“Today is a very exciting time to be working in behavioural science” - an interview with Dr Sophie von Stumm

Primary page content

The Department of Psychology’s Dr Sophie von Stumm hit the headlines this year with her innovative work on the imagination, the time of day your brain works best, and the connection between breastfeeding and IQ.

As the recipient of a Wellcome Trust Seed Award, she spoke to Wellcome about her current and recent projects.

What are you currently working on?

With my Trust Seed Award, I am trying to find out how useful digital language recorders are for studying children's early life development. The recorders, which were developed by the LENA Research Foundation, a non-profit organisation, are the size of a floppy disk and weigh about two ounces.

Children wear the recorders in tiny pockets at the front of their t-shirts, as they go about their day. The recorders document everything the child hears and says for up to 16 hours. The data from the recordings are then analysed by automated algorithms from the LENA software package; for example, they give a count of the number of words that the child said and heard.

What I want to find out is how well parents and children respond to the recorders, for example if they find them intrusive or if they forget about them quickly and get on with things. I also want to test how useful the recorded data is to study children's language development and their home environments.

My hope is that our participants love the recorders, and that we will collect some high-quality data that allows us to study how and why children differ in their early life development.

How are you going about answering this question?

At first I hired Katrina D'Apice, my fantastic research assistant, to help me with the study. Together we ordered the recorders from LENA in the United States and also the t-shirts for the children to wear the recorders with. This sounds like a simple task but it turned out to be our first challenge!

We learned quickly that children are just as picky as grown-ups when it comes to t-shirts (perhaps even more so), and we had to order a second set to get the colours and patterns right.

The next step was to find participants. We want to study 2 to 4 year-old children, because they are at a very exciting stage of language development. With this in mind, Katrina and I went to visit nurseries in London, advertised the study online, for example on Mumsnet, and we asked around our friends and colleagues, if anyone had a child of the right age and wanted to participate.

The Wellcome Trust Seed Award is an amazing help with the recruitment process, because we can offer our participants monetary compensation for their efforts. As we speak, the first families are trying the recorders at home for our study, and we hope to have all our data by the end of next year.

How did you get into this line of research?

I want to understand the causes and consequences of individual differences in cognitive development, focusing in particular on early life.

To study children's differences in cognitive development, I use longitudinal studies that repeatedly assess groups of children across time, often for many years. One caveat of these studies is that they are limited with regard to the number of variables that they measured, because traditional assessment methods are expensive in time and effort.

I was looking for ways to overcome this problem, and I wanted to comprehensively assess children's development and their surroundings. I stumbled across LENA, which builds on the latest technology, and I thought it might be a solution - we'll soon learn if I was right.

What aspects of this form of funding are most useful to you?

It is fantastic to have the chance to test something new and risky that may not work out. Such opportunities are very rare, especially for early career researchers, and the Wellcome Trust cannot be applauded enough for having created exactly such an opportunity with the Seed Award.

Second, the funding enabled me to expand my lab, including buying equipment and hiring staff. As a result, I am now regularly invited by other researchers to collaborate and also, I am more often contacted by prospective PhD students who are looking for a supervisor.

But most importantly, the funding enables the identifying of ways to produce better science in the future. Improving our assessment methods to collect 'big data' will lead to the knowledge and understanding about human behaviour that is needed to improve people's lives.

What are the big trends at the moment for early career researchers in your area?

In my opinion, psychology today is dominated by two movements: replication and big data. For early career researchers, these movements bring two challenging questions. First, how much can we trust previous results in our field of study? And second, where can we acquire the expertise and resources that are necessary to collect and handle big data?

Because thoughtful answers to both questions are beyond the scope of this interview, I will only say this: One factor that increases the replicability of psychological findings is testing large samples on accurate and detailed measures.

Modern technology allows us to do exactly that - by collecting big data.

While I don't want to claim that more data will necessarily lead to better findings, I truly believe that today is a very exciting time to be working in behavioural science, because we are witnesses to a revolution in assessment methods that will shape the future of this and all other scientific disciplines.

What effect do you think recent tech developments will have on behavioural sciences? What are the coolest things you’ve seen being done recently?

Recent technological innovations in data collection methods will revolutionise behavioural science. In terms of cool things, I am obviously fond of the idea of using digital recorders to study children's early life development but most recently, I've been working on a smartphone application.

At my lab, the Hungry Mind Lab, we launched the iPhone application moo-Q, developed by PSYT, in late August. It assesses people's cognitive function and their mood repeatedly across time. I designed moo-Q to study if changes in mood were associated with changes in cognitive function, without having people come to the lab.

It’s very difficult to get people to come repeatedly to a lab to complete tests, and lab-based measures of mood are always biased and slightly inaccurate. To date, moo-Q has been downloaded more than 23,000 times with about 4,000 active users from across the world. We will be looking at the data later this month, so stay tuned for our findings.

In an ideal world, how would you like to see the project develop?

In an ideal world, these things will happen: we will test and collect data from 100 families with children aged 2 to 4 years within the next year. All children will wear the recorders for 3 days, and the parents will complete additional questionnaires. We will then extract the data from the recordings, using the automated algorithms.

Next, we will find that there is a good correspondence between data from the recordings and from the questionnaires. With that, we will move on to testing relationships between the children's development and their home environment, where we make some amazing ground-breaking discoveries.

We will publish our findings in high-impact journals, and we will tell other researchers in the field about LENA conferences and seminar talks - everybody will be just as excited as us. Over the course of the study, Katrina will complete her PhD, and I will write funding applications for future studies that use LENA (ideally, successful ones). And then we'll find out why children are so different.

This interview was originally published in the Wellcome Trust's newsletter, Trustnet, and is republished here with their permission.

Find out more about the Hungry Mind Lab and the Department of Psychology at Goldsmiths.

Read about moo-Q - the app that measures your mood and brainpower across the day.