establishment psycho-bunk 5 —

what is memory, or intelligence?
incautious claims of ‘IQ’ genes

a briefing document

Index
Analysis of an article from the New Scientist
Discussion
End notes

related documents on intelligence and IQ

logicians, 'logic' and madness
intelligence and madness
writing down stats - distribution of a random variable

Analysis of an article from the New Scientist

Now to an item from the New Scientist for commentary:

“One form of a common brain protein makes us rather worse at remembering things, researchers have discovered. It is a first step towards finding the genes for intelligence.

“Human intelligence is partly inherited - studies of parents and children show that about half our cleverness, or lack of it, is down to genes rather than environment.”

This statement is far from adequate. You will have to struggle with this, ‘intelligence’: misuse and abuse of statistics, to understand why, in particular this beautifully artistic section.

Intelligence will be affected by vast numbers of genes, just as will ‘health’ or ‘fitness’. Most mutations are detrimental. Notice in this case the words “makes us rather worse at remembering”. [1]

Back to the article:

“Now Dominique de Quervain and colleagues at the University of Zurich in Switzerland have found one of those genes.

“People who inherit the less common form of a serotonin receptor have worse short-term memory than people with the more common form. It is not - by itself - a gene for intelligence.”

I wonder why that remark was inserted? It looks like this gene may be involved in laying down long-term memory, and is not to do with short-term memory. Read on....

From The Turing test and intelligence

“Karl Lashley, after studying the unit of memory for thirty years, supposedly stated that he was not even sure whether the engram (unit of memory) even existed.”

Over to the Encyclopedia Britannica:

“Lashley, Karl S(pencer)
(b. June 7, 1890, Davis, W.Va., U.S.--d. Aug. 7, 1958, Paris), American psychologist who conducted quantitative investigations of the relation between brain mass and learning ability.”

“His monograph Brain Mechanisms and Intelligence (1929) contained two significant principles: mass action and equipotentiality. Mass action postulates that certain types of learning are mediated by the cerebral cortex (the convoluted outer layer of the cerebrum) as a whole, contrary to the view that every psychological function is localized at a specific place on the cortex. Equipotentiality, associated chiefly with sensory systems such as the visual, relates to the finding that some parts of a system take over the functions of other parts.”

Back to the cited New Scientist item again:

“Their immediate recall was just as good....”

Notice, short-term memory was not effected.

New Scientist article again:

“...showing their attention and motivation were the same, ...”

I wonder from what the authors’ motivationometer is built?

And the article once more:

“while the difference between the groups was no worse a day later, showing the genetic difference had no separate effect on long-term memory.”

Back to the Enc. Brit.:

“ Memory is one of the most widely studied cognitive functions, and a number of different aspects of memory are recognized. The labels short-term memory, primary memory, and working memory refer to the temporary storage of information that is necessary for the performance of many cognitive tasks. In order to understand this sentence, for example, a reader must maintain the first half of the sentence in working memory while reading the second half. This working memory has been graphically described as the memory one uses to hold a telephone number in mind after looking it up in a directory and while dialling. The capacity of working memory is limited, and it decays if not rehearsed. Long-term memory, secondary memory, and reference memory refer to the storage of information for longer periods. The capacity of long-term memory is very large--in practice unlimited--and it can endure indefinitely. In addition, psychologists distinguish episodic memory, a memory of specific events or episodes normally described by the verb remember, from semantic memory, a knowledge of facts normally said to be known rather than remembered.

Almost certainly, memory is stored over wide areas of the brain rather than in any single location.”

site map

advertising
disclaimer

Discussion

A while back there were claims that a genetic marker IGF2R associated with intelligence had been located. That work has not since been replicated as far as I am aware. For more, read chapter 6 of Matt Ridley’s Genome.

Here you can find a recent [October 2003] item that may help an appreciation of the complexity of ‘memory’, while Repressed memory cautions the reader about the incredible lack of reliability of what is laughingly called ‘memory’.

We may be making steady progress understanding links to ‘memory’ and ‘intelligence’, but we still know fearfully little. Consider the gene in question in the (as so often, rather sloppy) cited article from the New Scientist—5HT2a. It is possible such a gene in its supposedly less propitious form will also protect us from aging as fast or some such.

Doctor to man who has undergone various tests: “I have two bits of bad news for you. The first is that you have AIDS. The second is that you have Alzheimer’s.”
Man: “Well, at least I don’t have that terrible AIDS thing.”

This ‘joke’ can be seen as a way of protecting yourself from over-stress. In lesser form than in the joke, this could allow you a long, steady, productive life, rather than a neurotic life where 5% of people may over-achieve, but 95% may be so inhibited by compulsive checking as to achieve less than does the possibly memory-inhibiting gene.

I repeat, this gene in the sample appeared not to effect short-term memory, but perhaps the ‘automatic?’ laying down of longer-term memories.

Now,

1. there are ways of learning to do that process more effectively, at will, for instance deliberate rehearsal or linking items to a list.
2. It is likely we will soon enough be able to swallow a pill to ‘correct’ any ‘lower’ efficacy of memory transfer. Ever heard of ‘side-effects’? In this area they may not be what you expect. For instance, a person who has learnt to run their own ship effectively, or has compensating ‘advantages’, for instance less worry as hypothesised above, may find themselves attempting to pilot a ship which has become unfamiliar.
3. You may like to know that ECT (electro-convulsive ‘therapy’, I will not comment on my view of this practice at the moment) is sometimes thought to operate by causing retro-active memory loss, thus stopping the experiences that have led to deep, ‘incapacitating’ depression.

To recap, we don't know enough. Like Karl Lashley, our understanding of memory is not great. Our knowledge of ‘intelligence’ is even less. We are making progress, but a great deal of humility and caution would be better at the current state of ‘knowledge’.

Finally, would you prefer a slow computer with an effective, stable programme,or a very fast, high-memory computer programmed by someone who is scatty and incautious, or even written by the author/s of the New Scientist article?

End notes

1. The standard test for short-term memory tends to be reeling off several random digits, for instance 67325, and making the victim (sorry, subject) repeat them back immediately. Few people can manage more than about seven digits, or maybe one or two more.

   Among the most interesting work on this subject is that done by Adriann de Groot, who tested chess players at various levels. One of his main tests was to expose a chess position for a few seconds and then see whether the player could reproduce the position on another board. He did many other tests, but this was the one test that clearly distinguished between top-level players and their slight inferiors. Top-class chess players can reproduce the positions without errors, whereas others cannot.

   I once asked a high-level player where the difference lay between his ability and that of top-level grandmasters. He simply replied, “Memory, it is just about memory”.

   de Groot came to the conclusion that the reason for this difference in top-level players was they had seen so many positions that they could chunk the board into seven or so units, containing familiar sub-patterns (for instance, a fianchettoed castled king position). Thus, they only had to remember seven chunks, equivalent to seven Chinese characters or random numbers, whereas lesser players had not such a complete store of basic patterns. de Groot confirmed this by setting up random arrangements of pieces on chess boards, whence immediately even top grandmasters could no longer reproduce them error-free.

   Incidentally, another useful concept in de Groot’s work was that he found chess players could study for some months, or even years, and make apparently very little progress and then, suddenly (probably through some higher-level understanding), their game would jump to a higher level. This is very useful to keep in mind for any educational or learning process.

email email_abelard [at] abelard.org

© abelard, 2003, 25 october


all rights reserved

the address for this document is https://www.abelard.org/briefings/memory_and_intelligence.htm

1490 words
prints as 5 A4 pages (on my printer and set-up)