Digit-symbol (coding) task

Post Reply New Topic

This source explains this task well:

https://www.brainbaseline.com/assessments/symbolsub

Poor performance on this task is associated with severe ADHD, autism, bipolar disorder, and schizophrenia. When I was 8, I took this test as part of the WISC-III, and scored only a 5 on it, my second lowest score--the mean is 10. When I took it again as part of the WISC-III at age 15, I did better, scoring an 8. Since I started having issues again, I wanted to see how I would perform today, but I had a hard time figuring it out, until I found this:

http://www.neurocogtrials.com/pdfs/BACS ... 202004.pdf

The way the test was done here was a little different; you would be presented the symbols and you had to write in the correct digits. I saw that the mean for the patients with schizophrenia on the first time was 36 and for controls was 54. I was curious how I'd do, but I wanted to be able to set up the test well in a non-biased way.

What I did was create an Excel workbook that would use a macro to generate the items and the key (because entering all that myself might "accustom" me to that version of the test). The parameters would be the same: I would pick the correct number substituting for 1 of 9 symbols in each item, there were up to 110 items with each correct item completed giving 1 point, and I would have 90 seconds. This test would actually be a related task, the symbol-digit modalities test. The worksheet initially looked like this:

[img][800:718]http://i57.tinypic.com/29vjfnk.png[/img]

I used the timer here to time myself, which gives a ringing sound when time is up:

http://www.online-stopwatch.com/countdown-timer/

I printed the test out each time I took it and took it by paper and pen. I took the test twice, with about an hour apart. On both tries, I scored a 52, which was well within 1st standard deviation from the mean, as shown on the study.

However, I noticed a problem with my method. My key was too large. Elsewhere, the font in the key is the exact same size as in the items, not much larger, as shown here:

http://www.springerimages.com/Images/RS ... 77-12-55-0

I then thought, Perhaps my score was influenced by the larger key making it easier to go back up and check the correct modality? So I redid the test to look like this:



Perhaps there are still some issues, such as the key being kinda close to the items, but I'm not sure how important it is. So I did 2 more rounds, using otherwise the same methods, such as using the timer set to 90 seconds.

I first took it just as I was getting off work, and my score was actually a little lower: 50. I noticed that the key being the same size as the items made it more difficult. After taking a nap and then eating something, I tried the test a second time, this time scoring only a 45, getting to the bottom of the 1st standard deviation for someone who has taken it a 2nd time. I guess my mind was still kinda fuzzy from sleeping.

Here are some issues I noticed as I was taking the test:

1.) My thoughts kept drifting away from the task, as I kept thinking about other things.
2.) I never memorized the key over the course of the test. Unless the symbol was immediately repeated, I always had to look up at the key.
3.) On that 4th time, I noticed that I would confuse symbols for other symbols occasionally, for example, thinking that the symbol just repeated when on second look it did not. This probably accounted for the lower score on that test.
4.) This task reminded me of my ROM hacking days, when I made ASM hacks where I would manually assemble the modified code (i.e. I would translate by hand the assembly instructions to the machine code operations). It seems that computer programming in general may be good practice for this sort of task, and I have been programming since the age of 10 or 11. It also may have contributed to my behavior in issue no. 2, as when you do manually assemble, you had better be darn clear you're putting in the right operation code or you could have a hard-to-find bug; since memory is buggy, it's always best to reference the translation.

These issues seem sorta ADHD/Aspie'ish.

Anyway, would anyone else be interested in taking this test?

EDIT: Also, I'm looking for feedback on whether I should make the items further away from the key.

EDIT 2: Please note that every time I take the test, I do a random sort on the symbols on the symbol sheet, so that they would be in a different order each subsequent time I take the test. The order looks the same on the above examples, but that is because I was using an old save to update the Workbook, and did not sort that old save between the first screenshot and the second screenshot.

That is the part of the IQ test that I score the absolutely lowest on. 1% percentile. Not joking. Don't know why I struggle so much on it. I do have autism and bipolar 1 with psychotic features. The previous diagnosis was schizoaffective disorder. It depends on the doc on whether I have bipolar or schizoaffective.

I actually did weirdly well on this when I was 15. I have a nonverbal learning disability, and this is supposed to be the hardest sub-test for people with NLD, yet I actually scored a 19 (maximum score). It was a total outlier among my other nonverbal sub-test scores. The next highest score was a full two standard deviations lower, and thanks to this one bizarrely high score, my nonverbal sub-test scores actually span fully HALF of the entire IQ range tested by the Wechsler tests.

I'm very bad at this.


It is like you have to draw your attention to external things. That is not how I function. Doing it felt like an insult towards myself.

We have computers for this and I can even make a program to do it for me.

Does mediation help with this? I mean make you numb towards your own humane dignity?

I scored incredibly high on coding. I believe it was my highest score. I remember the neuropsychologist who administered my IQ test was in shock, because she said she had never before seen anybody finish the full page/task in the time limit. I absolutely loved the coding task, too. I'd do it for fun on a regular basis. I also have NVLD, but I've always excelled in symbols and associations. That is probably why I was hyperlexic as a child and do so well in foreign languages. I actually still remember what the number 2's symbol was on the test- it was a mathematical perpendicular sign. That's why I perform so well on coding. After the first couple of lines, I've already started to memorize the symbols for the numbers, so I don't have to spend time looking up at the key. My nonverbal struggles on the WAIS come into play with such tasks as block design. And I supposedly scored in the 2nd percentile on the Rey-Osterrieth.

So here we have two women with NVLD who scored interestingly well in Coding. In my IQ test I scored 13 on Coding, the same result as in Similarities. I scored 14 on Block Design. But my Object Assembly and Picture Completion were relatively poor (9), my PIQ was 104, VIQ - 126, FSIQ - 117. I do not know if I have NVLD, one person concluded that I have not it or just have a very mild NVLD.