Hacker News new | past | comments | ask | show | jobs | submit login
Working memory at 5 is a better predictor of academic outcome than IQ [abstract] (nih.gov)
71 points by wslh on May 1, 2011 | hide | past | favorite | 14 comments



Thanks to the citation shown in the abstract submitted here, I was able to gain online access to the full text of the article through my alma mater university. The key issue here is that designers of IQ tests have known for a while that their empirically designed tests didn't tap all of the human abilities that might properly be called human "intelligence." The submitted paper test children with the WISC-R IQ at time 1, and the WISC-III IQ test at time 2. But now the Wechsler Intelligence Scale for Children (WISC) has gone into its fourth version, the WISC-IV, and the current Wechsler IQ test includes tests of working memory.

http://psychcorp.pearsonassessments.com/HAIWEB/Cultures/en-u...

Intelligence researchers get this. The older versions of IQ tests didn't have a broad enough variety of item content to tap important cognitive abilities such as working memory. Working memory is very important, and is a hot area in intelligence research in the last several years (as this paper reflects). The latest versions of the currently used tests are already taking this into account, and include working memory items in the item content of the tests, and subscale scoring to identify which test-takers have the most problems with working memory.

Another comment in this thread asked,

Does anybody know of a good introductory text about these various theories of memory? I'd love to learn more about them, but following references is a bit time consuming.

See

http://en.wikipedia.org/wiki/User:WeijiBaikeBianji/Intellige...

for a guide to background reading on these issues. The Alan Kaufman popular book IQ Testing 101 is a very good book, easy to read and quite up-to-date, on these issues.


I read a study some time ago which was examining the cognitive load required by humans to parse various natural languages. I cannot find any references to it at the moment, but the gist of it was that the grammatical structure of some lanuguages required the speaker/listener to keep more information in their working memory before the meaning could be extracted from the sentence (e.g. all the verbs at the end of the sentence). In computer analogies, think of it as being required to push the whole sentence onto a large stack which then needed to be completely popped before meaning was extracted, instead of pushing and popping in pieces as some natural language grammars allow.

This article thus had me wondering whether people who are raised speaking such languages natively have an inherent advantage in life because they are exercising their working memory earlier on and thus increasing it's capacity, which appears to convey more general advantages.


> (e.g. all the verbs at the end of the sentence)

Can happen in German. But I find written English actually worse, because it is often so ambiguous to parse---so you have to keep the words in mind and not just the partial abstract syntax tree (or whatever it is that my brains stores for complete sentences).


There are also some results from length of words; I remember reading that Chinese speakers had an effectively larger digit span because their numerals were single syllables as opposed to 'seven' or 'zero'.


And what is 'academic outcome' a predictor of?


Future income, among other things; more generally, success. It's actually fairly good at that. Don't let outliers and bias fool you.


The other day there was an article about the importance of motivation in intelligence and success. From my understanding of what this article is, that's a glaring hole in the theory.


This is a highly relevant point. The article you're referring to discussed the effect of motivation on IQ tests: motivated subjects did meaningfully better, which undermines the use of IQ as a test of general intelligence.

At 5, children will differ hugely in how motivated they are to do well on an IQ test, and parental influence may well play a large role.

A 5 year old won't exhibit much working memory if they're not concentrating on the test, and if they're not test-motivated by 5, they're less likely to be so when they're older.


success != academics.


This is hardly a surprise. Most "learning" in school is actually memorization. This is a skill I do not have and have not been successful at improving. For example: My short term memory is limited to two or three unrelated pieces of information. When I was working at Subway, I managed to improve this to four or five after months of daily practice. I left Subway to go to an out of state college. My short term memory went back to normal within a month. I went back to Subway for the summer, I was not able to regain that skill before school started again.

I'm definitely biased but I find memorization to be pretty useless as a primary method of learning. Unfortunately, testing actual learning would require the teachers to fully understand the material they teach and have the time recognize when someone was faking knowledge on examination. It's far, far easier to test memorization.


This is not about memorization, but working memory is like the memory cache (or registers?) of our brains.

There are more theories correlating working memory to IQ, specifically look at the dual n-back brain game.


Registers are the immediate computer analogy that came to mind for me as well.

Does anybody know of a good introductory text about these various theories of memory? I'd love to learn mire about them, but following references is a bit time consuming.


"... Does anybody know of a good introductory text about these various theories of memory? ..."

You might try, DeBono Edward, (Professor of Investigative Medicine at Cambridge when he wrote the book), "The Mechanism of Mind" ISBN 0224617095 January, 1969 ~ http://www.amazon.com/Mechanism-Mind-Edward-Bono/dp/03179055...

I've read it, it's old but forms the basis of DeBonos ideas on thinking. The book goes into detail on the physical formation of memory ~ http://c2.com/cgi/wiki?EdwardDeBono & http://en.wikipedia.org/wiki/Edward_de_Bono


Yes, it didn't sound perfect for me, but is a measure of how much juggling of information you can do before storing/retrieving it from another place.




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: