IELTS Academic Reading Practice Test 97

For centuries, people have dreamed of moving under the surface of the sea. The first known attempts came in the early 1600s, when a Dutch builder named Cornelis Drebbel made a wooden rowing boat covered with greased leather. It could travel a short way under the River Thames. Such craft were seen as oddities, not weapons. The first recorded submarine attack came during the American War of Independence. A one-man wooden boat called Turtle tried, without success, to fix an explosive charge to a British warship in New York Harbour.

Early boats were hard to control and could stay under only as long as their air lasted. Steam power could not be used while they were fully under water, and the air motors that were tried gave very little range. In 1800 an American inventor, Robert Fulton, built a hand-cranked boat called Nautilus for the French government, but it failed to attract sustained support from any navy. A real answer began to appear in the late 1800s, when good electric motors and storage batteries were developed. These allowed a vessel to run on an internal-combustion engine on the surface while charging batteries, and then switch to quiet electric power below.

The first boat to combine these systems well was designed by an Irish-American engineer, John Philip Holland. In 1900 the United States Navy bought his Holland VI and named it USS Holland. The boat was just over sixteen metres long, carried a small torpedo tube and could reach about five knots under water. Though simple by later standards, it set the basic plan used for nearly fifty years: a long hull with ballast tanks to control buoyancy, a petrol engine for surface travel (later replaced by diesel) and an electric motor for work under water.

During the First World War, submarines showed their importance for the first time. Germany used its U-boats to attack merchant ships in an effort to cut off supplies to Britain. In 1915 the sinking of the passenger liner Lusitania, with the loss of over a thousand lives, helped bring the United States into the war. Submarine warfare without limits proved highly effective in hurting trade, but it also turned neutral countries against Germany. By 1918, submarines had sent millions of tons of shipping to the bottom, and the weakness of surface ships had become clear.

Between the wars, designers worked to make boats faster and safer. The snorkel, a tube that let diesel engines run while the boat was just below the surface, was taken up by several navies. The German navy used it most during the Second World War. Advanced German designs such as the Type XXI, which came too late to change the result, had smooth hulls and larger batteries that allowed longer travel under water. After 1945, the winning powers studied these boats closely, and many of their features were copied in later national submarine programmes.

The next big change came with nuclear power. In 1954 the United States launched USS Nautilus, the first boat driven by a nuclear reactor. Because nuclear power does not need air, such boats could stay under water almost without limit, held back only by food supplies and crew strength. A few years later, the first boat able to fire missiles from below the surface entered service. These boats became the safest part of the nuclear defence force, because their ability to hide in the wide ocean made them very hard to find and destroy.

During the Cold War, submarines were split into two main kinds: attack boats, designed to hunt surface ships and other submarines, and missile boats, whose task was to stay hidden while carrying strategic weapons. Western and Soviet navies spent heavily on both. The growth of advanced sonar, long-range torpedoes and cruise missiles slowly turned the attack boat into a flexible tool of sea power. It could hit land targets hundreds of kilometres inland and also protect aircraft carrier groups.

Today, submarines also serve many peaceful uses. Research boats have gone down to the deepest point in the ocean, the Challenger Deep in the Mariana Trench. Tourist craft carry people to coral reefs and old shipwrecks. Special boats are used for underwater rescue, cable checks and scientific study. Small private submarines have even been used to smuggle drugs, and robot craft now handle tasks once done by crewed boats.

Several new tools are now shaping the submarine's future. Air-independent systems, which use fuel cells or Stirling engines, allow ordinary boats to stay under for weeks rather than days. This makes them quieter and harder to detect. Lithium-ion batteries hold more energy than older lead-acid types, so charging time falls and range rises. At the same time, craft without a crew are being built for long patrols. Some navies are testing large robot submarines that could operate for months with no one on board.

The submarine has evolved from a fragile oddity into one of the most complex machines ever built. The basic idea of controlled diving remains unchanged, but today's boats look nothing like Drebbel's leather-covered craft. Their journey from wooden shells to nuclear-powered stealth platforms mirrors the wider change in technology over four centuries. In the future, automation and new energy sources are likely to alter the balance between human and machine under the waves once again.

Twins have long interested scientists because they provide a natural experiment for separating the effects of genes and environment. Identical twins develop from a single fertilised egg and share virtually all of their DNA, whereas fraternal twins come from two separate eggs and share, on average, only half of their genes, just like ordinary brothers and sisters. If identical twins are more similar than fraternal twins on a particular trait, this suggests that genetic factors play a part in that trait. The English scientist Francis Galton wrote about twins in the 1870s, and his early interest helped launch a line of research that has made twin studies one of the central methods for investigating how much of human behaviour is inherited.

The classic twin method compares the concordance, or similarity, of identical and fraternal twins for a particular trait. For example, when one identical twin has a disease, researchers examine the probability that the other twin also has it and compare this with the corresponding probability for fraternal twins. A higher rate in identical twins indicates that genes play a role. From such comparisons researchers can estimate heritability, which is defined as the proportion of variation in a group that can be attributed to genetic differences. Twin studies have examined a wide range of traits, including height, body weight, intelligence, personality and susceptibility to mental illness.

Heritability is a population statistic rather than a fixed feature of an individual, so a high heritability does not imply that a trait is unchangeable or that environment is unimportant. For example, height is highly heritable in wealthy countries where most children receive adequate nutrition. In populations where many children are severely malnourished, however, the heritability of height would be lower because the environment matters more. Similarly, eyeglasses can correct poor eyesight even though vision problems have genetic components, showing that a trait can have high heritability and still be changed by a simple aid. These examples demonstrate why a high heritability estimate does not mean that a trait is fixed.

Twin studies also separate two kinds of environmental influence: shared and non-shared. Shared environment refers to factors that make children in the same family alike, such as parental income, housing and the local area. Non-shared environment includes experiences unique to each twin, such as different friends, illnesses, accidents or even different teachers at school. Research consistently shows that for many adult traits, shared environment has only a small effect, while non-shared environment and genetic factors are both significant. Shared environment is more important for some outcomes, such as school results in childhood and the risk of problem behaviour.

Twin studies rely on the assumption that identical and fraternal twins are treated in much the same way, which is known as the equal environments assumption. Critics argue that identical twins may be treated more alike because they look alike and often form closer emotional bonds. If this is true, their greater similarity could be due to treatment rather than genes. However, many studies have found that both types of twins are treated fairly alike, so the debate about this point continues, although the twin method remains widely used.

One of the most famous twin research programmes is the Minnesota Study of Twins Reared Apart, which began in the 1970s. It brought together twins who had been separated early in life and raised in different homes. The study reported striking similarities between some separated identical twins, including similar mannerisms, hobbies and even tastes in clothing. More formally, it found that identical twins reared apart were often about as similar as identical twins reared together, which strongly suggests genetic influences on traits such as personality and general intelligence.

Modern research has moved beyond simple estimates of heritability to examine epigenetics, which is how environmental factors can alter the way genes are expressed without changing the underlying DNA sequence. Identical twins start life with the same genes, but as they age and have different experiences, their epigenetic patterns diverge. This may help explain why identical twins sometimes develop different diseases, such as one twin developing schizophrenia while the other does not, despite their shared genetic risk. Epigenetic changes can be brought on by diet, stress and other life events, and studying these changes may lead to better ways to prevent disease.

Today, large twin registries exist in countries such as Sweden, Denmark and Australia, and some of these contain tens of thousands of pairs. This allows researchers to study rare outcomes and to combine twin data with modern genomic techniques. By identifying specific genetic variants associated with traits, scientists can examine how those variants interact with environmental measures, moving from broad heritability estimates to a deeper understanding of the mechanisms that link genes, environment and behaviour.

Twin studies still have limits, because genes may shape the environments that people choose, and this makes it harder to separate genetic and environmental effects. For example, a child who likes reading may ask for books, so genes help create part of the environment. Also, some twin pairs are not typical of the whole population. Even so, twin studies remain a central way to learn how nature and nurture interact, especially when combined with molecular methods.

Twin research has helped demonstrate that conditions such as autism, ADHD and depression have partly genetic roots, because for each condition identical twins are more likely to share it than fraternal twins. These findings do not point to a single cause but instead suggest a mixture of many genes and life events. They also help guide research into early help and treatment.

We live in an age that depends heavily on prediction. Governments ask economists to forecast growth, public health officials estimate the spread of disease, and businesses hire consultants to foresee market trends. Journalists interview political scientists who tell us which candidate will win, and think-tanks predict the consequences of policy changes. Yet a growing body of research suggests that experts, or at least those who make public predictions, often perform little better than chance. Understanding why this is so, and what we should do about it, requires looking closely at the nature of expertise and the limits of forecasting itself.

The most influential study of expert judgment was conducted over two decades by the psychologist Philip Tetlock. He collected thousands of predictions from hundreds of political and economic experts about the likely outcomes of world events. His conclusion was sobering: the average expert was only slightly more accurate than a dart-throwing chimpanzee, and even this modest advantage disappeared when the experts were compared with simple statistical rules. Tetlock also found that the more famous an expert was, the less accurate that expert tended to be, probably because fame encourages overconfidence and discourages changing one's mind.

Tetlock distinguished two styles of thinking among experts, which he called foxes and hedgehogs. The distinction comes from an old saying: the fox knows many little things, while the hedgehog knows one big thing. Hedgehogs are specialists who interpret the world through a single powerful theory, such as a particular economic ideology or a grand theory of history. Foxes, by contrast, are intellectually eclectic, drawing on many different sources of information and being willing to adjust their views. Tetlock found that foxes consistently outperformed hedgehogs, mainly because they were more likely to acknowledge uncertainty and revise their predictions in the light of new evidence.

Several psychological biases explain why experts often go wrong. Overconfidence is one: people systematically overestimate how much they know. In one classic experiment, when participants were asked to answer factual questions and then estimate the probability that their answers were correct, the average confidence was far higher than the actual accuracy. Experts are not immune to this; in fact, their specialised knowledge can make them more confident, not more accurate. Another bias is the tendency to see patterns in random noise, which leads forecasters to think they can predict events that are fundamentally unpredictable, such as the exact timing of a financial crash.

In addition, experts often fail to use base rates, the general frequency of an event in a population. For example, when predicting whether a particular start-up company will succeed, a technology expert may focus on the founder's enthusiasm and the product's novelty, ignoring the fact that most start-ups fail. A simple statistical model that uses only the base rate of failure would often outperform the expert. The same problem occurs in medicine: a doctor who relies on a vivid symptom may overlook the low prevalence of a rare disease, leading to a wrong diagnosis.

Another difficulty is that the world is not a stable system. Many social and economic events are influenced by feedback loops, where predictions themselves change behaviour. If a respected economist predicts a recession, consumers may reduce spending and cause a recession that would not otherwise have occurred. Or the opposite may happen: a confident forecast of growth may lead investors to take risks that destabilise the economy. Such reflexivity makes precise long-term prediction almost impossible, because the forecast alters the very thing it tries to predict.

Do these findings mean we should ignore expertise altogether? Not at all. Tetlock's later research, conducted in the Good Judgment Project, demonstrated that some people do forecast better than others, and that these skills can be taught. The best forecasters were those who thought in terms of probabilities rather than certainties, updated their beliefs frequently, and broke large problems into smaller, more manageable questions. They also sought out diverse viewpoints and were comfortable acknowledging when they had been wrong. These cognitive habits, rather than deep subject knowledge alone, seemed to be the key to their success.

For ordinary decision makers, the best advice is to be wary of any prediction expressed with absolute confidence, especially when it concerns a complex system. Ask whether the expert has a track record that can be checked, whether the prediction is falsifiable, and whether the expert has considered alternative outcomes.

Often, a range of possible scenarios is more useful than a single point forecast. Combining predictions from several independent experts, or using statistical models, generally produces better results than relying on any one voice. In other words, the future is uncertain, but our approach to predicting it can be improved by humility, diversity and a willingness to learn from error.