by Rik Smits
All the other body parts in which some kind of preference can be seen are used not so much to interact with our surroundings as to detect and observe them. In other words, they perform sensory rather than motor functions. The most prominent among them are the eyes and ears, and those are also the only two about which reasonably reliable facts on the matter are known. In both cases people have a clear preference.
We generally use one of our eyes for seeing and the other for measuring; one looks while the other mainly serves to measure the angle by which what we see differs between the two eyes, so that we can judge distance. This is known as depth perception. In many people, one eye works better than the other. Those of us who wear glasses or contact lenses know this, since the lenses are usually of different strengths. You might expect people to prefer to use their better eye as their seeing eye, but it turns out this is not so. It seems eye preference is settled early in life and remains the same even if the accuracy of one or both eyes changes, as often happens around the beginning of puberty. There seems to be an echo here of the stubbornness with which laboratory rats stick to their paw preference through thick and thin.
Eye preference can be tested in all kinds of ways, but most methods are less than reliable because measurements are distorted by hand preference. There’s not much point in taking account of which eye someone uses to aim a gun or a bow and arrow, for instance, since hand preference determines how such weapons are held and therefore which eye is used. What does work reasonably well is to observe how a person looks through the lens of a telescope or microscope, or the viewfinder of a camera. In the case of small children a pleasingly simple test has been developed. All you need is a hollow tube, a rolled up newspaper for instance. You look through one end of the tube and ask the child, who is moving around the room freely, to look back at you through it. The eye that appears at the other end of the tube is the child’s preferred eye. Although various studies into eye preference have produced markedly different results, it does seem that almost everyone consistently favours one eye or the other. Roughly two-thirds of us give priority to the right eye, the rest to the left.
Ears are another matter entirely. We don’t know whether we have a true ear preference for the simple reason that it’s almost impossible to think of a test that could determine which one it is. There’s certainly no test that could do so without any interference from other preferences, such as right- or left-handedness. All the same, it’s clearly a point worth examining, since in detecting sounds, two perfectly healthy ears turn out to work differently. Whether one or the other is dominant depends on what sort of sound they’re picking up. In general it seems we hear sounds that have to do with language, in other words speech, better with the right ear than with the left.
The most important means by which this has been studied involves what are known as dichotic hearing tests. Simply put, the test subject wears headphones and different words are sent to each ear at the same time. The subject then has to repeat what he or she has heard. It often turns out that words fed into the right ear have stuck in the memory rather better than those delivered to the left. If the sounds are unconnected with language, musical tones for example, then the results are the opposite, though the difference is not very great.
We need to treat these results with some caution. Dichotic tests place quite exacting demands on the test subject. It’s therefore perfectly possible that the effect arises in part for spurious reasons. In an attempt to produce an acceptable answer, a subject may for example come up with clever strategies that ultimately have nothing to do with the nature of the material presented. The differences do seem significant enough to be taken seriously, however, so we need to ask what the causes could be. We are not dealing merely with differences in sensitivity, since a less good ear hears both speech and other sounds less well.
The only remaining possibility is that this has something to do with differences between the two halves of the brain as they process sound. Each ear is directly connected to both cerebral hemispheres. We might therefore assume that the shorter connection has precedence, either because it’s quicker or because it’s less subject to disturbance along the way. Yet this is not the case. Speech sounds are received better by the right ear even though the centres that process speech are usually on the left. Conversely, the right half of the brain is more involved in processing melodies and yet we hear musical sounds better with the left ear. It therefore seems as if the nerve pathways coming from the opposite ear drown out those that run to the closer side of the brain, rather in the way that adjacent telephone lines can disturb each other. Perhaps we don’t so much have an ear preference as one ear that’s better at hearing certain sounds, depending on the place in the brain where that kind of sound is processed and the architecture of the specific nerve circuits that connect our ears with those processing centres.
If ears distinguish between linguistic and non-linguistic sounds, might not something similar happen with vision, in the sense that letters and words are more effectively viewed using one half of the retina and non-linguistic, or perhaps we should say non-symbolic pictures and patterns are better processed by the other? The suggestion seems natural, but it’s difficult to prove or disprove. Attempts have been made, but so many interfering factors and uncertainties come into play that we wouldn’t be justified in drawing any firm conclusions.
The existence of such a broad range of asymmetries naturally raises the question of whether there’s any connection between them. Is a left-handed person also left-eyed, left-footed and left-tongued? And if there is such a connection, does it have to do with the way the brain is org a n ized? Could it be, for example, that in many respects left-oriented people have a distribution of functions between the two halves of the brain that approaches the opposite of that found in the rest of us? One person who was deeply convinced that a strong correlation existed between the various forms of left and right orientation was an American called Beaufort Sims Parson. In 1924 he published a foolproof method of determining innate left-handedness with a machine he had built himself, the manuscope.
It was developed to detect a person’s hand preference objectively and reliably. After all, not only are people less than entirely trustworthy in answering questions about their hand preference, they are sometimes pushed in the direction of right-handedness by their environment, even more so in Parson’s day than in our own. This happens mainly, although not exclusively, at school. At home too, since most parents are right-handed, children are shown how to do things with the right hand and many appliances are designed to be used by right-handers. Conversely, it’s possible that a child of left-handed parents who is essentially right-handed might behave in some respects like a left-hander.
This kind of social pressure is entirely absent when it comes to eye preference. We are generally not even aware that we have such a thing. Parson not only believed it was connected to hand preference, he saw it as the cause. This idea had already been enjoying a degree of popularity for some years. He therefore felt that with his manuscope he had created an Egg of Columbus. Once eye preference had been determined, Parson reasoned, it would be clear whether the person in question was left- or right-handed.
Unfortunately for Parson and the other adherents of the eye-preference theory, it later transpired that their assumption was simply wrong. A majority of us are right-eyed, just as a majority are right-handed, but the proportions are quite different. Moreover it turns out there are not significantly more left-eyed people among the left-handed than among the right-handed. People have a hand preference and an eye preference, but it seems the two preferences develop entirely independently of each other.
More or less the same applies to feet and legs. The right-legged seem to be in the majority, but again the proportions are different from those we see with hand preference, nor is there any direct, demonstrable connection between leg preference and hand preference. It may even be that leg preference is a product of chance, like paw preference in animals. Som
e people say our preferred leg is the one we set off walking with, but what does such a thing really demonstrate? The only activity that brings to light a difference that looks even remotely like hand preference is soccer, and most people can’t kick a football particularly accur ately with either foot. It may well be that the roughly 30 per cent of us who are left-footed are matched by a roughly equal number of right-footed people, with a large group in between who have no preference at all. A proportion of 30 per cent is not too far from the one in four that characterizes a chance distribution.
As far as the tongue is concerned, our knowledge is extremely limited. It would be a little bizarre to ask large numbers of people to sing songs or recite poems with half their tongue between their teeth, and who can say whether or not it would render up anything useful?
28
Tallying Up
What is it exactly that’s so irresistibly funny about the eternally squabbling cartoon duo Tom and Jerry? Surely it has a great deal to do with the inexhaustible supply of unpleasant surprises, like the door with a brick wall behind it. Less funny was the discovery that medical and psychological breakthroughs of the second half of the nineteenth century were rather like the opening of exactly that kind of door. It was clear that hand preference had something to do with specialization by the two sides of the brain, but it would be a long time before the brain became accessible for direct investigation.
So people resorted to indirect methods, using psychological and statistical research to chart the traits that accompany left-handedness more often than we would expect based on chance alone. Scientists hoped this would at least help to determine whether hand preference was inherited and whether left-handedness was an abnormality, perhaps one that pointed to other problems. Although this would not lead to an explanation of the origins of hand preference, it might perhaps teach us something about the reasons why to this day one in ten newborn babies are left-handed. Over the years this approach has produced an impressive pile of reports and research data.
For practical reasons many such studies focus on groups that already differ from the norm in some way: children at schools for the disabled, people undergoing hospital treatment, residents of care homes or other institutions, prisoners, and pupils with all kinds of problems who have come to the attention of school doctors and psychologists. One major advantage of this approach is that you don’t need to go looking for subjects. You can concentrate instead on groups that tend to live in organized contexts, in buildings that are easily accessed by the staffs of universities, institutes and hospitals. Their lives are regulated by systems in which it’s normal to be investigated, so studying hand preference is a relatively effortless business, and they generally have little to do, which makes them ready and willing to take part in innocent-looking research projects. It’s a welcome diversion, something to distract them from their less than cheering daily routine. Furthermore, all kinds of medical and psychological data are available on such people, so it’s easy to make comparisons. All this has the incidental but far from trivial advantage that the research can be kept relatively cheap.
There’s another reason for concentrating on these particular groups, one that’s at least as important: the existence throughout history of preconceived ideas and prejudices about left-handed people, beliefs reinforced by the discovery of lateral specialization in the brain. Left-handedness is a minority phenomenon, and because aberrations are negative it demands to be corrected almost by definition. This is one of the many ways in which our tendency to think in opposites, in ‘we’ who conform to the norm and ‘they’ who are different, influences our attitude to our fellow human beings.
The crudest example of supposedly scientific research in the recent past that was in fact based primarily on unshakable prejudice – and not only in relation to left-handedness – is the work of Lombroso, who believed he could reliably read off character traits and aspects of a person’s make-up such as intelligence, trustworthiness and criminality from external features, including the shape of the face and head, body posture and, naturally, hand preference. Around 1900 he turned his attention to left-handedness and promptly found significantly elevated rates of it among criminals, especially female criminals. His categorization of left-handers as among the more civilized kinds of miscreants was presumably meant as a compliment. No fewer than one in three swindlers and racketeers were left-handed, he said, whereas in respectable people the rate was one in twenty. Murderers and violent assailants, the great skull-measurer claimed, display a far less striking tendency towards left-handedness, with figures of no more than about 9 per cent.
Lombroso’s conclusions turned out to be incorrect whichever way you look at them and his work has long since been consigned to the graveyard of science, but other consciously or unconsciously biased research persisted. In 1911, for example, psychiatrist E. Stier, on behalf of the German high command, carried out research into left-handedness in the army. Stier had published work on the subject before and he believed it was an inherited characteristic particularly common in primitive peoples. This not only reveals the man’s prejudice, it shows he was the kind of academic who paid little heed to facts and figures. We should therefore not be surprised that on average he found only 4 per cent of people to be left-handed, with the exception of the most stupid soldiers, among whom he discovered more than three times as many. Abram Blau, American psychiatrist and scourge of the left-handed, warned young parents as recently as 1961 in a newspaper article: ‘Don’t let your child be a leftie!’ Even the great British child psychologist Cyril Burt had an attitude that could hardly be described as free of value judgements. In his monumental 1937 work The Backward Child he characterizes left-handers as follows: ‘They squint, they stammer, they shuffle and shamble, they flounder like seals out of water. Awkward in the house, and clumsy in their games, they are fumblers and bunglers in everything they do.’ Talk about prejudice. A measured assessment is the last thing you’d expect after that.
The faces of German criminals illustrated in Lombroso’s standard work L’Homme Criminal. They include shoplifters (A), murderers (E), pickpockets (H) and burglars (I). The rest are all crooks of various other kinds. According to Lombroso there must be around twenty left-handers here.
Lombroso, Stier and Blau aren’t exactly shining examples to the rest of science – no sensible person would take their tall tales seriously today – but Burt is a different matter, since although controversy once surrounded the reliability of his twin studies, he was generally regarded as a meticulous researcher and is still held in high regard. That a serious scientist could allow himself to be carried away to such an extent by traditional prejudices, without noticing that in daily life he came upon so few of those floundering bunglers and squinting stutterers, demonstrates how perfidious prejudices can be and how hard it is to break free of them. Stereotypes undoubtedly had an influence both on the choice of research topics and on the interpretation of the data generated.
Over the past 70 to 80 years many hundreds if not thousands of studies, large and small, have examined the rates of occurrence of left-handedness in specific groups of people. The results are by and large disturbing, at first glance at least. Whatever the abnormality present in the group studied, there almost always seem to be more left-handers than would be expected based on chance alone: stutterers, dyslexics, children with special educational needs, sufferers from hay fever, asthma, allergies and other autoimmune diseases, epileptics, breast cancer patients – in every case they included a remarkably high proportion. There’s even a small study that suggests a clear link with alcoholism, another in which an unmistakable correlation is found with criminal behaviour, and a paper in which left-handedness in the more deprived neighbourhoods of a major American city is associated with heavy smoking.
Just as one swallow doesn’t make a summer, one study doesn’t make a hard and fast rule, even if the media often imply that it does. The results of this kind of research do something else entirely. They indicate t
he likelihood that an observed correlation between two phenomena is the result of pure chance. If that likelihood is small enough, then the scientist has a significant result and will write an article for an appropriate scientific journal, so that fellow scientists can work out whether or not what he or she says is true. This certainly does not amount to firm proof that the correlation exists in reality, let alone that one phenomenon causes the other. The publication of research results can better be compared with a decision by the police that there is sufficient evidence to arrest someone as a suspect in a criminal investigation. It indicates a reasonable suspicion of involvement, but no more than that.
No study is any more reliable than the assumptions on which it’s based. Errors of logic that creep in while an experiment is being set up, or while the data are being processed and interpreted, make the results as worthless as a car with wooden cylinders under its gleaming bonnet, or a computer with woollen circuits. Worse still, as long as no one brings those mistakes to light, the results will mislead. This is the origin of countless spinach-causes-cancer stories. Carelessness in carrying out experiments has a similar effect, as does the use of instruments that don’t do exactly what a researcher thinks they do, or reliance on assumptions that don’t square with reality. In short, it’s not at all easy to design and implement even the simplest study, and even the most careful, meticulous stickler of a researcher can run into problems with the most danger ous assumption of all: that he’s testing a representative sample.
In practice it’s hardly ever feasible to involve every member of a relevant population in research. Only more or less long-term, hugely expensive population surveys can do that, and they are the exception. They’re usually set up to look at common, fatal diseases such as breast or cervical cancer, and even so they are extremely rare. Researchers usually work with a manageable number of people who are presumed to be representative of the population as a whole, but it’s difficult to be sure whether this is truly the case, if only because no one knows exactly what a normal person is.