192 HOW THE BRAIN MAKES POTENTIALsome social classes and cultures than in others. Th at is, IQ scores are
mea sures of specifi c learning, as well as self- confi dence and so on, not
general intelligence. In chapter 3, I mentioned the paradox of attempting
to mea sure a supposed culture- free variable with an instrument couched
in the terms of a par tic u lar culture.
Sensible discussion on the matter could really stop there, in my opin-
ion. But even if IQ scores were genuine scientifi c mea sures, there is a
much deeper logical fl aw to consider. Investigations, interpretations and
conclusions are based almost entirely on correlation coeffi cients. As men-
tioned in chapter 1, correlations are an investigator’s honey- trap and the
cheapest weapon of the ideologue. It is inappropriate to infer causation
from them without further, properly controlled study.
For example, IQ is a mea sure of social class that correlates with a host
of factors that can also aff ect physical aspects of brain development
(though without necessarily aff ecting cognitive ability). Th ese factors
include stress in pregnancy or early childhood, malnutrition, exposure to
toxins, inheritance of cross- generational epige ne tic changes, and many
others to be discussed in chapter 10. For example, much evidence suggests
that moderate to severe stress, which is related to social class and to IQ test
per for mance, aff ects growth in the hippocampus and prefrontal cortex.^27
And diff er ent occupations and cultural be hav iors can themselves change
the sizes and connectivities in the brain (see below).
In other words, any IQ- brain correlations are prob ably due to joint
association with other factors. Or causation may be in the opposite di-
rection from that imagined. In fact, it seems astonishing that such pos-
sibilities are not taken into account when expansive claims about the
“neuroscience of intelligence” are made.
Th ere are also acknowledged diffi culties with MRI scans, in spite of
the basic brilliance of their invention. Remember that the fMRI estimates
activity in blocks of brain tissue by mea sur ing changes in the levels of
oxygen in the blood passing through it: the greater the oxygen level is,
the higher the activity will be. But the scans are known to be subject to a
variety of corruptions, such as noise (mea sure ment fl uctuations), visual
artifacts, and inadequate sampling.
As mentioned in chapter 1, the experience of having a scan, lying in
a claustrophobic cylindrical enclosure, is far from natu ral conditions. A
This content downloaded from 139.184.14.159 on Tue, 17 Oct 2017 13:54:00 UTChttp://www.ebook3000.com