and post test is compelling evidence for an increase fluid reasoning intelligence so devil will buy the 810 to 870 bump. The currency used for the purchase will be discussed at a later time…”

    Thanks… I’m glad that the evidence is compelling. What would I need to do to graduate the evidence from compelling to coercive? For methods of payment, I accept non-fiction books, classic music, and anything that contains organic dark chocolate; for dark chocolate is to my wife and I as fine wine is to amateur sommeliers

    Shaun