ESRC INTERVIEW Flashcards Preview

For the main research project undertaken in the second year we will use a within subjects experimental design with 35 participants following that of Green et al (2013). I undertook a power calculation using their effect size and we need 31 participants to achieve 80% power  at the 0.05% significance level to detect an effect size of 0.46.

Participants will act as their own controls with measures taken before and after an intensive period of rehabilitation (60 hours, spread over the course of a two week music-making camp) and at three month follow up.

The dependent variables will include

a) scores on the Assisting Hand Assessment (AHA: a standardized test of spontaneous use and performance of an affected hand during bimanual interactions in functional or play-based tasks (Holmefur et al., 2009);

b) scores on The Children’s Hand Experience questionnaire (CHEQ: a 29 item questionnaire of independence in daily living activities using the affected hand (Sköld et al., 2011) 

c) scores on the Strengths and Difficulties Questionnaire (SDQ: a 25 item parental report questionnaire which provides a brief measure of social function and psychopathology (Goodman, 1997)).

There are a three key areas in the literature that suggest music and movement are interlinked in a powerful way.  

First, the close correspondence of the auditory and motor systems in the brain has been assessed in a number of studies with professional musicians such as violin players and pianists who have highly developed motor regions which activate without any physical movement due to musical training. There are  stroke rehabilitation studies such as (Schneider et al., 2007) using music supported therapy showing extended motor learning due to the coupling of auditory feedback and movement. 

Secondly the cross-modal mapping from the experimental literature shows that there are certain natural correspondences between sound and movement that can be used to improve body kinematics (Sedda et al., 2011). 

Finally Rhythmic auditory stimulation - entrainment to  auditory cues has been shown as the strongest evidence for rehabilitation in a recent cochrane review. 

The evidence for the link between music and movement is compelling and it creates a strong rationale for using music in the design and evaluation physical rehabilitation. 

The work with Breathe will involve the use of cross-modal mappings based on the literature. However, in tandem with the proposed project I intend to run a series of lab based experiments further investigating the mechanism between auditory and motor systems. Specifically looking at the role of music in priming motor learning. I was involved with a project last year looking at motor sequence learning in healthy adults and we would like to extend this by setting up a motor learning sequence controlling for musical exposure in a population of adults at Homerton Hospital undertaking rehabilitation for broken arms. We hypothesis that an experimental group who undertake exercise mapped to auditory feedback will have further motor learning by exposure to the auditory stimuli without movement for blocks of time.

The first phase of the project will be to categorise the children who will be aged 3 to 10 into a number of groups.  Breathe’s clinical lead will work with us to characterise the key motor deficits of each child to clearly see the movements most suited for each individual. Drawing on the cross-modal mapping literature from experimental psychology we will pair the movements with sound patterns taking into account user preferences.

There are a two key groups that will require a different approach to map movement with sound. Firstly,  one group of children will only have very gross movement with an active  forward reach exercise undertaken to improve overall arm function. In this case a simple entrainment to the beat of motivating songs will be used in a similar way to that used in my masters project. Participants will undertake entrainment exercises by hitting digital drum pads that can be moved to suit the precise trajectory required.

The second group will be able to reach but the quality of arm movement will be jerky and stiff requiring a different approach. We will work out which aspects of the movements need to be trained and by using motion capture technology and sonification algorithms we will give real time feedback to the child. The aim in this case is to create cross-modal mappings that track the trajectory of the arm to encourage smooth movement.  The child will see a video of the exact movement pattern we want them to achieve and they will aim to move in the exact same way receiving real time auditory feedback. 

An example using the motion tracking would be a child undertaking the action of moving to pick up a cup in the middle of a table. They need to keep their elbow in and move forward smoothly to grasp the cup. If they move in the desired way with a smooth trajectory they will receive a pleasant musical sound matched to the acceleration profile of their movement. However, if they stray too far from the desired trajectory the auditory feedback will create an undesirable sound until they regain the desired trajectory again. 

For my MSc we recruited three adult three stroke survivors in a multiple case study design. They undertook a 5 week intervention programme where they entrained to the beat of their favorite songs by hitting bespoke digital drum pads. The interfaces were connected wirelessly to an iPad app and designed so that all participants could use the system without any support from anyone else. There were 15 music making sessions led by myself but all participants were encouraged to work on the exercises on their own as much as they could. All participants managed to undertake many hours of self-managment and showed improvements on a large range of physical and psychological measures. The majority of gains were still seen 3 months post study in 2 participants who have carried on using the devices in the home. In fact I just collected more data a few days ago for a 6 month follow-up which shows that they have continued to undertake physical exercise and retain some benefits.

The most novel aspect of the study was that as far as we are aware it was the first study to look at self management in the home using digital musical interfaces for stroke rehabilitation.

One of the key challenges will be keeping the attention of the children doe many hours of practice. However, we have a good strategy to deal with this as our protocol will be designed alongside the clinical lead and the breathe team to be individualised. Each child will have a set of exercises and a personalised experience based on the most important exercises relevant for their recovery as well as their musical preferences. The difficulty of exercises can be calibrated in real time to make sure their is enough challenge to keep them interested but too much to put them off.

I think by building a large database of cross-modal mappings for use with the motion tracking we will have a useful set of tools that could be used in studies within the healthy population. However, as our key objectives are for physical and psychological outcomes in hemiplegic children we will not be able to look in great detail at more general motor learning principles. To do this we would require a different study design with a control group.

There are 2 key phases to the project. The first phase of the project will be to categorise the children into groups. Breathe’s clinical lead will work with us to characterise the key motor deficits of each child to clearly see the movements most suited for each individual. Drawing on the cross-modal mapping literature from experimental psychology we will pair the movements with sound patterns taking into account user preferences. This first phase will take up the majority of the first year and end with some small feasibility trials to make sure all the technology and exercise programme are working exactly as the health experts see as the most efficacious. Then moving into the second year Phase 2 will begin and we will run a 2 week intervention with 35 children (due to a power calculation I undertook with the effect sizes from Green et al 2012). There will be pre and post tests and a three-month follow up using physical and psychological tests.

Yes I worked out the power required by taking the effect sizes from Green et al (2012). This means that 35 children will be required for the within subject study design. 

I calculated a power analysis for the study based on the pooled effect size of 0.46 from rehabilitation magic camps for hemiplegic children (Green et al., 2013) who used the same outcome measures that we will be using. UK and Israeli camps had a pooled effect size of 0.46. We need 31 participants to achieve 80% power  at the 0.05% significance level to detect an effect size of 0.46.

There will be a lot to organise as I will be embedded within the Breathe office as well as at Goldsmiths. By having regular meetings to discuss progress and using a strict time line with the Breathe team we aim to achieve our project milestones. If there are technical or unforeseen hold-ups there is enough flexibility in the PhD timeline to still achieve our objectives.

There are a few key global objectives as well as personal. To address the former we hope to achieve a robust musical protocol for Breathe to help children with hemiplegia and their families improve their quality of life, while at the same time extending the psychological literature on cross-modal mappings in music and illustrating how music can be a powerful motivator in the rehabilitation setting. Personally there will be value to my academic career by having expert guidance from my supervisors in Goldsmiths and expert health care support embedded within Breathe.

Firstly all participants will only be permitted to take part if they are undertaking no other physiotherapy programmes or botox injections or electrical stimulation. This will ensure the exercises are the only intervention itself is what will affect the dependant variables. By working with the clinical lead we will make sure that all the exercises are relevant and safe and by using three standardised robust outcome measures that have been used in a prior study means we think all variables will be controlled for adequately.

By working with the Breathe team and the Goldsmiths Ethics committee we believe the ethical issues should be covered as the families and children will be able to give informed consent and the technology used will be extremely safe with expert staff from Breathe will be working with the children directly.

We will run paired t-tests to assess changes in group differences on each set of outcome measures as modelling the relationship between the intervention and outcomes using simple linear regression on each DV. 

I have also been discussing how to further analyse the relationships between psychological and physical outcomes and my supervisor Daniel has said that the Rubin causal model can work well for smaller sample sizes to undertake the analysis. 

In the longer term we could make use of a large data base of Breathes with the aim to analyse at least 100 participants data using structural equation modelling to inform the relationships between the psychological and physical variables. 

Cross-modal mapping using specific trajectories and musical features will improve quality of movement in hemiplegic children as assessed by clinical health experts.

We should achieve a number of mappings that will inform a musical protocol for children with hemiplegia and help us to take user preferences and cross-modal mappings to inform the literature. 

An example using the motion tracking would be a child undertaking the action of moving to pick up a cup in the middle of a table. They need to keep their elbow in and move forward smoothly to grasp the cup. If they move in the desired way with a smooth trajectory they will receive a pleasant musical sound matched to the acceleration profile of their movement. However, if they stray too far from the desired trajectory the auditory feedback will create an undesirable sound until they regain the desired trajectory again. 

If there we can't recruit enough participants for the power of the study we will use a multiple case study design where we collect multiple data points during a baseline phase, intervention and follow-up. In this way we can undertake single case analaysis of the data and still acheive some good insights into the intervention. Also if we are not able to collect data at 3 month follow-up for any reason or data is missing we can take an average of each child’s score from pre- to post-camp testing to avoid assumptions of a future directional trend.

I think we will gain a good understanding of which cross-modal mappings are the most motivating and relevant for rehabilitaiton in hemiplegic children that will help inform a robust musical protocol for Breathe to use moving forward. 

There will be a camera that is positioned on a tripod to one side of the participants. There will be a number of light non-invasive sensors that are placed on each of the joints of the affected arm. There are small clusters of sensors that can be attached to the wrist very easily that give good kinematic acuracy in tracking the orientation and smoothness of movement. A number of musical algorithms will be used to create a cross-modal mapping between the movement and auditory feedback giving the participants ireal time feedback. 

I see myself as continuing as a post-doc with more research at the intersection of psychology, health and state of the art creative technology. Music will always be central to my aims with an ambition to continue understanding what features of music are the most relevant for rehabilitation. I also enjoy teaching and would like to supplement my reaserch career with a number of lecturing positions preferably with music psychology at the fore. 

I am very interested in the interplay between music and the brain. Partiularly the link between motor and auditory regions in the brain and the power of rhythmic entrainment on this.

A key question would be. Can music prime motor learning? And how much exposure to the audutory stimuli is required before the coupling is strong enough to extend motor learning with the negation of movement? 

I aim to further my career between three key areas, academia, the health exctor and the third sector continuing on my resaerch into rehabilitaion with music as a faramefework for the resaerch. If I can help design, evaluate and embed musical interventions for a number of populations this would be my ultimate aim. Helping those less fortunate to undertake exercise and gaining benefit both pysically and psychologically. 

1. What are some of the most exciting areas of reaserch in the Psychology depeartment currently? 
2. Do you have many collaborations with other instutions and partners in the third sector?