IELTS Academic Reading Practice Test 157

For most of human history, people with hearing loss had little to help them beyond cupping a hand to the ear. The first devices made for the purpose appeared in the seventeenth century and were purely mechanical. Known today as ear trumpets, they were funnels that collected sound over a wide area and directed it into the ear canal. They contained no electronics and could not amplify anything; their effect was to gather more of the sound already in the air. Some were small enough to be hidden in a hand, a fan or a hat.

Ear trumpets came in a remarkable variety of shapes. The simplest were cones of brass or silver; more elaborate versions used a tube to carry sound from a funnel to an earpiece. Makers advertised them alongside spectacles and artificial limbs, and wealthy customers could buy models disguised as walking sticks or built into the arms of chairs. They worked best when the loss was caused by damage to the outer or middle ear. For people whose inner ear had lost its sensitive cells, the gain was limited — and that was the great majority.

Electrical amplification arrived through the telephone. Alexander Graham Bell, who had taught deaf pupils and whose mother and wife were both deaf, was granted a patent for the telephone in 1876. Its transmitter turned the pressure of sound waves into a varying electrical current, and its receiver turned the current back into sound; within two years the transmitter had been improved by the carbon microphone. The same principle could make quiet speech louder, and within a few years experimenters were attaching telephone receivers to the ears of people with hearing loss. The carbon microphone was the critical component: it made boosting a weak signal possible.

The first portable electric aid is usually credited to an American inventor, Miller Reese Hutchison, who produced his device at the end of the 1890s. It combined a carbon microphone, a battery and an earphone in one case. The apparatus was heavy, needed a large battery and cost far more than an ear trumpet. Few were sold. Its importance lies in the principle it established: sound could be converted into electricity, strengthened and returned to the ear.

Vacuum tubes changed matters in the 1920s. A valve could amplify a signal far more cleanly than a carbon microphone, and the first commercial valve aid, made by Western Electric and named the Vactuphone, appeared in 1921. It was not an instrument anyone could wear discreetly: the case and its batteries were heavy, and users were advised to set the apparatus on a table in front of them. A few models were built into wooden cabinets the size of a small suitcase.

Miniaturisation then proceeded steadily. By the late 1930s an aid could be worn on the body, with a microphone and amplifier in a pocket-sized box and a thin cable leading to an earphone. Batteries remained bulky and were often carried in a separate pack strapped to the leg, hidden under clothing. Design became a small industry with competing manufacturers, though the devices stayed expensive and were still associated with old age and disability. Body-worn aids remained the standard design until the 1950s.

The transistor, developed at Bell Telephone Laboratories in 1947 by John Bardeen, Walter Brattain and William Shockley, transformed the field. It was far smaller than a valve, needed much less power and produced less heat. Because it drew so little current, it could run for days on a small battery. In 1952 the Sonotone company marketed a hybrid: a transistor and a vacuum tube in the same case. The first aid with no valve at all appeared the following year, and it was smaller still and far less demanding of power. Within a decade, valve aids had almost vanished.

The shape of the modern aid was settled in the same period. In a behind-the-ear model, the electronics sit in a curved case hooked over the top of the ear and deliver sound through a custom-made mould in the ear canal. In-the-ear versions, which appeared in the 1960s, put everything inside the mould itself. From the 1970s, integrated circuits allowed more functions to be squeezed into less space, and features such as volume control moved from a wheel on the case to a push button.

Digital signal processing was the next revolution. The first fully digital aid, the Senso, was launched by the Danish company Widex in 1996. Instead of making every sound louder, a digital aid could be programmed to suit an individual's audiogram, the chart of which frequencies a person hears well. It could also suppress acoustic feedback, the whistling that had plagued earlier designs, and use two microphones to favour speech from in front over noise from the side. Several programmes could be stored and switched by the wearer, for example between a quiet room and a busy street.

Today the technology is mature but far from universal. Modern aids stream speech and music from phones over Bluetooth, are recharged overnight like a mobile phone and can be adjusted with an app. Even so, surveys in many countries find that only a minority of people who would benefit actually use one; cost, comfort and the fear of appearing old are the reasons most often given. In the United States, a rule introduced in 2022 created a class of over-the-counter aids that adults can buy without a prescription or a medical examination.

A soft robot is a machine whose body is made not of metal and rigid joints but of rubber, gel, fabric or foam that bends, stretches and squashes. The term came into wide use in the 2010s, though engineers had built flexible machines long before. The inspiration is biological. An octopus has no skeleton, yet it can reach into a crevice, wrap itself around a rock and squeeze through a gap narrower than its body. A worm pushes through soil by changing the shape of its segments in sequence. Soft robotics asks how machines might move and grip in the same way, using the compliance of their own bodies rather than a motor at every joint. The result is a machine that can absorb a collision instead of being damaged by it.

The motivation is practical. A conventional industrial arm is powerful and precise, but it is also dangerous to anyone standing near it and hopeless with objects that bruise. A machine built of soft material can bump into a person without doing harm, so it can work beside them without a cage. A soft gripper can pick up a tomato, a peach or a raw egg without crushing it, and can hold an object of unfamiliar shape without being told its dimensions. Such work has usually been done by hand, and traditional automation has always struggled with it. Factories that pack food have therefore been among the first to use the new devices.

The commonest way of making a soft actuator is pneumatic. A block of silicone elastomer is moulded with a network of small chambers inside it. When air is pumped in, the chambers inflate and the block bends in a chosen direction; the pattern of the chambers, some reinforced with stiffer material, determines whether the finger curls, twists or extends. The idea is old. In the 1950s the American prosthetics designer Joseph McKibben built a sleeve that expanded when pressurised and shortened as it did so, and this artificial muscle has since been used in robots and in experimental suits for people with weak limbs.

Air is not the only option. Some soft actuators are powered electrically. A dielectric elastomer consists of a thin sheet of polymer sandwiched between two compliant electrodes; when a high voltage is applied, the electrodes attract each other, the sheet is squeezed thinner and it spreads sideways, producing movement. Others use shape-memory alloys, wires that shorten when heated, or tendons pulled by a motor. A quite different trick is granular jamming. A bag filled with particles such as coffee grounds is floppy while the air inside it is at normal pressure, but when the air is pumped out the particles lock together and the bag becomes rigid enough to hold a tool.

The best-known demonstration of how far the field has gone is the Octobot, built by a team at Harvard University and reported in 2016. It was a small, eight-armed machine with no rigid parts at all — no battery, no wires, no silicon chip. Its fuel was a hydrogen peroxide solution, which released gas when it met a catalyst. The gas inflated channels in the robot's arms, so that they moved in a steady rhythm. A soft fluidic circuit printed from the same rubber controlled the sequence. The Octobot could do nothing useful; it was a proof of principle.

Sensing and control remain harder than actuation. A soft body can be fitted with soft sensors: channels filled with a conductive liquid metal whose electrical resistance changes as the channel stretches. But the mathematics of a body that changes shape continuously is awkward. Engineers can describe the bending of a rigid link with a few numbers, whereas a soft finger has, in principle, an infinite number of possible shapes. Models therefore rely on simplification and on large quantities of data, and machine learning is increasingly used to work out what a soft arm is doing from the readings of the sensors embedded in it. That, in turn, means that the behaviour of a soft robot is often discovered by trial rather than calculated in advance.

Applications are multiplying. Soft grippers handle fruit, vegetables and bakery products, and are used in packaging lines where objects arrive in unpredictable positions. At the Wyss Institute in Massachusetts, engineers have developed soft exosuits — garments with cables that assist the wearer's legs or arms — for people recovering from a stroke. Soft robots small enough to travel through the gut are being explored for surgery and diagnosis, and segmented worm-like machines have been proposed for inspecting pipes and searching rubble.

The limitations are real. Soft actuators generally produce small forces, so a soft arm cannot lift what a metal one can; many designs need a tether to a compressor, which restricts where they can go; and elastomers fatigue, tear and degrade with repeated use. None of this means that soft robotics will displace conventional engineering. Its advocates argue, rather, that the two belong together: a rigid frame for strength and precision, and a soft hand or a compliant skin for the part of the task that requires gentleness. That combination, rather than a robot made entirely of rubber, is likely to be what most factories install.

The smart meter is a modest-looking box, usually fixed in a cupboard under the stairs, yet it represents one of the largest pieces of technical infrastructure that governments have asked households to accept. A conventional meter is read once a year and does nothing else. A smart meter records consumption at intervals, commonly every half hour and sometimes every fifteen minutes, and transmits the readings over a communications network to the supplier. Most installations include a small in-home display that shows what electricity is being used and what it is costing at that moment.

The idea is not new. The Italian utility Enel began replacing its customers' meters with electronic ones in 2001, and within a few years had installed tens of millions of them. Meter readers no longer had to visit, accounts could be settled without estimates, supply could be disconnected remotely, and faults could be identified from the data. Other countries watched, and by the 2010s large rollouts were under way in Sweden, Spain, Japan, Australia and Britain.

The argument made to the public was different, and it rested on feedback. The theory is that a householder who can see, moment by moment, what a tumble dryer or an old refrigerator is costing will use less. Trials have tested this, and the results are less impressive than the theory. Reviews of the evidence have generally found reductions of a few per cent, largest among households that had not previously thought about their consumption at all, and the effect tends to shrink once the novelty of the display wears off. A few per cent of a household bill is not nothing, but it is a long way from the promises made in some advertising.

The second argument is about the network. Electricity systems were built around large power stations that could be turned up or down to follow demand. As wind and solar generation grows, supply becomes more variable, and the ability to shift demand becomes more valuable. A smart meter allows tariffs that charge more at peak times, and it allows a supplier to reward a customer who charges a car overnight or lets a heat pump pause for half an hour. None of this requires the householder to watch a display; it happens through prices and automation.

These system benefits are, in my view, the strongest case for smart metering, and they are why I would not abandon the programmes. But they are benefits to the network, not to the household that pays for the meter. That distinction matters, because the money is real. In Britain, where suppliers were required to offer smart meters to every home, the programme has been repeatedly delayed, and its costs have been recovered through customers' bills. A household that already reads its own meter and already avoids waste pays more, and receives very little in return.

Privacy is the objection that deserves most attention. A reading every half hour is not a single number but a pattern, and patterns are revealing. Researchers have shown that load data can be used to infer when a house is empty, when its occupants go to bed, whether they cook or reheat food, and even which television channel is on, because different programmes draw slightly different amounts of power. The technique, known as non-intrusive load monitoring, requires no device inside the home: the meter alone provides the information.

Defenders of the technology reply that the data are encrypted, that they belong to the customer and that access by third parties requires consent. That is true in principle, and the rules differ from country to country. But consent is easily obtained in a tenancy agreement or a discounted tariff, and consumers rarely understand what they are agreeing to. The question is not whether the data are useful — they plainly are — but who may use them, for what, and for how long they are kept.

There are more mundane failures. In Britain the first generation of meters lost some functions when a customer switched supplier, reverting to a manual reading; standards for interoperability were written only after that hardware had been installed. Security is a further concern: a meter is a small computer attached to both a network and the electricity supply, and it must keep working when the network fails.

Fairness cuts both ways. A display is of little use to a tenant whose appliances are old and whose landlord pays for nothing, and the households with the least money to invest in insulation are the least able to shift their consumption to cheaper hours. On the other hand, prepayment by smart meter is less humiliating than a coin-operated box in a corridor, and remote reading means no more estimated bills for a family that cannot absorb a surprise demand.

My conclusion is therefore cautious. Smart meters are not the consumer revolution that their promoters promised, and the claim that they will pay for themselves in savings is, for many households, simply untrue. They are, however, a necessary piece of equipment for a system that must balance variable renewable supply with demand, and they can make billing fairer for those who struggle most. The case for them should be made honestly, on those grounds, and not on the promise of a smaller bill.