Brain Structure/Intelligence

MOFFAT at sscl.uwo.ca MOFFAT at sscl.uwo.ca
Wed Dec 13 15:16:57 EST 1995


>>Hi,
>>       I am an education major who has gotten out of his leauge in a 
>>paper I am writing.  In the paper I am trying to prove that the structure 
>>of the brain supports the multiple intelligence theory, rather than 
>>Spearman's g.  When I say brain structure, I am talking about 
>>modularity; the fact that the brain is made up of sub-systems and is not 
>>just a single thinking unit.  I have read articles by Ira Black (Cracking 
>>the Brain case) and Gould (Mozart and Modularity) as well as stuff by 
>>Gazzaniga.  Basicly my theory is that because the brain is made up of 
>>these nueral networks, it makes sense that people will have different 
>>abilities in different areas.  Like I said, I am an education major so if 
>>my logic seems to be flawed, please correct me.  Any comments would be 
>>welcomed.
>>Thank you
>>Bill Sallas

.
i don't study intelligence directly, but i do do research in cognitive
neuroscience regarding localization of abilities.
while i generally agree with the spirit of your thesis, i do not think that
your argument necessarily challenges the existence of a g factor in 
intelligence. 
It is certainly true that years of research in neuropschology have demonstrated
quite convincingly that different aspects of cognition can be essentially
independent and localized to different parts of the brain.  one of the simplest
examples of this is the fact that brain damage tends to disrupt function in 
highly specific domains of cognition (ie deficit in memory but not language;
or intact visual perception but deficient auditory comprehension...this list
goes on).
It is definitely true that cognitive abilities can be behaviorally and 
neurologically dissociated but I don't know that i would want to be in the 
position of saying that this fact alone precludes the possible existence of `g'.
these 2 ideas (g and localized abilities) are not necessarily mutually 
exclusive.
without the concept of g, it is very difficult to explain the phenomenon that 
virtually every cognitive test is positively correlated with virtually every
other cognitive test.  Take the WAIS-R for example (the most widely used 
intelligence test today)...the two most highly correlated subtests in this 
battery are vocabulary and block design...one involves telling the meanings of
words and the other involves arranging blocks to form a geometric pattern. these
tests are very highly correlated, meaning that people who have a high vocabulary
also tend to arrange the blocks very quickly and effectively. people who have
a low vocabulary tend not to be very efficient at manipulating blocks.
How does one account for this fact without the concept of g?...
However, it is very common
in neuropsychological testing of brain damaged patients (which i have done
extensively) that a person will have lost the ability to do block design but
their vocabulary is left intact...you see my point...highly dissociable 
abilities but neverless highly correlated in the general population.

different abilities can be independently localized in the brain but some other
underlying factor may still make some people generally good at all tasks and 
some other people generally poor at all tasks.

i don't know if anybody has written about this apparent paradox explicitly, but
i personally think there is a relatively simple explanation for it.
at the level of the whole brain cognitive abilities may be represented 
separately and independently.  However, i think the probable factor that 
underlies `g' is represented at the neuronal/cellular level.  ie something 
like better synaptic communication, faster transmission of impulses etc...  
this way, someone with generally more efficient neurons/electrical 
conduction could be generally better at most things yet still have their 
cognitive abilities parcelated quite independently at the level of the whole 
brain.

that's my two cents.  I hope this was of some help to you.
Scott Moffat.




More information about the Neur-sci mailing list