by Ian J. Deary
Deary, I. J., L. J. Whalley, H. Lemmon, J. R. Crawford, & J. M. Starr (2000). The stability of individual differences in mental ability from childhood to old age: follow-up of the 1932 Scottish Mental Survey. Intelligence, 28, 49–55.
A popular account of this work can be found at the following website: http://www.scre.ac.uk/rie/nl65/nl65deary.html.
The research paper that describes the large US Department of Labor Study that looked at mental ability test scores in tens of thousands of people from young adulthood to old age is:
Avolio, B. J. & D. A. Waldman (1994). Variations in cognitive, perceptual, and psychometric abilities across the working life span: Examining the effects of race, sex, experience, education, and occupational type. Psychology and Aging, 9, 430–42.
And the studies that followed up people after they had been tested during world wars I and II respectively:
Owens, W. A. (1966). Age and mental abilities: A second adult follow-up. Journal of Educational Psychology, 57, 311–25.
Schwartzman, A. E., D. Gold, D. Andres, T. Y. Arbuckle, & J. Chaikelson (1987). Stability of intelligence: A 40 year follow-up. Canadian Journal of Psychology, 41, 244–56.
There are two good summaries of the Schaie’s Seattle Longitudinal Study and I got much of my own information from them. The first is more accessible.
Schaie, K. W. (1994). The course of adult intellectual development. American Psychologist, 49, 304–13.
Schaie, K. W. (1996). Intellectual Development in Adulthood. Cambridge: Cambridge University Press.
The results from Salthouse’s research were taken mostly from the following papers.
Salthouse, T. A. (1996a). Constraints on theories of cognitive ageing. Psychonomic Bulletin and Review, 3, 287–99.
Salthouse, T. A. (1996b). The processing-speed theory of adult age differences in cognition. Psychological Review, 103, 403–28.
The research report from which I took Figure 10 was one of the Berlin Aging Study reports:
Lindenberger, U., U. Mayr, & R. Kliegl (1993). Speed and intelligence in old age. Psychology and Aging, 8, 207–20.
Chapter 3 Brainy?
Why are some people cleverer than others?
What we mean when we say that some people have higher psychometric intelligence than others is that some people reliably obtain more correct answers, and often achieve these faster, on a set of mental test questions. Previously we described the patterns into which these mental test scores assort. Later we look at whether scores on mental tests are of any use in predicting things in the real world. Here we ask the following question: why do some people score better on mental test questions than others? In fact, it is a more specific question than that: what is it about the human brain that makes some people better at psychometric intelligence test items than others? And we need to be prepared for some difficulties here. What we are attempting to do in this section is ask whether there are measurable aspects of brains and brain functions that differ between people and that also relate to psychometric intelligence differences.
In one sense we shall address this question of the origins of intelligence differences in the next section, when the genetic and environmental contributions to intelligence are described. Just a little reflection, though, tells us that these are rather distant causes. Finding that the genetic lottery and the environmental slings and arrows influence the level of some of our mental capabilities does not tell us what it is about the brain that makes some people cleverer than others. From antiquity this question has interested commentators on the human condition.
Prior to modern-day neuroscience, the guesses at the stuff that made for the more-efficient-brain were crude and followed the fashions of the times. For over 1500 years, the ideas about the more-efficient-brain were governed by the Greek and Roman physician-philosophers who thought that the well-tempered body had to have just the right amount of the four humours, blood, phlegm, and black and yellow bile. Such early efforts occasionally alighted upon one factor that was taken up by some scientists of the 19th century – the size of the brain – but research before the later years of the 1800s really told us very little.
One of the important things to point out at the beginning of this section is that our knowledge of the brain’s workings is still very incomplete. Even with the rise of the new brain sciences – neuroscience and cognitive science – we are still a long way from having a mechanistic account of how the brain thinks, emotes, and wills. Therefore, it cannot be surprising that our understanding of what makes some brains more efficient than others is still fairly rudimentary. It is possible, nevertheless, to offer some recent findings that provide intriguing clues.
For hundreds of years there have been simplistic hunches to the effect that people with greater mental powers might have brains that are bigger, faster, and/or finer-tuned. These hunches are hardly very clever; they are something that the man on the number 23 bus might have come up with given a moment or two’s reflection, even without much knowledge of the brain’s structure or function. Nevertheless, they have been tested and there is some scientific evidence worth recounting.
This is an area of research that I have spent some time in myself. Day in and day out I see it with its flaws and its small advances. What all of us in the research area know is that the main obstacle to progress is the lack of understanding of normal brain function and its variability. There have been great advances in understanding the brain and its functional units, but we are still a long way away from a mechanistic account of how thinking, feeling, and willing occur. The topics within this area of research that have attracted the most research effort are a rather mixed bunch. They are illustrated in Figures 11 to 14. In summary, I want to discuss how differences in psychometric intelligence relate to: brain size, the brain’s electrical activity, the efficiency of visual processing, and the speed of simple reactions.
Brain size
There is a modest association between brain size and psychometric intelligence. People with bigger brains tend to have higher mental test scores. We do not know yet why this association occurs.
Figure 11 is a picture of a 65-year-old man’s brain taken using a magnetic resonance imaging scanner. The man was taking part in one of my research team’s studies. We have not got to the stage of publishing the data from this study yet, but this will give a clear picture of how the research is carried out. The man took a large battery of mental ability tests and gave some blood for various assessments to be made. The last part of the study involved collecting data on the size of his brain and, specifically, the sizes of some particular parts of his brain – those we considered to be involved with memory and other areas of thought. What you can see in the image is a ‘slice’, in which the magnetic resonance scanner has taken a picture of the contents of his head from one ear across to the other. By moving our aim further to the front and the back of his head, we collected many images and eventually we were able to get a three-dimensional view of his whole brain. With these pictures, displayed on a very high-quality computer screen, one of our team drew around the outline of the brain. That is, she carefully, without knowing anything about the man, drew an outline around all of the brain ‘slices’ and worked out the brain area within each slice. Note these white outlines in Figure 11. Eventually, her information was compiled to give a measure of the man’s brain volume. She then repeated this procedure for 100 other men in the study. Thus, with a safe medical scanning machine that involves no radiation, we can now measure the size of people’s brains while they are alive, and we can ask if the size of the living brain is related to intelligence test scores. Let’s turn to results from other laboratories.
11. A picture of a living human brain taken using a magnetic resonance imaging scanner. Note the white line drawn around brain tissue to measure the area taken up by the brain in this ‘slice’.
Nancy Andreasen is a renowned researcher into schizophrenia. Among other research, she and her team have examined the structure of the brain in people
with that illness. The device her team used was a magnetic resonance imager like the one we used in our own research. Before the advent of magnetic resonance imaging, researchers had recourse to all sorts of methods that have been lampooned in the scientific and popular literature on research into intelligence. Brains were weighed after people died, skulls were filled with lead shot or other handy materials to find out how big the brain was that once resided there, and, more often, the size of the head (hat size, effectively) was measured. None of these archaic measures approaches a satisfactory way of getting at brain size (though there is a modest positive correlation between head size and brain size), but they were all born of the frustrating inability to get at brains and their sizes while people were still alive. That changed forever with the wider availability of magnetic resonance imaging machines. For the first time, the human brain was seen in situ, in vivo, in the living being. Accurate pictures of its shape and size could be reconstructed and its overall dimensions were at last available. The first person to correlate intelligence test scores with brain size – measured using magnetic resonance imaging – was the late Lee Willerman from the University of Texas at Austin. His path-breaking study in 1991 did find a modest association between brain size and cognitive ability: people with better scores on mental tests tended to have larger brains. But the study was limited by the fact that it mostly tested students, who are a rather narrow group of people with respect to their range of mental abilities. Better, then, to describe a more normal group, such as the healthy volunteers tested by Andreasen’s team.
Andreasen and her team collected the largest set of data which correlated normal, healthy humans’ brain sizes with their intelligence test scores. They had a broader – more normal – spread of intelligence test scores than Willerman’s students, meaning that we can be more confident that these results will apply to the general population. In 1993 they examined 67 people (they are now up to about 100). These volunteers underwent a brain scan in the Mental Health Clinical Research Center at the University of Iowa. They took a standard group of mental tests, one of the Wechsler test batteries that we saw in Chapter 1. The researchers then computed a correlation between the size of the brain and the score on the mental tests. They did find a modest association, a correlation of about 0.3 to 0.4. They then asked more detailed questions about whether the size of different areas of the brain was related to people’s prowess in specific types of mental ability. Results then and since are inconclusive on that issue.
Now, in psychology, people will rarely if ever believe that a finding is secure on just one or two studies. Many things can happen in a single study that can spuriously give rise to a positive result. Therefore, sensible researchers wait for many similar studies to be conducted in different, independent laboratories before they begin to accept that a finding is secure. This is certainly the case in the present topic. And so some researchers make it their work to collect all of the studies on a topic and put them together to see what the overall finding amounts to. This was done in the field of brain size and intelligence differences.
Key dataset 6
A group of researchers led by Tony Vernon gathered together all the studies up to 1999 that had examined the size of the living brain using modern brain-scanning machines and had correlated the brain volumes with the persons’ scores on mental tests. Like they did, let us omit all those studies that included clinical groups (people with illnesses) and look only at healthy samples. There exist 11 such studies. Overall, that amounted to 432 people who had their brains scanned to measure the size and they all took some mental ability tests too. It is important that such an exercise of averaging across different research studies tries to find all such studies: they have to include any that showed nothing or even indicated that people scoring better on cognitive tests had smaller brains (there are none). That done, the average correlation was about 0.4. That is a moderate effect size: not a huge association, but large enough to state securely that people who score better on mental tests do tend to have bigger brains.
To the best that we can judge, then, the untutored guess that the cleverer person is literally more ‘brainy’ has some modest force. The finding is fascinating more in what it does not tell us than in what it does. The relationship between the size of the brain and better scores on cognitive tests begs for some explanation, some mechanistic account. It is fair to say that the best anyone has at present is yet more guesses. Some have suggested that the bigger brains have more nerve cells. Some suggest that the nerve cells are the same in number, but they have more connections in the bigger brain. Others have come up with the idea that the bigger brain comes about because cleverer people have thicker fatty layers surrounding the nerve cells; these ‘myelin sheaths’ are the electrical insulation that surround nerve cells’ cables and help them to send messages more quickly. There are other suggestions, but they are all speculative. The work of the next decades in this field will be to find out why this brain size – cognitive ability association occurs.
The brain’s electrical activity
The evidence is mixed, but there is some indication that the brain’s electrical responses show differences between people of different levels of intelligence. People with higher intelligence, on average, appear to elicit faster, more complex, and differently shaped electrical responses. The main problem in this line of research is that, of the 100+ studies available to date, hardly one exactly repeats the previous studies, so we do not have a check on the trustworthiness of the findings.
Have a look at Figure 12. It is a trace of the electrical activity of the human brain. (In fact, it is an average of one person’s brain’s activity over many encounters with the same stimulus, as I shall explain below.) Going along the bottom of the Figure from left to right, the time span is about half a second (500 milliseconds). Going from bottom to top, we are measuring electrical activity in just a few millionths of a volt. The nerve cells of the brain transmit messages along their lengths by electrical discharge. Also, the chemical messages that one nerve cell sends to the next make alterations to the electrical status of the brain’s cells. As long as we are alive – alert, awake, asleep, whatever – our brain is electrically active and this activity can be measured using very sensitive equipment to give a picture, the electroencephalogram (or EEG). For example, we know that the brain’s electrical activity is faster when we are doing mental arithmetic than when we are relaxing.
12. A graph of the brain’s electrical activity. This is an average of one person’s brain’s activity to a number of ‘oddball’ stimuli.
A big advance was made in this area over 30 years ago when psychologists first became able to measure the brain’s electrical activity in response to simple, discrete stimuli. The EEG activity mentioned above is an amalgam of all that is psychologically happening to us at any one time. If we tried to get people to perform a small, specific psychological act and we then look at the EEG we would learn nothing, because the small amount of the brain’s electrical activity that was related to that single act would be swamped by the rest of the activity. It would be like trying to hear a distant skylark’s song standing beside the M1 during the rush hour. Researchers hit on the idea of teasing out the tiny electrical response to simple mental acts.
First let’s discuss their approach. They test people in a quiet laboratory, where sounds and other distractions are minimized. With their subjects sitting comfortably, they record the electrical activity of the brain by placing some small, metal electrodes on the surface of the scalp. The person being tested would, for example, listen to a long series – perhaps hundreds – of tones, just simple sounds. Most of these sounds, which occur every few seconds, are the same. However, the occasional one is different, perhaps much lower in pitch. These occasional different tones which break up the stream of repetitive normal sounds are called ‘oddball’ tones, because they are different from the norm. The experimenter asks the person to listen out for the occasional ‘oddball’ tones, perhaps to count t
hem just to make sure they paid attention. The experiment keeps going until over 50 or even 100 oddballs have been heard. The experimenter saves all the brain’s electrical responses to each of the oddball sounds and keeps a separate store of the brain’s responses to every one of the normal tones. Now, any one of the oddball’s electrical responses is a chaotic-looking squiggly line. If you looked at all 50 or 100 of the squiggly lines representing the brain’s electrical response to each of the various oddball sounds, they would all look different. However, hidden within each one of the responses is a very small, fairly constant ‘signal’, which is the brain’s specific response to the oddball sound. By averaging all the squiggly lines one can take out the EEG that was nothing to do with the oddball and just leave the oddball-related electrical activity. This is because the electrical response to the oddball sound is the only ‘constant’ pattern in the dozens of responses; it emerges intact when the rest of the chaos averages to a flat line. It’s then that you get a wavy line like that shown in Figure 12: an average of how the brain responds, electrically, to a sound that is different from other sounds in a stream of simple tones. This average electrical activity of the brain to a stimulus is called the ‘event related potential’ or ERP. Its shape has characteristic peaks and troughs.
The arrow in Figure 12 indicates when the stimulus – the oddball sound – came on. Note too that, after about of a second (at 300 milliseconds) there is an especially large positive wave (upward-going) of electrical activity. I have labelled this P300. It is called the P300 for the following reasons: ‘P’ because it is electrically positive and ‘300’ because it occurs about 300 milliseconds after the stimulus which elicits it. The P300 occurs in response to the oddball sound only, not to the normal tones. It is thought to reflect brain activity related to noticing difference or novelty. In most humans it typically, as we see here, occurs about second after the oddball sound starts. There is an earlier positive peak, labelled P200. (This earlier positive peak is discussed further below.) In the person whose responses I used for Figure 12 you can see that the ‘P200’ occurs a bit earlier than 200 milliseconds after the oddball sound. There is an even earlier negative electrical trough, called N100: a negative electrical wave at of a second (100 milliseconds) after the oddball sound comes on. So, when the brain notices even small stimuli and makes decisions about them, we get predictable types and patterns of electrical responses from our brains. For the oddball sound, the N100, P200, and P300 are typical electrical events. Other types of event have their own characteristic waves. These diagrams, then, can tell us about how fast and vigorously the brain responds on average to events in the outside world, and they reflect the decisions we have to make about these events.