by Ian J. Deary
Again, as we noted with the inspection time findings, it is really quite exciting to find that something as complex as an intelligence test score can be related to something as simple as reaction time. However, it would be reductive to think that intelligence is about faster and more consistent reactions. The association, though consistent, is not large and only a small portion of intelligence differences at best could ever be explained by reaction time’s speed and variability differences.
Again, as we found with inspection time, although most researchers acknowledge that the association between reaction times and intelligence is a real advance, they disagree strongly about what the connection means. Some psychologists, again, think it is an indicator that the person with higher intelligence has a brain that is a faster and more consistent processor of information. That is, they assume that the simple procedure involved in reaction time can tell us about some basic limitations or operating characteristics of the brain. On the other hand, those who dissent from this view say that reaction time is in fact rather complex and can be affected by some of the things that affect our performance on intelligence tests. Really, this is mostly a replay of the argument current within the inspection time researches – that is, whether speed in reaction time is a cause or merely a symptom of intelligence differences.
There is one possible reason for the association between reaction times and intelligence test scores that will have occurred to some and that needs countering. It would be easy to assume that the association between reaction times and intelligence comes about because reaction time involves working quickly and accurately and so does performing on intelligence tests. But, in fact, the association between reaction time and psychometric intelligence tests is found also with those intelligence tests that are not speeded, where people are left to take as long as they like to complete the questions.
One more bit of detail on reaction times. Have another look at the reaction time box drawn in Figure 14. If you think about the activity of completing a single reaction time trial you can imagine the sorts of mental processes you go through. Attend to the target buttons; notice which one has been lit; lift finger from the home button; get to the lit target light and press it as fast as possible. This involves a combination of decision-making and reacting. Some psychologists have been keen to separate the thinking and doing parts of reaction times, and this is how they did it. Instead of having a single timer in the box, they have two, to give a measure of the person’s ‘decision time’ and their ‘movement time’. Here’s how.
As before, the task is a choice reaction time test with all eight of the semi-circle’s buttons being possible targets. The person puts their preferred finger on the home button. Gets ready. Attends to the target lights. One of the target lights comes on and the first timer starts. The clock is ticking ready to measure the speed of the person’s response. Here’s the difference. This time, the first clock stops when the person’s finger is lifted from the home button: that is, the first timer calculates how long it took the person to decide to lift their finger and make a start toward the target button after the target light came on. This first time is the person’s ‘decision time’. As soon as the first timer stops, the second one begins – that is, when the person takes their finger off the home button. It stops again when they put their finger on the target button. This second timer calculates the time between the finger coming off the home button and going on to the target: that is, it is measuring the time it took the person to move from the home button, having decided which button was correct. This is called the person’s ‘movement time’. Thus reaction time can be split into decision and movement sections and measured separately: both the speed and the variability of the decision time and the movement time can also be assessed. It’s a surprise to many people that the decision time takes about of a second and that the movement time is much less, about only of a second. That is, it takes almost twice as long to lift the finger off the home button as it does to go from the home button to the target.
Both decision time and movement time relate to intelligence test scores. People with higher intelligence test scores have faster decision and movement times. With regard to variability, it tends to be the variability of only the decision time that relates to intelligence – people with better intelligence test scores are less variable in their decision times – whereas there is no relation with variability of movement time.
What research is currently going on in this area?
One idea that ties up a lot of this field is that brighter people have a faster ‘mental speed’. This broad idea, that cleverer people are somehow mentally faster, is an old and vague one. I can certainly trace it back at least as far as the 17th-century English philosopher Thomas Hobbes, and it has never really gone out of fashion. Psychologists today often refer to the ‘mental speed’ or ‘information processing speed’ ‘theory’ of intelligence. What they mean by that is that people who score better on intelligence tests might in part be cleverer because some key aspect(s) of the brain proceeds faster. My principal problem with this overall idea is that my colleagues can’t make up their mind how to measure this mental speed. Some use reaction times. Some use inspection times. Some use the brain’s electrical responses. Some even measure how long it takes electrical impulses to travel along people’s nerves. But these are all different measures, and it is an odd theory that can be tested without a common yardstick, and some of these mental speed ‘yardsticks’ don’t relate to each other very well at all. The truth is that we do not have an agreed measure of how fast the brain processes information, and that is because the workings of the nerve cells and their networks are largely mysterious. We must summarize by concluding, therefore, that intelligence is related to many things that involve speed of processing information, but that scientists have difficulty in conceptualizing ‘mental speed’ in a uniform way. I think it is likely that that will change quite quickly with new methods of brain scanning. At present, though, we need to acknowledge what findings there are. Those described above are real and interesting, but their limitations must be acknowledged.
There are more and more studies of brain size and intelligence appearing nowadays; in normal adults, in children, in old people, and in groups of people with illnesses. The focus is moving on from just finding out yet again that bigger brains tend to go with higher intelligence. The search is on for the explanation. Researchers are beginning to examine the way that people’s brains cope with the inspection time task by having them perform it in brain scanners and watching the activity in different parts of the brain as they do the test. There are more studies appearing of how drugs that affect the brain also affect inspection time, reaction times, and mental test performance. There are studies coming out on how ageing affects the speed of processing of information (see Chapter 2).
To follow this area up …
This was the one chapter in the present book for which I was not able to pull out a few key sources and describe them in more detail. A research colleague from my own department and I wrote a short, general overview of biologically oriented approaches to intelligence as follows.
Deary, I. J. & P. G. Caryl (1997). Neuroscience and human intelligence differences. Trends in Neurosciences, 20, 365–71.
The best sources for follow-up material are the chapters by Tony Vernon, Ian Deary, and David Lohman in the following book:
Robert J. Sternberg (ed.) (2000). Handbook of Intelligence. Cambridge: Cambridge University Press.
My own recent monograph on this area is aimed at fellow researchers and students and is thus technical rather than popular in style.
Deary, I. J. (2000). Looking Down on Human Intelligence: From Psychometrics to the Brain. Oxford: Oxford University Press.
Here’s the article in which Nancy Andreasen first reported the association between in vivo brain size and intelligence in normal people.
Andreasen, N.C. (et al.) (1993). Intelligence and brain structure in normal individuals. American Journa
l of Psychiatry, 150, 130–4.
Here’s a review of the inspection time research that I wrote for the nonspecialist reader.
Deary, I. J. & C. Stough (1996). Intelligence and inspection time: achievements, prospects and problems. American Psychologist, 51, 599–608.
Chapter 4 ‘They f—— you up, your mum and dad’ (Larkin)
Are intelligence differences a result of genes or environments or both?
Most people who are curious about human intelligence want to know whether there is much information about its origins: do genes have an appreciable effect?; what is the impact of the environment? Let’s start with a simple result: people in the same family tend to be more alike in their intelligence test scores than unrelated individuals. Like many other human characteristics, being clever tends to run in families. And the closer the family relation in an extended family, the closer is the resemblance in intelligence level. However, that is a near-useless finding because it cannot possibly tell us the origins of the contributions to intelligence: we share an environment, as well as genes, with our parents. Perhaps the environment they provided – the nutrition, the books, the schooling, the encouragement, the health care, the not smoking, and so forth – helped shape our intellectual capabilities? Maybe. But maybe it was the genes they gave us, the 50% of our genes that we share with our mothers and the 50% with our fathers. We can’t pull these two effects apart. The same people who mixed up our genetic cocktail also produced the environment. How can we find a way to study the effects of each separately?
Research in this area focuses on the study of twins and the study of people who are adopted. Sometimes in this area twins are called ‘experiments of nature’ and people who are adopted are called ‘experiments of society’. In what comes next I want to explain how these groups can help us to understand the origins of human intelligence differences.
Key dataset 7
Twins
Everyone knows that there are two types of twins: identical and non-identical. The key thing for researchers is that identical twins have exactly the same genes. What happens is that a sperm from the father fertilizes an egg from the mother and creates an embryo. At a very early stage the embryo splits into two. Therefore, what might have been one being becomes two genetically identical beings. Non-identical twins are only as genetically alike as any brother or sister. They have on average 50% of their genes in common. What happens is that two sperm from the father fertilize two eggs from the mother, creating two separate embryos, which develop into two genetically non-identical human beings. So, identical twins have 100% of their genes in common and non-identical twins only 50%. Therefore, we have a remarkable natural occurrence whereby we have types of people whom we know are always the same age and are either genetically identical or share 50% of their genes.
15. A diagram of the environmental and genetic influences on intelligence for identical twins reared together.
Now look at Figure 15. This refers to a pair of identical twins brought up in the same family. There is a box each for twin 1 and twin 2 of the pair. Since they are identical twins, they have to be the same sex, so twin 1 might be John Smith and twin 2 might be James Smith. The boxes just represent the twins and something about them that interests us, such as their score on an intelligence test. The first box, then, could be John Smith’s intelligence test score and the second could be James Smith’s intelligence test score. So we have two intelligence test scores from our two identical twins. Next we want to think about the influences on those test scores, specifically the influences of environment and of genes, and we want to ask which of these influences are shared by John and James Smith and which are not.
In the Figure, note the label G and that there is an arrow pointing from it to both of the twins of the identical twin pair. G stands for genes, and the arrow pointing to each of the twins from the same G captures the fact that they have identical genes. Now look at Figure 16, which refers to non-identical twins brought up together in the same family. Again G represents the effects of genes on measured intelligence, but notice the difference between this and Figure 15. Here there are two different circles with Gs in them to signify that the genes of these two twins are not identical. However, we do know that non-identical twins share half of their genes on average. So we can join their sources of genes with an arrow labelled to indicate this.
16. A diagram of the environmental and genetic influences on intelligence for non-identical twins reared together.
Before going into more detail on this, it is worth focusing in general on the environment and how it might be partitioned. Anyone brought up with brothers and/or sisters has two separable aspects to their environment. There are those aspects of the environment that they share with their brothers and sisters. For example, they might share feeding patterns and diets, family outings and holidays, the home’s books and other educational resources, the parents’ attitudes, and so forth. Then there are those aspects of the environment that are their experiences alone. They might have had different illnesses, have different friends, read different books, have different hobbies, even experience the ‘same’ events very differently, and so forth. Therefore, when we think about the environment we need to be more specific. It can at least be divided into that which we have in common with our siblings and that which we have to ourselves, our shared and private experiences. The environmental effects we share with our siblings are called the common (C) environment. (It is also called ‘shared’ or ‘between-family’ environment in the research literature.) The environmental effects we do not share with our siblings are called the unique (U) environment. (This is sometimes referred to as ‘unshared’ or ‘within-family’ environment.) To recap. When we ask about the effects of the environment on intelligence – or anything else – we can be more specific and ask if it was our family upbringing that had the effect and/or our unique experiences that we did not share even with members of our close family.
17. A diagram of the environmental and genetic influences on intelligence for identical twins reared apart.
Back to Figures 15 and 16. In both, C and U are experienced in the same way by each member of any twin pair – identical or non-identical – brought up in the same home. They share a common environment (denoted by a single C with two arrows): being a member of a particular family will produce an effect of the environment that they share. There are separate U circles for each member of each twin pair. This represents the fact that they have some non-shared aspects of their environment that can affect their level of intelligence.
Let’s recap. If we ask about the influences on the intelligence of the identical twins reared together, we see three sources: genes, which they share 100%; common environment, which they share 100%; and unshared environment, which they don’t share at all. For the non-identical twins reared together: genes, which they share 50%; common environment, which they share 100%; and unshared environment, which they don’t share at all.
Next, let’s look at twins (identical and non-identical) who have been separated very early in life and have been brought up in completely different families. This is a rare occurrence, so there are not many studies on it around the world. In situations where it does happen, it is extremely difficult to trace and test the twins involved. Figure 17 shows a pair of identical twins reared apart. The two twins in such a situation still have 100% of genes in common. They will still have a portion of the environment that they share with other members of the rearing family, and they have their unique experiences too. However, they have no ‘shared’ environment with their twin because they were separated from them to be brought up in different families. So, unlike Figure 15, Figure 17 has two different C circles, one for each identical twin.
In summary, for identical twins reared apart, the influences on their intelligence test scores may be summed up as follows. There are genetic influences, which they share 100%; there are aspects of the environment that they share with the siblings in the rearing family, which they share not a
t all with their twin; and there are aspects of the environment that are unique to themselves.
18. A diagram of the environmental and genetic influences on intelligence for non-identical twins reared apart.
Figure 18 pertains to non-identical twins reared apart. As with the identical twins reared in different families, these non-identical twins have no ‘shared’ environment with each other. Therefore, we can summarize the contributions to their intelligence test scores as follows: genes, which they share 50%; ‘common’ environment which they don’t share at all with their twin; and unique environment which of course they don’t share at all.