Intelligence_A Very Short Introduction
Page 7
Almost as soon as researchers were able to collect these diagrams of brain responses, some asked about individual differences between people. That is, they noticed that the peaks of the waves in the panel occurred after a shorter time in some people than in others. In some, the peaks were taller than in others. Perhaps brighter people, then, had a faster brain? Perhaps their electrical responses to events were that bit faster than people with lower intelligence test scores? Before going into that possibility and some others, it’s useful to inform you that my colleague Peter Caryl and I tried twice during the 1990s to read and assess all research studies ever reported in the scientific journals that looked at psychometric intelligence and the brain’s electrical responses. What we found and reported to the research community was, frankly, a mess. It’s probably fairly obvious that these types of study are technically difficult to set up. They need specialized equipment and there are many different ways they can be done. The problem was that we found hardly any studies that repeated the same procedures, so it was hard to draw out any well-attested, replicated results. We did, though, find some hints at regularities in the research. I would stress that these are indications only: none of them is certain, but all are interesting possibilities for making some links between the brain’s activity and the scores on intelligence tests.
First, the timing of the peaks and troughs of the electrical response. Some researchers got the idea that cleverer people had a faster brain electrical response to simple stimuli. There does seem to be some evidence that these occur some thousandths of a second earlier in people with higher psychometric intelligence. So, if the trace shown in Figure 12 was an average person, a brighter person might, on average, have their electrical peaks occur a bit to the left. The largest focus of research here has been the timing of the P300 wave peak, which might appear at a bit less than 300 milliseconds in higher intelligence test scorers, and a bit more in lower scorers.
Second, the overall complexity of the electrical response. Some researchers got the idea that brighter people had a more consistent brain electrical response to stimuli. Therefore, all of the 50 or 100 or so electrical responses to, say, an oddball tone would be very alike in a person with high intelligence. When averaged up, they should retain much of the complexity of the originals. On the other hand, people with lower mental test scores were thought, perhaps, to have more variable electrical responses. Therefore, when their responses were averaged, a cruder, less detailed waveform resulted. Quite a few research teams have tried to test this idea – that on average brighter people have a more complex brain response and less bright people have a simpler-looking response. It even has a popular name, the ‘string length measure’ – if one laid a string over a more complex (more squiggly) electrical response, then it would be longer than if one laid a piece over a simpler response. The result? Hard to tell. Some studies suggest that this idea does work and some that it doesn’t. Researchers are trying at present to find out why this discrepancy exists.
Third, the shape of some selected parts of the electrical response. Some researchers got the idea that more intelligent people had a differently shaped brain electrical response to simple stimuli. Again, take a look at Figure 12. Notice a large kink in the trace, from N100 to P200, where the electrical voltage swings from negative (going down) to quite a high positive value. That excursion happens between about and of a second after the event – that is, after the oddball sound (or whatever) occurred. We think that this electrical activity is something to do with our attempts to tell one thing from another, in making a simple discrimination. It’s been noticed that some people have steeper swings in this part of the electrical response than others. In other words, the slope that climbs from N100 up to P200 is steep in some people and flatter in others. And it seems to be the case that, on average, people who score better on intelligence tests have steeper slopes in this part of the brain’s electrical response to simple events. This statement is based on just a few studies, and the results require exploring in more and bigger groups of people.
The efficiency of visual processing
There is a well-established, moderate association between the efficiency of the early stages of visual perception and intelligence test scores.
Think of the situation where you enter a dark room, put on a light briefly, and then switch it off again. For a short time after returning to darkness you have an image of things in the room, a fleeting, fast-dissolving ‘shot’ of the scene. That very brief access to visual information after the actual stimulus has been taken away is called our iconic memory and it lasts just a fraction of a second. Next think of the situation where you are watching the television and something – an image or a word perhaps – flashes up on the screen very quickly and then disappears. Pop music videos are common culprits in producing this sort of event. In a group of people, some people will catch the information and some will not. It will have come and gone too fast for some and not for others, even if everyone attended closely. Therefore, there might be individual differences in how efficiently people extract information from iconic memory, and researchers, including me, have asked whether this relates to intelligence differences. The test we use most often is called inspection time.
13. (a) The stimulus materials used in the inspection time test that measures people’s ability to process visual information quickly. (b) The stimulus is followed by a ‘masking’ figure that has thick lines of equal lengths.
Look at Figure 13a. Notice the simple shapes with two vertical lines that are joined at the top. One of the vertical lines is longer then the other. In one of the images the long line is on the left and in the other the long line is on the right. When you look at each image you’ll find it very easy to tell whether the long line is on the right or the left because there is a large difference in the lengths of the lines. These two simple shapes are the stimuli we use in inspection time tests. What we do in this test is randomly present one or the other of these two shapes to a subject and ask them to tell us whether the long line was on the right or the left. Now, the immediate problem you will see with this test is that everyone will get all of the answers correct, because the question is very easy. There are two ways used to make it harder. First we can flash one or other shape to the person for a short amount of time, measured in thousandths of a second. If it is presented before the eyes only fleetingly, it is harder to tell where the long line was. Also, immediately after the shape is shown, we can remove it and replace it with another type of image, something to ‘wipe’ the impression of the figure from the eye and its brain. This second, interfering image is called a ‘mask’ and one type of mask is shown in Figure 13b. It has thicker lines and the lines are the same length.
Let’s look in detail at what happens to a person taking part in an inspection time test. Usually they sit in a quiet, dim room in a psychology laboratory. They look at a screen that is about 50 cm away – this might be a computer screen, or a panel of light-emitting diode lights, or a screen on some special device. They get a brief warning that something is about to happen, usually a little cross-hair or a dot on the screen. One of the two shapes at random from the section (a) of Figure 13 appears on the screen only briefly. After the figure is taken away, the experimenter replaces it with the masking shape (shown in Figure 13b). The person tells the experimenter whether the long line was on the left or the right. The experimenter records whether that answer was correct or wrong. Now, it is especially important to appreciate that the person giving the answers does not have to answer quickly or within a given time. The experimenter only needs to know whether or not the person is correct, not how fast they responded.
The test is repeated, sometimes hundreds of times. About half of the time the long line is on the right and half on the left, but it is not possible to predict the order. The shapes are shown for varying lengths of time. Sometimes the shape is shown for a long time, for example a quarter of a second. Almost no one will make errors when they see the stimulus for that length of
time. Sometimes the shape appears for just a few thousandths of a second. In that case, no one will be able to ‘see’ the stimulus at better than chance level. (Note that even just guessing will get the correct answer 50% of the time.)
What we find in this test is that, as the two-line shape is presented for longer times, the person is more likely to be correct in identifying the position of the long line. But we also find striking differences between people in how well they do on this simple inspection time test. Some people can report the position of the long line accurately even when it is shown only briefly, whereas others are guessing at no better than chance at the same duration. Therefore, researchers wondered whether there was a relation between this simple aspect of the efficiency of visual perception and intelligence test scores. The first studies of this type were done in the mid-1970s by Ted Nettelbeck and his colleagues at the University of Adelaide, and to date there have been dozens of other studies, involving many hundreds of people in four continents. The overall answer is yes, there is a moderate association between how good people are at the inspection time test and how well they score on intelligence tests. The correlation is about 0.4. People with higher intelligence test scores seem on average to be more efficient in processing visual information when it is presented only briefly. They can accurately tell what has been shown to them when others see only a blur. Therefore, this test of the efficiency of processing simple visual information relates to human intelligence differences.
How much does that tell us about what it means to be clever, at least as defined by a score on intelligence tests? Some researchers say it tells us quite a lot and some say it tells us not very much. Some take the view that the inspection time test is a simple function, an indicator of some basic limitation in the brain’s ability to cope with incoming information. They have even compared it to the clock speed of a computer: that is, the people with the better inspection times have been likened to computers with faster clock speeds. They go about the world taking in and chewing up information at a faster rate than others. There is support for this view from various sources. People’s inspection time slows as they get older. There are studies of illnesses and chemical compounds that slow down inspection time; and these factors also seem to affect psychometric intelligence. So, maybe, then, one smallish contribution to being brighter is having a brain that can process simple information rather quickly. This would agree with the old adage that people who are brighter are ‘quick on the uptake’. Essentially, the more intelligent person might be able to sample the world faster, making distinctions that go by too quickly for others.
But it is only fair to tell you that there are other views. Those who read the research reports firmly accept that there is an association between inspection time and intelligence, but some psychologists explain this in a different way. They say there could be other reasons for brighter people doing well on inspection time tests – it might have nothing to do with how fast their brain processes information. It could be that more intelligent people are more motivated or more relaxed or quicker to learn any task. Therefore, inspection time might be just another thing they do well because they try harder, or they don’t get so nervous in the lab, or they pick up the idea of the task better and quicker. If any of these ideas were true, it would be the case that inspection time was really acting just like an intelligence test, and not testing something basic about the brain. Another view is that people with better intelligence test scores might find some trick or strategy for doing the inspection time test better. Even though the task is meant to be a simple one that everyone does in the same way, the brighter person might pick up some strategy that gets them a better score: nothing, then, to do with how fast they process simple visual information. For example, some people can spot a small apparent movement after the two-lines figure is removed (with the best equipment this does not happen), and they try to use that to make better decisions. It would be fair to say that there have been some (but not enough) attempts to test the alternative ideas I have outlined in this paragraph and that there is not much evidence – if any at all – to support them.
Two views, then, about why people’s inspection time differences have a moderate association with intelligence test scores. (1) It’s because the less intelligent person’s brain processes information at a slower rate on average. In this view, inspection time would be a cause of intelligence differences; just one among others, obviously, but this would be an important finding. (2) It’s because inspection time is essentially just another test that bright people find a way to do relatively well. In this view, inspection time would be merely a symptom or consequence of intelligence differences. Who’s right? We do not know for certain. On balance, there is little evidence for the latter view, but that’s partly because these ideas are vague and hard to test in experiments. At present it is worth keeping open the possibility that we have discovered a way of testing some important limitation of the brain’s ability to make discriminations and decisions. It is exciting to have found psychometric test scores relating to something that at least looks very simple. Researchers need to do more digging to find what it is about the brain that causes these differences between people in their ability to cope with simple information.
Reaction time
People with higher intelligence test scores have, on average, shorter and less variable reaction times.
Look at Figure 14, which shows a box that measures people’s reaction times. First, let’s describe the equipment and how it is used by experimenters. Recall that inspection time was intended to assess how well people could make a discrimination when the visual figure was shown for only a very brief time. It was to do with speed of taking in visual information from the world. There was absolutely no need to respond quickly. Reaction time, on the other hand, is to do with how quickly people can make a correct physical response to a signal. The box shown in Figure 14 has eight buttons arranged in a semi-circle. At the bottom of the box there is one other button, equal in distance from the other eight. This separate button is called the ‘home’ button and the other eight are called ‘target’ buttons. Each button contains a light and is also a press-down switch.
14. The reaction time equipment used to measure people’s reaction times, decision times, and movement times.
Here’s the sequence of events when someone is having their reaction time measured. The person being tested places their preferred finger on the home button. One of the eight buttons in the semi-circle around the home button lights up. The person, as quickly as possible, lifts their finger off the home button and presses the target button that lit up. The process is repeated dozens or more times.
Here’s how the person’s reaction time gets measured in that simple event. When the target light comes on a timer starts immediately. The timer only gets switched off when the person being tested presses the target button down. The time that lapsed from the target light being turned on until the person pressed down the target button is the person’s reaction time. As a general guide, the time this sort of reaction takes ranges in different people from under a second to about of a second. Because not every single reaction is identical, the dozens that are collected by the experimenter are used to calculate an average for that person. But note that all those separate reactions can tell us something else, apart from the average. Some people are relatively consistent: their individual reaction times are all about the same, falling into a small range of values. Other people are more variable, with quite a wide spread of faster and slower reactions. Therefore, we can measure how fast a person can react on average and we can also measure how variable/consistent they are in their reactions.
Before we proceed to look at how well reaction time relates to intelligence test scores, I need to add a few more details to the reaction time test. First, note that in the situation I described the person was reacting to one out of eight lights. Because they have to press the correct button out of eight, they have to make a choice of which is the correct light, and
so this procedure is called choice reaction time. The choices can be any number, though two, four, and eight are the most commonly used in the choice reaction time procedure. When there is only one target button – imagine the box in Figure 14 with just the home button and one target button – the person just waits for the light to come on and responds to it. In that case, there is no choice to be made and the procedure is called simple reaction time.
Choice reaction time and simple reaction time form the basis of many different procedures in psychology. There have been measures of reaction time since the mid-19th century. There was some faltering interest around the start of the 20th century in whether reaction times, because they seemed so basic, were related to intelligence test score differences. But the work really began in earnest in the late 1970s and early 1980s when a type of psychology called ‘cognitive’ psychology came into vogue and began to study the timing of human mental processes. Since then dozens of studies, involving in total thousands of subjects, have looked at the association between reaction time and intelligence test scores. The researcher who brought reaction time to the study of intelligence and did much of this work is Arthur Jensen from the University of California at Berkeley. The finding: there is a small but consistent association between speed in simple and choice reaction time experiments and psychometric intelligence. The correlation is often about 0.2 or a bit higher. People with better intelligence test scores are, on average, faster in their reactions. Also, just as consistent a finding is that people with better intelligence test scores are more consistent in their reaction times. People who don’t score so well on intelligence tests have, then, slower and more variable reactions on average.