At the BSA Conference 2012 and the London Open Day, a team of researchers from the Speech and Brain Research Group at Oxford University presented their work on brain imaging of people who stammer. For anyone who missed them, here they explain their findings.
Brain scientists make use of many different research tools to try to understand how the human brain achieves its unique ability to speak. We have applied these tools to the study of stammering. We hope to find differences in brain structure and function that might explain why some people sometimes struggle to produce fluent speech.
How is the research done?
Using tasks that measure changes in behaviour such as reaction times or errors, we can record what the brain can do and what people who stammer do differently from people who do not stammer. We can also measure precisely when the brain does something by recording its electrical activity using electrodes on the scalp or by measuring tiny magnetic fields produced by the brain activity (using a neuromagnetometer). These methods can track brain activity with millisecond precision and are called electroencephalography (EEG) and magnetoencephalography (MEG). We also want to know exactly where in the brain changes are occurring. To see which brain areas are activated when performing tasks, we use magnetic resonance imaging (MRI) scans. MRI machines also give us exquisite pictures of the brain's structure – the grey matter made up of cell bodies and the white matter made up of the fibres that carry communication to and from the cell bodies. All of these methods allow us to record how the brain is organised and when and where activity occurs during different tasks such as speaking.
Sometimes, we also want to test whether activity in a certain brain area is crucial for performance of a task and for that we need to use interference techniques. We interfere with the brain's function by stimulating it either directly (and painlessly!) using very weak electrical currents or using a large magnet to induce a current. Transcranial magnetic stimulation (TMS) induces electric currents in the neurons nearest the surface of the brain by generating a rapidly changing magnetic field around a coil placed on the scalp. This stimulation can interfere with on-going processes in the brain area to which it is applied. The effect is temporary and the person stimulated is often unaware of any change in their behaviour. Transcranial electrical stimulation (TES) is a relatively new tool in the brain scientist's kit that can alter brain function by direct application of a very weak electrical current across the scalp. When TES is applied during learning or training it can speed up these processes and prolong their effects. We hope to find out whether TES can help to improve speech fluency in people who stammer by prolonging the effects of known fluency enhancers, for example delayed auditory feedback or speaking in unison with another person.
What have we found?
To date, most brain research on stammering has used MRI scanning to see where differences in function and structure occur. Increases in brain activity are measured by increases in the amount of oxygenated blood supplied to a brain area during a task relative to another condition, such as resting. In our own studies, we scanned the brains of people lying in the scanner reading sentences on a computer screen. The research subjects could hear themselves speaking (via headphones) but sometimes we manipulated this feedback so that it was delayed (like an echo) or had a higher pitch (like speaking after inhaling helium). These manipulations of feedback can help some people who stammer to speak fluently. Speaking and listening activates the left side of the brain more than the right but roughly the same areas on both sides show an increase in activity when compared with resting during silence. These brain areas are involved in planning speech, moving the vocal folds, tongue and lips during speech and listening to speech. In people who stammer, we saw less activity in each of these brain areas but it was especially obvious in the area involved in planning speech. This area has lots of names, depending on whether its location or its function is being described. The area is located in the ventral (lower) part of the premotor cortex, which is part of a region called Broca's area on the left side of the brain (see green area, fig. 1). If Broca's area is damaged by a stroke, a patient experiences Broca's aphasia, in which speech is severely impaired, hesitant and dysfluent.
Abnormal brain wiring
The area we identified as showing abnormally low activity during speech production in people who stammer receives inputs from other brain areas involved in monitoring speech production (red arrow, fig. 1) and sends outputs to the areas that are involved in the execution of speech movements (orange arrow). We used another kind of MRI scan to examine these connections with other brain areas. This kind of scan is called diffusion-weighted imaging, which measures the tiny movements of water molecules (diffusion) inside brain tissue. Water molecules move mostly along a fibre rather than across it – so if we see a strong organisation of water diffusion in one main direction, then we assume that is because there is an underlying structure, such as a fibre tract, that is oriented in that direction.
Using diffusion imaging to study stammering, we showed an abnormality in the organisation of white matter fibres underlying the ventral premotor cortex (part of the green area). A difference in the same areas has been reported several times in studies of adults, adolescents and children who stammer, using different scanners and different analysis tools – so it appears to be a particularly robust finding. There are differences in other white matter pathways too but this area is of interest to us because it lies directly underneath the area of the brain we found to be functionally underactive during speech production in people who stammer. It is likely, therefore, that the white matter difference we observed affects the communication of this brain region with other parts of the speech network.
Is stammering due to poor co-ordinated activity between brain areas?
To test the idea that brain areas involved in speech production might have a problem communicating with each other in people who stammer, we looked at how well these areas co-ordinated their activity when speaking. In normally fluent speakers, we found that the areas involved in execution of speech movements (pink area) and listening to speech (blue area) synchronised their activity during speech production. In people who stammered, however, the opposite occurred: the areas became less synchronised than they were at rest. When the people who stammered spoke and received altered feedback of their speech - either delayed or at a higher pitch – the brain areas synchronised their activity in the same way as seen in fluent speakers (see fig. 2). This exciting finding gives us a clue as to how altered auditory feedback can help enhance fluency - by improving the co-ordination of activity between brain areas. This idea that speech fluency is influenced by the co-ordination of activity between brain areas is something we want to follow up in our next studies. We will do this using MEG, which will give us very precise measurements of when brain activity changes in different areas. One previous study found that in people who stammer the brain areas involved in executing speech movements (pink area, fig. 1) were active in advance of those areas involved in planning speech (green area), whereas normally speech is planned before the movements are executed.
What next? Stimulating fluency
In our new studies (starting soon) we are hoping to use brain stimulation to prolong the effects of successful fluency-enhancers, such as altered feedback or choral speech or speaking in time with a metronome. We have carried out our first study using TES, which showed that brain stimulation improved the learning of new words that we trained (fluent) people to produce. This improvement was seen one hour after the stimulation and training and persisted until 24 hours later. In our next study, we will test whether TES could improve fluency in people who stammer. We also hope to use MEG and MRI to see whether improvements in fluency due to TES are reflected in changes in the brain?s function and structure. This will help us to understand whether the differences observed in the brains of people who stammer are a consequence of stammering or might cause people to stammer.
Want to get involved?
If you would like further information about any of our studies, or if you would like to volunteer to take part in them yourself, please contact us via email at firstname.lastname@example.org. If you were previously a participant in our research, we would like to thank you for your valuable contribution, which has already helped us improve our understanding of the brain science underlying stammering.
Kate Watkins, Jennifer Chesters, Emily Connally and Riikka Mottonen,
The Speech and Brain Research Group, Department of Experimental Psychology, University of Oxford, www.psy.ox.ac.uk/research/speech-brain-research-group
From the Autumn 2012 issue of 'Speaking Out', pages 12-13