Legislation to make trial data transparent could be threatened (Image: Christian Thomas/Getty)

A NEW front opens today in the ongoing struggle to improve transparency in medical research. The latest battleground is a legal challenge against a UK regulator by a company that runs clinical trials for pharmaceutical companies. Campaigners fear that victory in court could upend moves towards greater openness. It is important this does not happen.

The argument for greater transparency is well known: the results of up to half of all clinical trials are never published. Positive results are around twice as likely to be published as negative ones, so the evidence on which doctors base decisions is skewed, potentially portraying treatments as more effective, and less risky, than they really are.

In January 2013 the campaign was launched to push for change, headed by the charity (SAS) and campaigning journalist Ben Goldacre. With supporters now including the UK’s Medical Research Council, the National Institute for Health and Care Excellence and over 500 other organisations, AllTrials was instrumental in putting transparency on the agenda during drafting of new European Union clinical trials regulations.

The outcome was that from 2016 all trials in Europe will have to be registered in publicly accessible databases, and their results reported within a year of wrapping up. It was deemed a major victory.

Judicial review

However, Richmond Pharmacology, which runs small, early-stage clinical trials, including “first in human” studies, launched a judicial review in March, challenging the move towards greater transparency by the UK regulatory body for clinical trials, the Health Research Authority (HRA). At a late stage, Richmond asked the judge to rule that there was no general legal requirement for researchers to publicly register any clinical trial. This would contradict existing law and the new EU law that is due to come into force next year, says Sile Lane, campaign director for SAS. If the HRA lost, it could set progress on transparency in the UK back to a point before the AllTrials campaign began.

“We saw a serious risk that the public’s expectations of transparency, the hard won progress of two decades, and new advances in patient safety, would be unwittingly sacrificed in a squabble between a company and the regulator,” says Tracey Brown, director of SAS.

That’s why SAS will act as the voice of AllTrials at today’s court hearing in Manchester via written arguments. It has tried to get the scope of Richmond’s legal arguments narrowed to technical points that would not threaten trial regulation, create loopholes that future trials could slip through or nullify the new EU regulations.

It is hopeful that this will happen because Richmond has already narrowed its arguments – but it has also switched arguments four times already, “so we cannot make any assumptions,” says Brown. Richmond said that it could not comment because the case is still ongoing.

“A victory will be if the hearing is dull as hell,” Brown says. Let’s hope it’s a dull day in court then.

Simon Makin is a science writer based in the UK. His blog is

Talking in your sleep might be annoying, but listening may yet prove useful. Researchers have shown that sleeping brains not only recognise words, but can also categorise them and respond in a previously defined way. This could one day help us learn more efficiently.

Sleep appears to render most of us dead to the world, our senses temporarily suspended, but sleep researchers know this is a misleading impression.

For instance, a study published in 2012 showed that sleeping people can learn to associate specific sounds and smells. Other work has demonstrated that presenting sounds or smells during sleep boosts performance on memory tasks – providing the sensory cues were also present during the initial learning.

Cat or hat?

Now it seems the capabilities of sleeping brains stretch even further. A team led by from the Ecole Normale Supérieur in Paris trained 18 volunteers to classify spoken words as either animal or object by pressing buttons with their right or left hand.

Brain activity was recorded using EEG, allowing the researchers to measure the telltale spikes in activity that indicate the volunteers were preparing to move one of their hands. Since each hand is controlled by the motor cortex on the opposite side of the brain, these brainwaves can be matched to the intended hand just by looking at which side of the motor cortex is active.

Once the volunteers had repeated the task enough times for the process to become automatic, they were taken to a bed in a dark room. Here, they were instructed to continue the task as they drifted off to sleep.

Once the EEG recording confirmed they were asleep, the researchers presented the volunteers with a new set of words. The volunteers brains’ continued to respond in the same way – preparing to make the movement appropriate to each word’s category, even though they were no longer moving

their hands. Fresh words were introduced to ensure that the volunteers were still analysing the words’ meanings rather than merely responding to learned associations.

Automatic for the people

“This opens the door to a lot of questions about how much linguistic processing happens during sleep,” says at Northwestern University in Evanston, Illinois, who is . “That’s unexplored territory.”

Kouider suggests this unconscious processing is possible because the task can be automated in a way that bypasses the prefrontal cortex, a region known to be heavily suppressed during sleep. “When you sleep, some brain regions sleep, while others remain totally awake,” he says. “Sleep is much more local than previously believed.”

This hints at what the limitations of unconscious processing might be. The prefrontal cortex is critical for executive functions such as planning, problem-solving and task-switching. “When you have two tasks you have to switch between, I’m not sure you could do that [in your sleep],” says Kouider.

On waking, the volunteers weren’t able to recall any of the words they processed while asleep but Kouider’s group is now investigating whether the approach can be extended so that new information is retained. “If you have a learning procedure, if it’s automatised enough, and if it’s simple, you might be able to learn it even during sleep,” he says.

The team is also investigating more complex linguistic processing. “We’re now looking at whether you can process a full sentence while sleeping, and detect whether it’s meaningful or not,” he says. “Or whether you can even pull out information relevant to the sleeper from a mixture of voices.”

Journal reference:

A reimagining of the island setting of Shakespeare’s play (Image: RSC/Lucy Barriball)

, The Other Place, at the Courtyard Theatre, Stratford-upon-Avon, UK

A special retelling of The Tempest shows how people with autism may be able to tap into the rhythmic heart of Shakespeare’s plays

THE jury is out about the ability of drama to help people with autism, and plays by Shakespeare would seem to be unlikely candidates. But their intensity may help explore the emotional expression that autism can make difficult.

This is the aim of a new version of The Tempest, adapted for children and young people with autism by , an actor with the Royal Shakespeare Company. Her pioneering method relies on iambic pentameter, the verse form Shakespeare used. She also works with the Shakespeare and Autism project at Ohio State University, and the play is an RSC-Ohio co-production.

Hunter’s adaptation places the actors and young people around a reimagining of the play’s island setting (see picture). The performance is bookended by calming exercises that use the similarity in rhythm between heartbeat and iambic pentameter. At the start, for example, everyone taps a heartbeat on their chest as they chant “hel-lo”.

Then there are games, like “throwing” looks for others to “catch”. Others are based on emotionally charged interactions from the play. To end, everybody taps out a heartbeat “good-bye”.

Hunter says that Shakespeare explores how “we use our eyes and mind to find reason and love”, which is why she used these skills in games. She says his characters lend themselves to teaching self-expression. In The Tempest, Caliban personifies anger, and Ferdinand’s reaction to Miranda is one of literature’s more extreme examples of intense love.

Researchers at Ohio State are evaluating the methods. who plays Trinculo, is also director of the programme at Ohio’s theatre department. She says that participants show greater gains in social functioning and language skills compared with young people on a waiting list to join the project.

As we leave, a mother asks her son: “Was that fun?” He seems not to understand, but moments later, a sing-song voice echoes down the street: “Good-bye, good-bye…”

This article appeared in print under the headline “Acts of emotion”

Could drama workshops help children with autism-spectrum disorders? Results from a pilot study called suggests this might be the case.

The research involved 22 children aged between 7 and 12 and consisted of one 45-minute session every week for 10 weeks. During this time, groups of four children entered an enclosed themed environment, such as a forest or outer space. These environments were designed to engage all senses simultaneously, using lights, sounds, puppetry and interactive digital elements.

Trained performers used improvisation techniques to encourage the children to engage creatively with the environment and each other, both physically and verbally. The hope was that the sessions would help develop the children’s communication, social interaction, and imagination skills – the “triad of impairments” seen in autism.

Children were assessed before the intervention, and again between two and six weeks after the sessions ended. As well as looking at whether behaviours used to diagnose autism changed after the drama sessions, the researchers also assessed emotion recognition, imitation, IQ and theory of mind – the ability to infer what others are thinking and feeling. Subjective ratings were also gathered from parents and teachers and follow-up assessments were conducted up to a year later.

Their own story

At the early assessments, all children showed some improvement. The most significant change was in the number of facial expressions recognised, a key communication skill. Nine children improved on this. Six children improved on their level of social interaction. The majority of these changes were also seen at the follow-up assessments.

“If you look at other studies of this nature, that’s quite a significant effect,” says at the University of Kent, the lead psychologist on the project.

Project leader , a professor of performance at the University of Kent who presented the Imaging Autism results at the at Goldsmiths, University of London, earlier this month, credits the approach’s effectiveness to engaging all the children’s senses at once. “Children are moving and thinking and interacting at the same time,” she says.

“It’s an opportunity for children to create their own narratives in an unconstrained, unfamiliar environment,” says Wilkinson. “They find this empowering, and we know from the psychology literature that individuals who are empowered enjoy increased attention skills and an improved sense of well-being.”

Iambic calming

“I think it quite plausible that drama experience will aid at least some autistic children in learning about social interactions,” says of University College London. But she stresses that the small size of the pilot study and the lack of a control group means the study is not sufficient to draw conclusions from.

For the researchers, the results are encouraging enough to warrant further study. They plan to collaborate with psychologists at the Centre for Embodied Cognition at Stony Brook University in New York later this year on a larger, more rigorous study. “What we have demonstrated is that this is something that is well received and could work,” says Wilkinson.

Other groups are thinking along similar lines. A project at Ohio State University in Columbus, called , claims the similarity between the heartbeat and the rhythm of iambic pentameter can help children to feel safer when communicating. The at Vanderbilt University in Nashville, Tennessee, has also carried out a two-week at a summer camp for children with autism.

“The studies indicate we are all on to something,” says Shaughnessy.

Roughly are affected by autism spectrum disorder (ASD). Rates in many countries, including the US, have risen sharply in recent years but no one is sure why. It is still not clear whether this is prompted by something in the environment, increased awareness of the condition and changes in diagnoses, or a result of people having children later.

The environmental case is hotly debated. There is some evidence that maternal infections during pregnancy can increase the risk. Other studies have pointed to a possible link with antidepressants while others have looked at elevated levels of mercury. But determining prenatal exposure to any substance is difficult because it is hard to know what substances people have been exposed to and when.

To get around this, and colleagues at the University of Chicago analysed US health insurance claims containing over 100 million patient records – a third of the population – dating from 2003 to 2010. They used rates of genital malformations in newborn boys as a proxy of parents’ exposure to environmental risk factors. This is based on research linking a , including pesticides, lead and medicines.

Toxic environment?

The team compared the rates of these malformations to rates of ASD county by county. After adjusting for gender, income, ethnicity and socio-economic status, they found that a 1 per cent increase in birth defects – their measure for environmental effects – was associated with an average increase of 283 per cent in cases of ASD.

Rates of autism were several times greater in some counties than others, which the researchers interpret as being the result of environmental toxins.

“If toxins are driving the geographic variation in autism incidence, there are practical implications,” says Rzhetsky. “Parents should pay attention to their kids’ environment, policy-makers should clean our environment, and scientists should incorporate data about children’s environment in future genetic studies.”

However, not everyone is convinced. at the University of Oxford points out that . “I find the equating of rates of malformation with environmental causes quite worrying,” she says.

Alternative explanations

Rzhetsky counters that the clustering of malformations is harder to explain in terms of genetics. “Given how much the US population moves around and mixes, it seems unlikely that some counties would have several times as many genetic defects as others,” he says.

However, the variation across counties might also be due to differences in standards of medical care, says at the Broad Institute in Massachusetts, leading to more cases being picked up in some areas than in others.

What’s more, at University College London points out that because the study used insurance claims the sample is not random. For instance, if someone has a genital malformation as well as ASD, it might increase their chance of making a claim. “There are all sorts of potential biases,” he says.

“There’s clearly value in analysing these large datasets, but interpretation is challenging when it’s not done in a controlled research setting,” says Neale. “I wouldn’t bet against the stance that fetal exposure to toxic compounds is bad for development, but by itself, this study does not give us a clear picture of the extent to which these environmental risk factors contribute to autism.”

Journal reference:

It brings new meaning to having an ear for music. Musical aptitude may be partly down to genes that determine the architecture of the inner ear.

We perceive sound after vibrations in the inner ear are detected by “hair cells” and transmitted to the brain as electrical signals. There, the inferior colliculus integrates the signals with other sensory information before passing it on to other parts of the brain for processing.

To identify gene variants associated with musical aptitude, Irma Järvelä at the University of Helsinki, Finland, and her colleagues analysed the genomes of 767 people assessed for their ability to detect small differences between the pitch and duration of a sound, and musical pattern. The team compared the combined test scores with the prevalence of common variations in the participants’ DNA.

Genetic variations most strongly associated with high scores were found near the GATA2 gene – involved in the development of the inner ear and the inferior colliculus. Another gene, PCDH15, plays a role in the hair cells’ ability to convert sound into brain signals.

Hearing the bigger picture

, an auditory neuroscientist at the University of Oxford, cautions that these findings should not be taken as evidence that genes determine musical ability. He points to the case of the profoundly deaf girl featured in the film . She became a superb pianist despite only hearing the world through cochlea implants, after meningitis damaged her inner ear. “Her case clearly demonstrates that even severe biological disadvantages can often be overcome,” he says. “She would do extremely poorly at the pitch discrimination task used in this study.”

Schnupp says that although the findings alone will not help identify who is destined to be a great musician, they may be valuable in understanding how the auditory system turns sound waves into perception.

Järvelä agrees that with something as complex as musical ability, genetics is not the whole story. “The environment and culture where you live are also important in musical ability and development,” says Järvelä. Her team now plans to study the interplay between genetic and environmental factors in the development of musical ability.

Journal reference: Molecular Psychiatry, DOI:

Coffee has long been a friend of students working through the night, but it does more than just keep us awake. A study provides the first convincing evidence that caffeine enhances long-term memory in people – provided the dose is right.

The effects mirror similar results seen in honeybees, where a boost to memory from caffeine-laden nectar may help bees return to certain plants.

Researchers strongly suspected that caffeine enhances memory, but studies that tried to show this in people weren’t conclusive, as any apparent benefits in memory could have been due to increased attention, a known benefit of caffeine.

Studies in animals such as rats, meanwhile, suggested that it enhances memory consolidation – the process of strengthening memories between acquiring them and retrieving them – which should affect long-term memory.

Seahorse similarity

To investigate further, , a neuroscientist at the University of California, Irvine, recruited 160 adults who normally consume only minimal amounts of caffeine. The volunteers first studied images of objects, before randomly receiving a pill containing either 200 milligrams of caffeine – equivalent to two espressos – or a placebo. Receiving the caffeine after studying the images helped to isolate the effect of caffeine on memory, as you wouldn’t expect alertness to matter at this point.

When the volunteers returned 24 hours later, they took a memory test involving images they had seen before, unseen images and images that were similar but not identical, such as a seahorse with a different shape to one they had seen previously. The volunteers had to classify these as “old”, “new” or “similar”.

Yassa’s team recorded no differences in accuracy for identifying old or new images between the volunteers who had taken caffeine and those who hadn’t – but this was expected because this part of the task was so easy.

Too much caffeine

However, the group who had received caffeine were significantly better than the placebo group at identifying which images were “similar” rather than “old”, which Yassa says is a harder task.

He concludes that caffeine enhances long-term memory by improving the process of memory consolidation. “This doesn’t mean people should only drink coffee after they’ve studied, and not before,” says Yassa. “I think you would get the boost regardless.” That’s because the process of consolidation is likely to begin as soon as new memories form.

However, caffeine isn’t much use once consolidation is finished. The team ran a second experiment in which caffeine wasn’t administered until one hour before the memory test, to check for any effects on memory retrieval. They found no such effect. “So let’s say you studied without coffee and decided to drink a cup right before an exam – that’s not going to help you retrieve memories better,” says Yassa.

Caffeinated bees

The results have impressed of Newcastle University, UK, who last year showed the link between caffeine and long-term memory in honeybees.

“We were often asked: what are the implications for humans?” says Wright. “So it’s nice someone actually did the experiment and found evidence that caffeine has a direct effect on memory consolidation.”

Finally, Yassa’s study also revealed that the dose of caffeine is important. When they repeated the experiment with 100 milligram and 300 milligram doses they found that neither was significantly different from placebo. This could be because other effects kick in at higher doses that negate the benefits for memory consolidation. Also, participants given the 300 milligram dose reported side-effects like jitters and headaches, says Yassa.

Wright’s team found a similar effect in bees. “In high concentrations it looks like [caffeine] is bad for learning – so don’t drink too much!” says Wright’s colleague Julie Mustard at Arizona State University in Phoenix.

Journal reference: Nature Neuroscience, DOI: 10.1038/nn.3623