We’re looking at new ways of detecting concealed information – in particular, detecting when people are lying about recognising something or someone. When most people think of lie detection, they think of Polygraphs and measuring heart rates, but these are easy to “cheat” by staying calm and have accuracy rates as low as 35%.
Our research uses computational methods to analyse brain activity instead. When we see something familiar, such as a face or a name, there is a specific spike in brain activity called the P3. Using Rapid Serial Visual Presentation, we are able to show images or words at a rate of about 10 per second, which means they appear on the fringe of awareness – where your brain “sees” every image but you aren’t fully conscious of it. Your brain has templates for important or familiar items, and if one of the images we show you matches a template, it breaks through into your conscious awareness and we record the spike in brain activity. As we present the images so quickly, you can’t control your response since by the time you realise you’ve seen it, your brain has already reacted.
I find the whole concept fascinating and my supervisor’s initial results were remarkable: when showing people their own names, the test was 100% accurate. What I’m doing now is applying this method to a wider range of items, such as locations and email addresses. It’s interesting to take something that we know works in a specific situation and apply it more widely. If it works, it could make a real positive impact. I’m at the writing up stage now and it’s exciting to soon be able to show everyone what I’ve found.
I studied psychology initially and that’s where my interest in neuropsychology and brain activity developed. I decided to take Kent’s computer science conversion Master’s and discovered an academic in the School of Computing was doing work on brain activity and lie detection. I got in touch and luckily, he agreed to supervise my Master’s dissertation. I was then able to move on to the PhD.
Research was always something I liked, so for me I was taking my favourite part of all my previous degrees and just doing more of it.
I see my supervisor every week and can contact him by email any time. He’s in the office next door to mine so it’s very easy to get in touch. It’s been a very positive experience.
I have an office that I share with other PhD students in the Computational Intelligence Group, it’s a nice set-up. Although our research doesn’t overlap often, we all get on and help each other out if need be. When you start your PhD, you can choose whether to have a laptop, a Mac or a PC, which is great.
They are encouraging and genuinely interested in our work. We have weekly research seminars where we can present our work, you’re asked lots of questions and get useful feedback. It’s a good experience and prepares you for your viva, where you are going to get grilled!
Kent’s Graduate School also offers lots of workshops. They run a kick-start your PhD session and lots of other training to help you gain additional skills.
I got funding from the University which was topped up by the School.
I’m looking at two main options. I love research and am looking at post-docs and continuing my research in this area, but I’m also looking at jobs in the public sector, such as the civil service, or in a private company where I can use my research and analytical skills to have a positive impact on people’s lives.
In terms of your research, have an open mind and don’t worry if it takes a different direction to the one you were anticipating. That’s one of the great things about research, it can lead you somewhere you weren’t expecting to go.
Also, get involved in School life. I became an ambassador for the School. I’ve really enjoyed it and gained a lot of valuable experience. Take advantage of the workshops and everything else that is on offer too. For example, the University’s annual employability festival is great; I spoke to lots of people who were keen to talk to me about their companies and what they could offer me.