Day 2 – Impact of music experience on language processing
After the first symposium of Day 2 I retired to the very grand coffee/tea room for a warm beverage and a catch up with a few friends. It is really lovely to be able to see colleagues at these conferences, and to find out how everyone is doing with their work. It sounds like even though funding times are tough for academia music psychology is still managing well, with many people completing exciting moves to new labs, publishing work or getting grants. Overall, the mood was buoyant and positive.
Boosted by a mild dose of caffeine we all headed back into the main Hall for symposium 2 which had been organised by Mathias Oechslin from the University of Geneva and was called ‘Impact of musical experience on cerebral language processing’. The aim of the symposium was to examine the latest research on the question of whether musical expertise facilitated or affects language processing.
1) Nina Kraus – Selective auditory attention and hearing speech in noise
I have written previous blogs recently about Nina’s fine work in the area of speech processing in musicians and her labs specialism which is looking at the frequency following response in the brainstem. As usual she did not disappoint with her dynamic presentation method. I remember a time when Nina was the only person who wore a head mike at these conferences, allowing her to move about the stage – this time lots of people copied her act!
In this presentation she focused on her work looking at enhanced auditory working memory in musicians. If you are interested you can see the details of Nina’s papers for free on her fantastic lab pages (also lots of great presentations)
Her conclusion was that hearing speech in noise is not only a bottom up skill, but one that is influenced by top down cognitive processes as well – in particular the capacity of auditory working memory. She showed evidence of improvements to hearing speech in noise across a wide range of individuals with musical experience including school age kids, adult musicians, and older adults with musical training. This so called ‘biological advantage’ for language processing afforded by musical training was also linked to increased MMN response and more activity in memory areas of the brain during speech in noise, compared to basic sensory areas.
Overall, Nina is amassing a large body of evidence to show the advantage conveyed by those with musical training. What remains is to determine whether this is a biological predisposition or a consequence of training in any way (and, if so, which aspects of musical training?)
2) Daniele Schön – Speech segmentation
Daniele has developed a rather cool skill as a result of his research using the Saffran paradigm. This paradigm tests implicit learning and requires the presentation of around 20 minutes of continuous syllables or tones, in which are buried some hidden patterns. You then determine whether the participant has learned any of the hidden patterns. Daniele has been using a combined music/language version where the stream of nonsense syllables are also pitched, so they sound sung. And he can sing them!! He can actually reproduce the stream himself with remarkable fidelity. For a music psychologist like me, that counts as a good party trick!
So does combining the pitch and syllable versions of the Saffran paradigm lead to better learning? Daniele has shown that the answer is yes, people learn hidden patterns better when the syllables are sung. Is there an advantage for musicians? Francois & Schön (2011) demonstrated that everybody struggles with ‘hard’ versions of this test, but that musicians show a performance advantage on the ‘easier’ versions. Such a finding is more convincing to me than just showing that musicians are better at everything – here there is an advantage but a definite processing limit. He also used EEG in his study and found an N400 ’familiarity’ effect in musicians but not non-musicians, backing the argument that they have learned more about the stream.
His follow-up study was to try a longitudinal study with 37 children – a brave and complex paradigm choice. He looked at children’s ability after random assignment to the usual choice of a painting class or a music class, for two years.
He found that:
1) Sung materials were again much better learned than the spoken ones (advocating the use of song in language learning)
2) Musical training facilitated learning of both the linguistic and the musical structures (again backed up by the early EEG measures which showed that the music group needed less time to segment the stream).
He concluded that musical practise has a beneficial effect on speech segmentation. The question now, is why?
The penultimate talk of the session was from Martin Meyer who talked about his work with EEG and responses to the fast changing acoustical signals within speech, including voice onset time. I am afraid my sprained wrist was hurting by this point so I was unable to take detailed notes. But I can tell you that musicians showed different N1 responses compared to nonmusicians, and that Martin argues that musical expertise facilitates processing of these sub-segmental cues in speech by modifying our auditory system.
The end of the session was to feature Ani Patel’s presentation of his new hypothesis which is has called OPERA. Sadly, we were told that Ani couldn’t make it to the conference (an announcement that triggered an audible groan of sadness from the audience) and also, unfortunately, the symposium had over-run so there was only 5 minutes for the brave session chair to briefly present his ideas (to a rather hungry crowd; at this point I was glad of my large morning bowl of Scottish porridge!).
Ani’s new hypothesis attempts to define why, and under what circumstances, music training might benefit speech processing networks. He argues that plasticity occurs when 5 conditions are met:
1) Overlap – Anatomical overlap in the brain regions that process aspects of the speech and music signal
2) Precision – Music places higher demands on these networks than speech
3) Emotion – Musical activities that engage the network elicit strong emotional responses
4) Repetition – Musical activities are repeated
5) Attention – Musical activities are associated with focused attention
In general, I like broad outline theories like OPERA. It gives a blank set of hypotheses which we can then work on to try to show where there is support and where there is conflict; and where there is conflict, to modify, improve and build on the foundations of the original hypothesis.
I am strongly in favour of moving away from simply showing that musicians have advantages and focusing more on why and how these advantages occur. I have to say that many of the principles outlined in OPERA seemed to me to resemble basic learning mechanisms (e.g. repetition and attention) so I am not sure that they can ever be expanded upon or falsified. But I admit I am sufficiently intrigued by the OPERA hypothesis to wait for the details from the man himself before drawing further conclusions. I look forward to hearing more.