Could melody analysis spot the bum notes in our brain patterns?

Music in the brain.

The culmination of Steven Spielberg's 1977 classic Close Encounters of the Third Kind gave us one of the great moments in sci-fi movie history. It starts with a simple, unforgettable five-note melody – the musical phrase used by US government scientists in an attempt to communicate with a huge UFO that has just descended on the rocky Wyoming countryside. There's a hushed pause as the last note drifts in the night air, followed by a deafening blast as the alien vessel gives its rumbling response. Before we know it, the two parties are bouncing arpeggios off one another; a beautiful, joyful symphonic conversation.

As well as providing a wondrously optimistic antidote to the end-of-days mentality of today's chrome-plated alien blockbusters, that final sequence perfectly articulates the idea of melody as a universal language, one that might be used to explore previously unfathomable mysteries.

In many ways, the human brain is almost as alien to us as the extraterrestrial visitors of Close Encounters. Our relationship with our brains is a little like the relationship most of us have with our PCs – we use them intuitively and we understand the surface processes, but we've only the most basic understanding of how everything works under the hood.

With researchers still struggling to get to grips with the brain's inner workings, the treatment or management of an incredibly varied disorder like epilepsy remains an uphill struggle for medical institutions. One of the keys to better management of epilepsy is the ability to forecast impending seizures, the disorder's main cause of death. Clinical studies in this area predominantly focus on the use of electroencephalogram (EEG) data, which records the brain's electrical activity, but the difficulty involved in accurately interpreting this data is a recurring problem.    

A just-launched European research project, spearheaded by the Italian Association for the Research on Brain and Spinal Cord Diseases (ARCEM), aims to prove that the universal language of melody could provide a solution to this problem. The project has pulled together neuroscientists, IT specialists, musicians and music analysts in an attempt to predict impending seizures using a method called data sonification.

Data sonification is the process of expressing visual data, like EEG read-outs, in the form of melodies. The project is investigating whether tying EEG data sets to musical melodies could help researchers, and eventually doctors, to spot the abnormal brain activity that prefaces a possible seizure.

What advantages does turning EEG data into melody streams bring over the current methods of studying visual brain pattern data to predict epileptic seizures? According to the project team, turning visual data into audio melodies could help researchers to sequence and conceptualise the brain's activity over time, increasing the possibility of catching the hidden signs that could signal a seizure. These signs would be expressed as an abnormal or jarring sound in the melody, essentially turning seizure prediction into a clinical search for a bum note.

Data sonification also brings a more human advantage – our intrinsic ability to spot the off-key note in a melody. The ear is naturally attuned to audio patterns and detecting irregularities within them. The project team believes that expressing EEG data musically could help doctors and researchers identify seizure-indicative anomalies more easily than looking at graphs and read-outs.

Unfortunately, it'll be a while before we find out if this intriguing approach will bear fruit in the field of epileptology. The project has only just begun to gather the huge volumes of EEG data needed to validate the data sonification technique; the team hopes to present some preliminary results by the end of 2013.  

Although it's too early to prove the effectiveness of data sonification for predicting seizures, there's something beautiful about the idea that music, the universal language, might hold the key to furthering our understanding, not of visitors from another planet, but of the inner recesses of our own minds. Spielberg would be proud.

Link to full feature:

Photograph: Getty Images


Chris Lo is a senior technology writer for the NRI Digital network.

Photo: Getty
Show Hide image

Who'll win the Richmond Park by-election?

There are three known unknowns that will decide the contest. 

It’s official: Zac Goldsmith has resigned as the Conservative MP for his Richmond Park seat, and has triggered a by-election there, where he will stand as an independent candidate.

Will it be a two-way or a three-way race?

The big question is whether the contest will be a three way fight between him, the Liberal Democrat candidate Sarah Olney, and an official Conservative candidate, or if CCHQ will decide to write the thing off and not field a candidate, making it a two-horse race between Goldsmith and Olney.

There are several Tory MPs who are of the opinion that, given that latitude to disagree on Heathrow has been granted to two Cabinet ministers, Boris Johnson and Justine Greening, similar leeway should be extended to Goldsmith. It’s win-win for Downing Street not to contest it, partly because doing so would put anti-Heathrow MPs, including Johnson and Greening, in an impossible position. Theresa May isn’t averse to putting Johnson in a tricky spot, but Greening was an early supporter of her leadership bid, so her interests come fairly high up the prime ministerial radar.

But the second reason not to contest it is that Goldsmith’s chances of re-election will be put in a serious jeopardy if there is a Tory candidate in the race. Everything from the local elections in May or the Liberal mini-revival since Brexit indicates that in a three-way race, they will start as heavy favourites, and if a three-way race results in a Liberal Democrat win there will be bloodletting.

Although people are talking up Goldsmith’s personal vote, I can find little hard evidence that he has one worth writing home about. His performance in the wards of Richmond Park in the mayoral election was actually a bit worse than the overall Tory performance in London.  (Boris Johnson didn’t have a London seat so we cannot compare like-for-like, but Sadiq Khan did four points better in Tooting than he did across London and significantly outperformed his general election performance there.) He did get a big swing from Liberal to Conservative at the general election, but big swings from the Liberal candidate to the Tory were a general feature of the night, and I’m not wholly convinced, given his performance in Richmond Park in 2016, that it can be laid at Goldsmith’s door.

If he wins, it’ll be because he was the Conservative candidate, rather than through any particular affection for him personally.

But will being the Conservative candidate be enough?

Although on paper, he inherits a healthy majority. So did Robert Courts, the new MP for Witney, and he saw it fall by 19 points, with the Liberal Democrats storming from fourth to second place. Although Goldsmith could, just about, survive a fall of that magnitude, there are reasons to believe it may be worse in Richmond Park than Witney.

The first is that we already know, not just from Witney but from local council by-elections, that the Liberal Democrats can hurt the Conservatives in affluent areas that backed a Remain vote. But in Witney, they barely squeezed the Labour vote, which went down by just over two points, or the Green vote, which went down by just under two points. If in Richmond Park, they can both damage the Tory vote thanks to Brexit and squeeze Labour and the Greens, they will win.

Goldsmith's dog-whistle campaign for the London mayoralty will particularly help squeeze the Labour vote, and thanks to Witney, the Liberal Democrats have a ready-made squeeze message. (In Witney, Green and Labour votes would have been more than enough to elect Liz Leffman, the Liberal candidate.)

But their good performance in Witney and Goldsmith's mayoral result may not be enough on their own.  Ultimately, the contest will come down to the big question that will decide not just the outcome in Richmond Park but the future of the Liberal Democrats.

Have the voters forgiven the Liberal Democrats for going into coalition?

We know that Brexit can help the Liberal Democrats at the direct expense of the Conservatives. What we don’t know is if Brexit is enough to convince 6,000 Labour voters in Bath to vote tactically to get Ben Howlett out in exchange for a Lib Dem, or for 7,500 Labour voters to back a Liberal candidate in Hazel Grove to defeat William Wragg.

One of the reasons why the Liberal Democrats lost votes directly to the Tories in 2015 was fear: of uncertainty and chaos under an Ed Miliband government propped up by the SNP. That factor is less live in a by-election but has been further weakened due to the fact that Brexit – at least as far as Remain-backing Conservatives are concerned – has brought just as much uncertainty and chaos as Miliband and the SNP ever would have.

But the other reason was disgust at the Liberal Democrats for going into coalition with the Conservatives. If they can’t win over enough votes from the parties of the left, we’ll know that the party still has a way to come before we can truly speak of a Liberal revival. 

Stephen Bush is special correspondent at the New Statesman. His daily briefing, Morning Call, provides a quick and essential guide to British politics.