task, is a causal determinant of intelligence, what then causes there to be
differences in perception speed (see Richardson, 1991)? What would be the
biological basis for mental speed, or for any other component of cognition
measured by information processing tasks?
Research on the Physiological Basis of Intelligence Anotherwaytogetattheun-
derlying nature of intelligence is to examine neurophysiological correlates of
individual differences in cognition. Typically, research and theory studying the
neurophysiological basis for intelligence has assumed, at least implicitly, that
intelligence is a unitary phenomenon. For instance, researchers have speculated
that the brain of a highly intelligent person has more synapses among neurons
(Birren, Woods, & Williams, 1979), more efficiently metabolizes energy (Smith,
1984), or more efficiently reconfigures connections among neurons (Larson &
Saccuzzo, 1989). Unfortunately, it is difficult to obtain clear-cut evidence for or
against any of these conjectures, because the research on the physiological
underpinnings of individual differences in cognition is meager and inconclu-
sive. One of the main difficulties lies in measuring the critical physiological pro-
cesses, which are likely to be dynamic phenomena reflected in the way neurons
communicate with one another.
Are Smart Brains Metabolically Efficient? Recently, brain imaging technology,
such as positron emission tomography scanning (PET scans), has allowed re-
searchers to study metabolic activity in various sections of the brain of an alive
and awake person. Some studies suggest that people who do better on intel-
ligence tests tend to display lower neural metabolic activity. Haier, Siegel,
Nuechterlein, et al. (1988) found that performance on the Raven’s Matrices was
negatively correlated with overall cortical metabolic rate. Subjects who scored
higher on the Raven’s Matrices test (a nonverbal intelligence test) tended to
have lower overall cortical metabolic rates than subjects who scored lower on
the test. The authors speculated that people who are good at reasoning tasks
have more efficient neural circuits which therefore use less energy than the
neural circuits of people who have more trouble with the reasoning tasks.
Haier, Siegel, MacLachlan, et al. (1992) measured cortical metabolic activity
during the initial stages of learning the complex computer gametetris,and
again several weeks later after subjects practiced the game. They found that
subjects who improved the most on the computer game displayed the largest
drop in cortical metabolic activity while playing the game. Similar results have
been found by Parks, et al. (1988).
In apparent contradiction to these studies, though, is research that has un-
covered a positive correlation between metabolic rate and performance on IQ
tests. This research, however, has usually used elderly subjects, some of whom
have Alzheimer’s disease and other forms of dementia (e.g., Butler, Dickinson,
Katholi, & Halsey, 1983; Chase et al., 1984). Aging and disease may alter the
normal functioning of the brain.
Even if the negative correlation between performance on intelligence tests
and cortical metabolic activity proves reliable, interpretation problems remain.
It is not clear what makes neural circuits more efficient. Is it the density of the
neurons, the ease with which neurons affect the activity of other neurons, the
number of glial cells that support the neurons, or any of a number of other
Individual Differences in Cognition 787