Category Archives: Music and Neurosciences IV

Day 4 – Recent findings in autism

Good morning dear reader. I realised yesterday that I still had one blog from the recent Music and Neurosciences conference that I had left to upload! So here you are:

I was particularly looking forward to the symposium entitled ‘Music: A window into the world of autism’. My friend Prof Pam Heaton was due to speak and I have always enjoyed her presentations, plus I was interested to hear about the other types of work going on in this area. The goal of the symposium was to present recent advances in research examining the neural underpinnings of musical processing in Autism Spectrum Disorders (ASD) and the therapeutic potential of experimental treatments in ameliorating some of the associated deficits.

1)      Pam Heaton – Why are musical skills preserved in autism?

The first speaker was my friend Pam, who did a great job of outlining the symptoms and consequences of ASD for the audience. She pointed out that the first report of ASD from Kanner (1943) remains a very accurate description of the condition to this day, in particular in its emphasis on the language and social abnormalities that characterise a child or adult with ASD, and also the high incidence of heightened musical abilities. In Kanners original study 6/11 children displayed unusual music-related behaviours including one in particular who was able to identify 18 symphonies within the first two years of life.

The paradox of ASD in the auditory world seems to be a neglect of speech (Klin, 1991) combined with focused attention of the contour, syntax and even emotion of music.

Pam presented some pilot work she is conducting right now looking at ASD children and their reaction to different auditory sounds in their environment. She is looking at 10 children and age matched controls and coding a number of factors in their responses to different sounds: orientation co-ordination, positive affect, vocalisation, and anticipation/ increases in focus. She found that children with ASD showed more attention to music compared to either speech or environmental sounds.

In a second study she looked at emotion recognition of happy, sad and fearful states in vocal gestures (like laughing), vocal pseudo words and music. She has so far found no differences between ASD and control children in their ability to detect emotion in these stimuli. Interestingly however, while ability to detect emotion in all three types of stimuli is highly correlated in control children, the ASD children show no correlation between their identification of emotion in vocal gesture and music, and pseudo words. She concluded that speech processing may be less focused on semantic content in ASD and more focused on the finer perceptual aspects (so the more ‘music like’ aspects such as pitch and contour). An example of this is being able to recognise a word spoken by their mother but fail to recognise the same word spoken by their father.

2)      Krista Hyde – Brain and behaviour correlates of auditory processing in ASD

I know Krista of course from her many years of work in congenital amusia. She now has her own lab and has moved into the area of studying auditory processing in ASD, in both children and adults. She talked about her work on high functioning adults with ASD. She pointed out that ASD has up to 90% heritability and is 4 times more likely in males. She also built on Pam’s introduction by describing the auditory processing features of ASD:

  • Hyper/hypo sensitivity
  • Hyper focus on the local sound source
  • Diminished perception of speech
  • Enhanced musical ability, including pitch perception and pitch memory. Absolute pitch is also much more likely in ASD than in a control population

She presented the Enhanced Perceptual Functioning (EPF) model of ASD, which is based on a similar model that was developed in the visual world. It is the equivalent of seeing the tree before the wood, rather than seeing the wood before the trees.

She also presented the Neural Complexity Hypothesis (NCH) in ASD (Bertone et al. 2003) which postulates that an over-activity in primary visual and auditory areas in ASD in response to simple stimuli is combined with an under-activity in the same areas for complex stimuli.

Her three studies (one behavioural, one brain structural, and one brain functional) presented data that nicely fitted with both the above hypotheses. In the first study, ASD adults show better perception of simple tones, especially when their speech onset is delayed. In the second study she found larger left planum temporal volume in adults with ASD (Foster et al. 2011 – Human Brain Mapping). Finally, simple tones elicit larger activity in primary auditory areas in adults with ASD compared to controls (reflecting their hyper sensitivity) while complex tones elicit less activity in the same area (reflecting their hypo sensitivity).

Her conclusion was that there is particular sensitivity in ASD to the processing of low level features, and pitch in particular. Next she intends to collect DNA samples to make a further link between genes, the brain and behaviour.

The final two talks were  by Istvan Molnar-Szakacs and Catherine Wan (session chair). Both presented great brain data that supported the work of the previous two speakers.

Istvan has been working with Katie Overy on the Shared Affective Motion Experience (SAME) model of music perception, which holds that music is perceived not only as an auditory stimulus, but also as intentional, hierarchically organised sequences of expressive motor acts behind the signal.

They postulate that the human mirror neuron system allows for co-representation and sharing of affective musical experiences. With fMRI they have showed that children with ASD exhibit normal responses to emotional music in their mirror neurons and emotion areas of the brain, supporting their reliable ability to identify emotion in music (as showed by Pam earlier)

Catherine is working on a fascinating new intervention for nonverbal ASD children that is called Auditory Motor Mapping Training (AMMT). AMMT grew out of work on music therapy for stroke patients (Melodic Intonation Therapy) and in principle the method is very similar.

There is bimanual tapping on two drums that play a tone as you hit them (one high and one low) and children are taught words using the drums. Her pilot work, on 6 nonverbal ASD children has shown that in just 8 weeks children can learn some simple words, which is a big breakthrough when in some cases the children had not spoken for up to 8 years of their life. Using brain imaging of white matter tracts in the brain (Diffusion-Tensor Imaging) she has shown abnormalities in the language tracts of the brain that may be related to treatment outcomes. I will look forward to hearing more about her studies as they progress.

Day 4 – Recent findings on amusia

As you might expect, dear reader, I have a slight bias towards any conference symposium on congential amusia. I enjoyed all the symposia at M&N IV of course, but there is nothing quite like the engagement you get from hearing speakers you know inside out from years of reading their papers and admiring their work. So I sat near the front row (like a proper geek), pen at the ready and brimming with excitement to hear the latest developments.

WARNING: This blog will be longer than usual – I just can’t help myself!

For those whose lives are not so connected with the topic I should outline first of all that all the speakers in this blog discussed ‘congenital amusia’, a neurodevelopmental disorder that impacts negatively on pitch perception and production. As a gentleman pointed out to me during my poster presentation, we should state more clearly that this is a different condition to acquired amusia, whereby an individual develops music processing difficulties as a result of neural damage (stroke or injury) or degeneration.

1)      Simone Dalla Bella: Memory disorder and vocal performance

Simone began by discussing his studies of vocal performance in amusia. Whereas most people learn to sing in tune, this is in fact a complex behaviour that is underpinned by several brain components which all need to function optimally in order to bring about successful production. There are two ways in which singing can fall down:

  • Failure in perception – this is thought to be the primary cause of poor pitch singing in amusia
  • Failure in auditory motor mapping – this is a more common cause of poor pitch singing and may be behind all those people who are not amusic but couldn’t carry a tune in a bucket

But what about memory? Simone pointed out that there are now a handful of studies showing memory problems in individuals with amusia (at this point I went my usual scarlet colour upon seeing my name on the screen!) But no one had yet looked at how their memory problems influenced their poor pitch production or the poor production of ‘poor pitch singers’ who do not have amusia.

Simone showed two studies that convincingly demonstrate that singing with lyrics and singing with a model (i.e. imitation of a song) can positively influence performance in both amusics and poor pitch singers as compared to singing on a new syllable (such as ‘ta’) or singing from long-term memory. This opens the possibility for future singing aids that aim to compensate for the influence of poor memory representation of the ability to sing.

2)      Lauren Stewart (AKA ‘the boss’ to me!): Congenital amusia – is there potential for learning?

Lauren presented work she completed this year with one of our masters students, the lovely Sue Anderson. Sue is simply the best singing teacher I have ever seen, and for her masters project on the Music, Mind and Brain course (Goldsmiths) she wanted to see if she could improve the amusics ability to sing. This went in tandem with a challenge from Professor Graham Welch (Institute of Education) who had told us for years that he didn’t believe there was such a thing as amusia and that all these people needed was proper exposure to music to allow them to perform like anyone else.

So Sue embarked on a 7 week singing program with 5 of our amusic individuals. She tested their perception and production abilities before and after the intervention. She based her training on breathing exercises, building vocal strength, improving pitch response to feedback and a number of other tasks.

Before I even knew the results of the study I was impressed, as the feedback from the amusics who took the program was all really positive; they loved it! The results showed a rather mixed bag of changes, with most of the participants making small improvements in their abilities, but on average still remaining in the amusic range for perception. But there were certainly larger gains in confidence, vocal control and also willingness to try to listen to music in the future rather than simply ignore it. All in all, a fascinating study that shows the real potential of a learning experience, beyond that of simple skill gains.

3)      Psyche LouiBehavioural and neural correlated of normal and disordered singing  

Psyche is well known for her brain imaging work in the field of amusia, and in particular for her findings regarding the disrupted organisation of one particular brain pathway in amusia, the arcuate fasciculus (AF). Today she was presenting some new work looking at the source of musical knowledge and how well we learn when exposed to a completely new musical scale.

She was interested in testing implicit learning, the principle by which we all absorb the knowledge of our own musical system without explicit conscious knowledge. This is how we find it so easy to listen to the music of our own culture (which follows the rules we have internalised) over the music of another culture which can sounds strange and disorganised to our ears.

In order to test this system she did a really clever thing – she invented a new musical grammar! She used a variant of the little known Bohlen-Pierce musical scale which is a 3/1 scale with 13 logarithmic divisions and composed a new finite state grammar. By this method she created a new musical system that she calls the “Martian system” (with one of my favourite cartoon characters as its mascot!)

So how do amusics and controls compare on their ability to learn this new music? Using both behavioural and brain imaging (DTI) studies she showed that controls can learn the structure of the new music quite well after minimal exposure (30mins) whilst amusics did not.  Once again she showed reduced AF volume in the amusics and significant correlations between AF tract volume and learning ability. She concluded by saying that the acquisition of musical structure (i.e. the ability to learn about music) depends on brain structures like the AF, and this may be one reason why the amusics in her study found it difficult to learn implicit rules about a new musical structure.

4)      Isabelle PeretzLearning speech but not musical sounds in amusia

I can’t help but be a bit in awe of Isabelle. Not least because without her initial work on amusia I would not have a job! But the sheer output of her lab and the scope of her ideas cannot fail to impress (even when you are not directly financially grateful)

During this session she presented the results of studies conducted in collaboration with Jenny Saffran, who is famous for her work on language learning. In particular, she is known for the development of a paradigm that shows our ability to learn transitional probabilities in a continuous stream of sound (with around 21mins exposure). You can expose people to either syllable streams or tone streams and they seem to be able to spot the hidden patterns within them, even in the absence of conscious knowledge. So can amusics do this?

She presented 5 experiments (yes, 5!) where she tested amusics and matched controls, although you got the feeling she had done quite a few more while struggling a little with the paradigm. Her results were as follows:

  • Simple tone streams – both groups can learn syllable streams. Nobody learned the tone streams
  • Diatonic piano tones – as above, expect this time the controls learned the tone streams
  • Reduced number of targets – as above (although it appeared from the graph that half the amusics did learn the tones here)
  • Sung sequences (combined) – Much better learning overall, but amusics still worse
  • Sung sequence without amusics – Everybody learns well

Overall, the amusics seem to learn the language stimuli as well as the controls but they are much poorer with the tone stimuli. She argues that this result illustrates how lack of musical exposure cannot explain amusia, although it must be said that she is working with an already compromised system in adult amusics – it is not clear how children with amusia would perform in similar circumstances. But Isabelle is soon to publish the first case of amusia in children (a very good paper that I was lucky enough to review) which will hopefully bring forward new tests that will enlighten the role of development and exposure in congenital amusia.

In the meantime I’m off back to my lab to chat to the people who really matter, those with amusia, about these findings and see what they think. I’m really looking forward to seeing them again in the lab over the summer and to exploring their music abilities together. What a super bunch of people they are to work with – I’m a jolly lucky psychologist!