ICMPC 12 Day 1: First afternoon session
I missed the start of the afternoon session as I was chatting away to long lost colleagues and friends. They are not actually ever lost of course, but they do spend most of the time a very long way from London so I enjoy catching up with how they are getting on in life and in work. The sad consequence of this is that I kept missing things on the first day!
In the afternoon paper session (two in each session) I managed to catch Erin Hannon’s talk about pattern learning in music and language. Erin started off with a discussion of the evidence both for dissociation and overlap between music and language processing:
Dissociation evidence = Patients with double dissociation, aphasia (language problems) but not amusia (music problems) and vice versa.
Overlap evidence = There is lots, but Erin talked about children who have better auditory verbal skills when they learn an instrument. Also, there is evidence that people who speak a rhythm based stress language (Japanese or Finnish) do better with temporal processing tasks while those who speak a tone based language (Mandarin Chinese) do better with pitch based processing tasks. Finally, there is the way we create patterns in both language and music concurrently, such as in infant directed speech.
Personally I have always been far more convinced with the overlap evidence but then it is important not to ignore important things like double dissociations.
In any case, complete overlap or dissociation would make little sense based on our knowledge of how the brain is organised so there is bound to be a degree of one or the other – I believe in more overlap than dissociation although this is mediated by your degree of musical experience/expertise.
Feel free to disagree – I may disagree with myself later in life!
Erin has conducted (and is conduction – at least two are ongoing) experiments based on long established paradigms that show how we learn sequential patterns. These are often called the ‘Saffran paradigms’, after Jenny Saffran, who showed that infants and adults can learn the rules of structure within an auditory stream following simple exposure. This is thanks to our ability to track statistical regularities in the environment and to learn these patterns over time.
Erin presented 5 experiments in total – too much detail for me to get down in my notes – but essentially her main finding centres on a simple manipulation where you create stimuli that can be heard as music of language (sung syllable triplets). You then tell people that they will hear music and language and see if there is a difference in their response during a rule based learning task, where they hear novel patterns and say whether they match the previously heard stimuli.
Erin’s group have tried this experiment across a variety of adult populations (infant studies are in the pipeline) and have found that most people (including tone language speakers – note, this is different to the information in the conference abstract) are biased towards the syllable pattern when making their responses. Musicians are different – they pay more attention to the melody patterns even when you tell them it is speech.
These studies add to our knowledge about the role of domain specific strategies/processes in our perception of the auditory world. I am very much looking forward to the developmental infant studies to come, to see whether the biases Erin has found in adults are present in some infants before a lifetime of musical experience begins…