We are pleased to announce the newest member of the AXNS team – Romy Lorenz as Public Engagement Curator. Here she writes of her background, and her motivations for joining.
Hooking up a brain to a computer – I remember very well the first time that I learned about brain-computer interfaces (BCIs). I was completely enthralled by the broad application areas: ranging from communication devices for patients completely locked-into their bodies unable to move or communicate otherwise over additional control in gaming to mental state monitoring in cars or airplanes. The fact that this was not pure Science Fiction but an actual field of research gave my young life a calling. From then on (that was 2010 back at the Berlin Institute of Technology) I focused my further education on the acquisition and analysis of electroencephalogram (EEG) data. EEG is a non-invasive technique that records the brain’s electrical activity and is most commonly used as an input channel for BCIs in humans.
However, three years later and with research experience gained in labs in Beijing, San Diego as well as Berlin, I became somewhat disenchanted with the field. I started asking myself: Why are there still no patients benefiting from BCIs installed at their homes, allowing them to communicate with their families? How is it possible that we are still extracting the exact same EEG signals from the brain as 30 years ago? Why hasn’t there been any breakthrough in recent years although in the eyes of the public BCIs seem to be the answer to nearly anything? While I realized that this was rather a limitation of the very technique employed (EEG) and not a lack of genius, passion or motivation of the scientists involved, I also became aware of the responsibility we have as researchers towards the public. Having the media disseminate unrealistic short- to mid-term futuristic scenarios does not contribute to a constructive controversy about the limitations and challenges faced with trying to bridge the gap from the laboratory to real-world applications of so called neurotechnologies. Those facts motivated me to slightly diverge from my research path and move to London.
Currently, I am a third year PhD Student at Imperial College London and I work in the Computational, Cognitive and Clinical Neuroimaging Laboratory (C3NL) under supervision of Robert Leech (Division of Brain Sciences) and Aldo Faisal (Department of Bioengineering). While I am still hooking up brains to computers, I am now using a less widespread technique for such purposes: functional magnetic resonance imaging (fMRI). FMRI is an indirect technique of measuring brain activity by detecting associated changes in blood flow. It requires lying very still in a very expensive and non-portable scanner. So why would I possibly prefer using such a technique for developing BCIs, instead of the much cheaper and portable EEG? Rather than looking at BCIs through the lens of assistive technology, I now apply BCIs as a new experimental paradigm in cognitive neuroscience. Cognitive neuroscience is a field that studies the neural correlates and mechanisms underlying mental processes such as attention, memory, language and other higher-level thinking. Since the introduction of fMRI in the early 90s, the approach of mapping brain function (such as cognitive processes) to certain brain areas usually follows the same principle: a neuroscientist designs a task and investigates the brain regions that respond to it. However, this mapping between cognitive task and neural processes is often not straightforward as the same region can be activated by inherently different tasks. To overcome this limitation, I have developed together with my supervisor Rob and brilliant collaborators from Imperial College and King’s College London a novel framework ‘The Automatic Neuroscientist’ that combines real-time fMRI and state-of-the-art machine learning techniques. The idea is quite straightforward: we start by focusing on a particular brain region and try to find a set of tasks that maximally activate that brain region. This is done in form of a closed-loop BCI: the human participant’s brain activity in response to a current task is analysed instantaneously using real-time fMRI and based on this, a clever algorithm proposes what task will be presented to the participant in the next iteration. The benefit of this framework is that we can test over many more possible tasks in a far more efficient manner than the ‘standard’ paradigm would allow. By considering more tasks simultaneously, we hope to accelerate our understanding of how cognition and the brain interrelate. I am now working on extending this BCI to non-invasive brain stimulation. Our aim is to personalize stimulation therapy by finding stimulation parameters that are most optimal for a patient’s underlying pathology.
Despite being highly enthusiastic about my work and neuroscience in general considering the fast-paced and high-impact research carried out in the field, I likewise consider myself a neurosceptic as I am aware of the limitations of the techniques applied and the considerably narrow scope of questions it enables us to answer. Public engagement in neuroscience and neurotechnology should therefore not be limited to communicating the potential benefits and impact on society but provide a holistic information landscape that allows the public to question, criticise and debate latest headlines in the news. Using art as a medium for this purpose is a playful, aesthetic and stimulating starting point to foster dialogue between scientists and the public. This made me curious to get in touch with the AXNS collective and I am very thankful to now join Rachel, Miranda and Rachel in their mission.