Features

How good are we at predicting upcoming words during a conversation? If someone begins a sentence with the words 'You never forget how to ride', you'll be surprised if it doesn't end with 'a bike'. But if the sentence begins with 'You never forget how to ride an', then, phonologically, the bike is out of the question – but an elephant isn't.

In a new study published in the journal eLife, an international team including Oxford's Dr Matt Husband investigated, using neuroimaging techniques, whether our brains have the capacity to make very specific predictions about upcoming words, such as their initial sound.

Dr Husband, from the Language and Brain Laboratory in Oxford's Faculty of Linguistics, Philology and Phonetics, talks to Arts Blog about the research.

What was the prevailing wisdom on our ability to predict upcoming words?

For the last decade, our ability to predict upcoming words has become integral to our understanding of language comprehension. The idea that we predict upcoming words has gone from the fuzzy intuition that we all, more-or-less, seem to experience during conversations to a detailed mechanism proposing that we comprehend language in part by predicting what will be said next.

The key empirical questions of the last decade have moved from 'do we predict?' to 'what is it that we predict?' We might, for instance, only predict a very abstract, high-level meaning of what will be said, or our predictions could be quite fine-grained – perhaps even all the way down to the probability for what specific form a word will take.

Arguably the most high profile evidence for this 'specific word' view of prediction is DeLong, Urbach and Kutas' 2005 Nature Neuroscience paper titled 'Probabilistic word pre-activation during language comprehension inferred from electrical brain activity'. They investigated how the brain responded to the expectations set up by sentence fragments such as 'The day was breezy so the boy went outside to fly…' when it continued with the expected phrase 'a kite' or an unexpected phrase like 'an airplane'.

DeLong and colleagues cleverly manipulated the indefinite article in their study, taking advantage of a phonological rule of English in which the indefinite article is realised as 'a' before consonant-initial words and as 'an' before vowel-initial words. This allowed them to ask whether the brain was predicting the sound form of a word before actually seeing that word. They examined electrical brain activity elicited by articles that were comparable with the highly expected yet unseen noun ('a', followed by 'kite'), or by articles that were incompatible with the highly expected noun and heralded a less expected one ('an', followed by 'airplane'). They reported increased brain activity for articles that were incompatible with the highly expected noun compared with those that were compatible with the highly expected noun, suggesting that predictions could be very specific and fine-grained.

What does this new research find?

It is perhaps surprising given how foundational and widely cited the DeLong, Urbach and Kutas study is that a direct replication of this effect on 'a/an' had not been reported in the published, peer-reviewed experimental literature. The study reported in our paper is the first large-scale attempt to directly replicate these results, using both their original methods and analysis and new analysis techniques currently available to us. This was a massive effort, coordinated between nine UK university labs and collecting 334 participants – ten times what an experimental study of this type would normally have. The participants read sentences that were presented one word at a time, while electrical brain activity was recorded at the scalp. Each sentence contained an expected or unexpected combination of an article and a noun, such as in the kite/airplane example given earlier.

Surprisingly, even with such high statistical power, our study did not replicate the key finding of DeLong, Urbach and Kutas, suggesting that either the effect on 'a/an' is, in general, too weak to detect using current neuroimaging techniques, or that the predictions we make during language comprehension are not so specific as to single out a particular word.

What are the implications of this study?

Our failure to find the 'a/an' prediction effect stands as an important reminder that it should take a large body of research to fully convince a field to adopt a new theoretical proposal. The results of a single published study, even in a high-impact journal, need to be considered with care.

Our failure to find the 'a/an' prediction effect should also caution us when it comes to how detailed our predictions can be. It may be that we still predict broad meanings and perhaps some specific features of upcoming words, but very fine-grained, phonetically detailed predictions may be out of reach for the human mind.

The study was led by Mante Nieuwland, cognitive neuroscientist at the Max Planck Institute for Psycholinguistics (MPI) and the University of Edinburgh, and involved researchers from nine UK laboratories: University of Birmingham, University of Bristol, University of Edinburgh University of Glasgow, University of Kent, University College London, University of Oxford, University of Stirling, and University of York.

The Oxford laboratory research was supported by the John Fell Fund.

Dr Marco J Haenssgen discusses the application of management thinking to solving the growing global problem of antimicrobial resistance.

You may have heard about superbugs, drug-resistant bacteria, or antibiotic and antimicrobial resistance (AMR) – all referring to one of the most pressing health challenges that the world is facing currently. AMR is high on public health agendas, it has attracted several hundreds of millions of pounds of research funding, and it risks to become one of the leading causes of death in the world by claiming an estimated 10 million lives annually by 2050. The World Bank argues that this will have an economic impact similar to the 2008 global financial crisis. Poor countries will be hit hardest, but rich countries are by no means safe because drug resistance is a global problem and drug-resistant bacteria can also be imported from abroad. The UK experienced this very recently, for example.

AMR means that certain types of medicine become less useful. The problem arises when bacteria and other microbes develop a tolerance to antibiotics and other antimicrobial drugs, which happens for example if we keep using antibiotics for the wrong purpose, like to treat flus and colds. At the same time, new medicines to fight superbugs are still far on the horizon, and so diseases like tuberculosis are becoming more difficult to treat or even life threatening.

A part of the response to the superbug crisis therefore involves stimulating the supply of new antimicrobials, and reducing the demand for and unnecessary use of antimicrobial drugs. Typical suggestions to reduce the demand for antibiotics among the general population thereby include reducing infections through improved public health, and to increase public awareness about superbugs. As a social scientist, I would argue that this is unlikely to solve problematic antimicrobial demand and overuse in the general population. The problem is likely to persist even if everyone in the world was aware and educated about AMR because health behaviour is not solely driven by what we know (other determinants include e.g. poverty, lacking access to qualified doctors and nurses, fear, different cultural beliefs, or people’s understanding of what “good care” is).

The supply-and-demand definition of a market for medicine does not help to resolve this problem. The definition is common in neoclassical economists, where markets are defined as an allocation mechanism for products and services. Akin to a “market place,” supply and demand depend on the price of these goods and services. According to this simple model, we could lower demand for antimicrobials by changing people’s preferences, by ensuring that they don’t get sick so often, or by offering them other medicines instead. These suggestions are not insensible, but the focus on a single product or family of products is a barrier to understanding the nature of the demand for antibiotics among the general population, and to find more comprehensive solutions. We can find some impulses for an alternative in strategic management.

Business leaders know that they don’t compete only with comparable products for their customers. An example I have been trained with in management school is a construction firm in the Middle East (let this be our “Customer”). The customer wins a valuable contract by the government to build the next skyscraper, deadlines are tight and stakes are high. The construction firm will therefore have a demand for the best and most reliable construction equipment available (diggers, cranes, and all the other things that get boys excited about). Let a manufacturer of such equipment be our “Supplier 1.” Quite obviously, Supplier 1 competes with other manufacturers of equipment, for example on the basis of product quality, price, or other purchase-related services like quick order fulfilment.

Is Supplier 1 right in considering only the product market for construction equipment? What is it that actually matters to the Customer? Certainly no skyscraper construction can happen without equipment, and so to consider competing manufacturers of the same product is not absurd. But the important consideration for the Customer is to fulfil the construction contract without delay so as to avoid financial penalties by the government. That is why high-quality and breakdown-proof equipment is important, as it helps to limit the risk of delays. Similarly useful would be insurance to cover the penalties for delays, in which case our Customer could make do with cheaper equipment. Supplier 1 is therefore not only in direct competition with other equipment manufacturers, but also with insurance companies, and if the function of avoiding financial penalties for delays can be met by an insurance broker, then suddenly there may no longer be a demand for high-quality and reliable equipment by the Customer. The market does therefore not just comprise products, but more general functions that the customer aims to fulfil, and different types of solutions or technologies can fulfil these functions.

This is the conceptualisation of strategic market segments following Abell (1980), and one line of management teaching suggests that businesses should not only be concerned with competing producers of similar products, but indeed with solutions from other industries that help customers fulfil their needs.

Though it may appear “off the beaten track,” we can apply this definition of strategic markets quite usefully to AMR and the demand of antimicrobial drugs. If we consider the case of people’s antibiotic use, then the conventional supply-and-demand logic can easily trap us in a focus on prices, different types and brands of antibiotics, or perhaps relative prices with other medicines. A strategic market definition draws attention to other aspects: functions (the ultimate goal of taking medicine), technologies (the range of solutions to reach this goal, including medicine), and consumers (different segments of the general population). We could therefore consider:

• What function(s) do antibiotics fulfil when people demand them? For example, some people might just take medicine in the hope to get better quickly, especially if they cannot (afford to) take time off and their family depends on their work.

• What other solutions help people to achieve the same function(s) that antibiotics fulfil? If antibiotics provide peace of mind, then this might not be an intrinsic characteristic of antibiotics, but of receiving some form of pharmaceutical treatment more generally. The same peace of mind could be brought about by labour laws that provide paid sick leave, so people don’t have to worry about their income when they get sick.

• Do these functions matter equally to all consumers? Strategic marketing starts from the premise that consumer groups differ in their needs, and the functions of antibiotics may be distributed unevenly across a population.

If we apply this strategic management definition of a market, then we can broaden our understanding of and response to people’s antibiotic use. For example, while awareness campaigns might change some people’s behaviour, what we think to be superior knowledge or a better solution may not be deemed superior by the population groups whom we serve, so we need to understand their needs and objectives first. The reason for overusing antibiotics might not have been because people did not know what they were taking, but for example because they were desperately trying to keep working and sustain their family.

The strategic market logic thereby permits us to formulate new premises for analysing people’s medicine use. A selection of such premises is exemplified below:

As a simple frame of mind, the strategic management market definition and this list of premises can be useful to understand people’s demand for antibiotics, but it can also be applied to other health behaviour and interventions beyond AMR. Interdisciplinary approach like this – applying social sciences thinking to global health problems – thereby help us to understand why interventions fall short of expectations, and they can help to stimulate new ideas for action and interventions beyond awareness-raising and education campaigns.

Further reading:

Haenssgen, M. J., Charoenboon, N., Zanello, G., Mayxay, M., Reed-Tsochas, F., Jones, C. O. H., et al. (2018). Antibiotics and activity spaces: protocol of an exploratory study of behaviour, marginalisation, and knowledge diffusion. BMJ Global Health, 3(e000621). doi: 10.1136/bmjgh-2017-000621
Haenssgen, M. J., Charoenboon, N., Althaus, T., Greer, R. C., Intralawan, D., & Lubell, Y. (2018). The social role of C-reactive protein point-of-care testing to guide antibiotic prescription in Northern Thailand. Social Science & Medicine, 202, 1-12. doi: 10.1016/j.socscimed.2018.02.018

California, Brazil and South Africa have all recently experienced major drought, threatening serious disruption to supplies for major cities (‘Day Zero’ events). How can England prepare for drought without harming the environment or driving up water charges? 

Dr Matthew Ives and Mike Simpson of Oxford's Environmental Change Institute, discuss their research on strategic water planning - conducted with Professor Jim Hall and newly published  in the Water & Environment Journal. 

Many people find it hard to believe that a country so blessed with rain as England would have any need to undertake intensive water conservation measures. But, contrary to popular opinion, the United Kingdom isn’t as wet as some believe. In fact, some parts of England have rainfall rates per person as low as the world’s most arid regions, such as the Middle East.

Convincing people to use less water and investing in long-term leakage reduction solutions will be critical for the avoidance of drought-induced interruptions to water supplies for large numbers of businesses and households in England.

Additional consequences of failure to act would include high costs for new infrastructure, such as for desalination or transfer pumping, while the extra energy this uses may mean additional carbon dioxide emissions. These stark conclusions are the headline results from recently published research into future-proofing England against the spectre of severe drought.

This twin-track approach represents a bold challenge to the water engineering community. Technological and social solutions to address leakage and demand reduction already exist, with many currently implemented in the UK or overseas.

Smart metering, available on a voluntary basis in much of England, can drive down the costs of finding and managing leaks, as well as encouraging reduced use of water. Satellite and remote-sensing technologies pioneered in drier parts of the world, like Israel and California, can be used to identify leakage sites.

The sheer number of people in the relatively small urban areas of England require an enormous amount of water. Unfortunately, while many of the most densely populated areas are in the South and East, much of the rain falls in the North and West. One regularly proposed answer to this problem is to transport water across the UK, in particular from Wales and Scotland, to support temporary dry conditions in the Southeast of England. Could this pipeline idea be a solution? Maybe technologies such as desalination could be used? Or the development of a new generation of larger reservoirs? What about increasing the efficiency of our existing water system?

Developing solutions to meet England’s future water needs calls for a national perspective, which can answer strategic questions about our water infrastructure strategy. Using our purpose-built National Infrastructure Systems Model (NISMOD) we assessed all of the different investment options available to England’s water companies for future-proofing the country’s water supplies. With a twist. We included the options available to individual companies, such as reservoir extensions and desalination plants, alongside options requiring a national perspective, such as inter-company transfers and demand management campaigns. And we pitted all such options against the spectre of future uncertainty around climate change and population growth.

We termed this analysis ‘navigating the water trilemma’ as it involved finding solutions that not only provided England with future water security but solutions that were also affordable and did not put too great a strain on the natural environment. This study highlighted the value of the flexible, ‘trilemma-friendly’ options like leakage reductions and demand reductions.

Our analysis points to the unavoidable answer: leakage reduction and demand management are the most cost effective and widely applicable components of future water strategy for England. Early investment in both of these solutions would allow a sensible and frugal culture of water use to be developed without recourse to panic during the inevitable drought events, such as experienced in the summer of 1976.
When we look at the impacts of drought in places which have the resources of England but have not taken sufficient preparation, the results are clear.

In Australia, hugely expensive new desalination works were developed in response to an extended drought, with long-term costs to public finances. Over recent years in California, restrictions on water use have been seen as deeply socially disruptive. However, many Californians now see responsible water use as a normal part of daily life.

Our research and new modelling capabilities were used to great effect by the National Infrastructure Commission (NIC) in their assessment of England’s drought preparedness. Their analysis, produced on the basis of our work, proposes a dramatic and ambitious change in approach. The NIC concluded that the equivalent of an extra 4 billion litres of water per day would be needed across England in case of significant drought. The report proposed that two-thirds of this should be made available through developing efficient pipe systems as well as shifting to the lowest household water use rates in the developed world. The NIC recommended that this should be supported by transfers of water between regions and, where appropriate, new water infrastructure including reservoirs and water recycling schemes.

Without improved national co-ordination and large-scale investment in water supply, the NIC’s report suggests that large parts of the country have a one-in-four chance of having their water cut off during a drought. Emergency measures, such as road and ship tankers, could cost up to £40 billion up until 2050, while the costs of building greater resilience would cost only half this amount.

Improving water resource efficiency is a fascinating challenge with many lessons to be learned from around the world. Technological solutions including sensing and monitoring of water supplies can be complemented by social solutions such as education and identifying the factors that influence people to make better use of water. Organisations such as ECI and the Centre for Ecology & Hydrology are well-placed to influence how such ideas are researched and how this research can become reality.

With some planning and vision, water supply in England can be future-proofed and it doesn’t have to be expensive. Adequate early investment, the development of a culture of water saving and some new technological and social ideas should make our occasional long, dry summers something to look forward to. When the alternative is expensive, environmentally damaging short-term solutions and regularly running out of water, surely the choice is clear?

This article is based on research in the Water and Environment Journal (WEJ), and the National Infrastructure Commission’s report “Preparing for a drier future”

Marina Filip, Postdoctoral Research Assistant, and Feliciano Giustino, Professor of Materials, both in the Department of Materials, explain how elementary geometry and modern data analytics can be combined to predict the existence of thousands of new materials called ‘perovskites’, as shown in their recent publication in PNAS.

Perovskites are a broad family of crystals that share the same structural arrangement as the mineral CaTiO3 . The extraordinary appeal of perovskites is their unusual chemical versatility, as they generally can incorporate almost every element in the Periodic Table. This leads to an incredibly diverse array of functionalities. For example, two major scientific discoveries of our times prominently feature perovskites, high-temperature superconductivity in perovskite cuprates (Bednorz and Müller, Nobel Prize 1987) and the recent discovery of the perovskite solar cells (Snaith, University of Oxford 2012).

In our own study we wanted to understand what makes certain combinations of elements in the Periodic Table arrange as perovskite crystals and others not, and whether we could anticipate how many and which perovskites are yet to be discovered.

It turned out that Norwegian mineralogist Victor Goldschmidt asked exactly the same question in 1926. Based on empirical observations, he proposed that the formability of perovskites follows a simple geometric principle, namely: The number of anions surrounding a cation tends to be as large as possible, subject to the condition that all anions touch the cation. This statement is known as the ‘no-rattling‘ hypothesis, and essentially means that if we describe a crystal using a model of rigid spheres, in a perovskite the spheres tend to be tightly packed, so that none can move around freely. Using elementary geometry Goldschmidt’s hypothesis can be translated into a set of six simple mathematical rules that must be obeyed by the ions of a perovskite.

Goldschmidt’s hypothesis had been used in one form or another in countless studies over the last century, in order to explain the formation of perovskites in qualitative terms, but its predictive power had never been assessed quantitatively. We realized that unlike 1926, in 2018 we benefit from a century of research in crystallography, documented in publicly available databases of crystal structures, such as the Inorganic Crystal Structure Database, and more than 50,000 published scientific papers on perovskite compounds. Using internet data-mining and statistical analysis, we were able to collect and study a library of more than 2000 chemical compounds which are known to form in various crystal structures, and use them to test the predictive power of Goldschmidt’s hypothesis. We found that this very elegant geometric model is actually capable of discriminating between compounds which are perovskites and those which are not with a higher success rate than sophisticated quantum-mechanical approaches.

In our study we used this simple model to screen through nearly four million compositions, and predict the existence of more than 90,000 new perovskite materials that have not been synthesized yet. This library of predicted compounds offers the exciting challenge of uncovering the functionalities of these novel perovskites to the community working on the synthesis and characterization of new materials. Most importantly, our discovery may lead to the realization of entirely new functional materials for a broad range of technologies, from applications in energy, electronics and medicine.

The full paper can be read in the journal PNAS.

From defending a besieged castle to spying for an exiled king, a new project is discovering the untold story of female activism during the British Civil Wars...

It wasn't easy to be born female in the early modern period. The ideology of the era held that women should be controlled by their father or their husband, and that the ideal woman was obedient and quiet, concerned chiefly with domestic matters and the rearing of children.

But don't be fooled by the dominant narrative, says Dr Emma Turnbull, lecturer at Jesus College, Oxford. When you look more closely at actual women's lives, it's clear the picture is far more complicated.

As part of a Knowledge Exchange Fellowship with The Oxford Research Centre in the Humanities (TORCH), Emma has been working with the National Trust to take a closer look at the stories of some of the fascinating women associated with their properties. Focusing on the English Civil War and Interregnum period (1640-1660), she has been examining how some women were able to break out of the confining boundaries of their society's gender roles in this era of huge social upheaval.

Take Lady Mary Bankes, who defended Corfe Castle against the first of two Parliamentarian sieges, beginning in May 1643. Along with her maidservants and a small group of soldiers, she personally patrolled the battlements, heaving rocks and hot embers over the walls. She died in 1661, and a memorial plaque in the church where she was buried commemorates her 'courage and constancy above her sex'.

Then there was Katherine Murray, mistress of Ham House. Left behind when her husband, William Murray, a close personal friend of Charles I, went away with him to war, she was forced to rely on her wits to preserve the family property in Richmond, right beside the heart of Parliamentarian power in London.

Or her daughter, Elizabeth Murray, who took over the property after her parents' deaths. In the period of Cromwell's rule she became a kind of double agent, joining the secret royalist organisation the Sealed Knot to work towards the restoration of the monarchy, while also keeping the company of Cromwell. Later, along with her second husband, she would become one of Charles II's most trusted advisers.

What is particularly fascinating about all of these stories, according to Dr Turnbull, is how all of these women used and manipulated traditional notions of femininity to serve their own ends. This is something that has often been missed in traditional accounts of these women, which take their outward affirmation of conventional gender roles at face value.

Katherine Murray and Elizabeth Bankes, for example, claimed they were only safeguarding their children's or their husband's rights when they were defending their property. Although Elizabeth Murray was suspected several times of being a double agent, she was never intercepted. 'What she intends I have not learnt', Sir Richard Browne, Charles II's agent in Paris, reported in the autumn of 1656. Elizabeth thus skilfully exploited the inability of male agents, on both sides, to decode her political motives.

'Focusing on them as mothers or as wives or as domestic beings doesn't really do justice to the level of their engagement and activity. Often they represented themselves in a way that has led us to undermine them. And we've kind of fallen for that,' says Dr Turnbull.

It is also important to recognise that these women were a part of a wider female culture. For example, Elizabeth's later political manoeuvring would not have been possible without the example of her mother. 'It's clear that Elizabeth was learning her craft from her mother. There's a sense of continuity between strong women in the family.'

Dr Turnbull has also drawn parallels between a miniature of Katherine Murray, painted by John Hoskins in 1638 and kept at Ham House, and similar portraits of Charles I's Catholic wife, Henrietta Maria. The fact that Katherine should choose to present herself in an analogous way to the queen, despite their different faiths, suggests that there was a distinct feminine culture at court that bound women together across divisions of religion and politics.

Dr Turnbull hopes that this fresh perspective will inform how this period is presented at heritage properties in the future. Her work has already contributed to a new 'Object in Focus' tour at Ham House, telling Katherine Murray's story through her portrait miniature.

'So often women in the early modern period are presented, in heritage properties, as domestic beings,' she says. But what I wanted to emphasise in this project is that, in spite of the formal and informal barriers to their activities, women had a stake in the political conflict. Elite women, like Katherine Murray and Mary Bankes, were moving around the country, they were visible, and they had a role in protecting themselves and their property.'

Read Dr Turnbull's articles on Katherine Murray, Elizabeth Murray and Mary Bankes here and here.