Keynote Speech

Keynote Speakers

Titles, Abstracts, and Biographies

Prof Roi Cohen Kadosh

Title: Beyond Explanation: Engineering Cognition through AI and Neurostimulation

Abstract: Psychology, neuroscience, and computational modelling have long sought to explain how the mind works. My research asks a more ambitious question: can we use that knowledge to change cognition in precise, personalised, and scalable ways? In this keynote, I will present a programme of work at the intersection of cognitive neuroscience, computational modelling, and artificial intelligence that aims to do exactly that. Focusing on non-invasive brain stimulation, I will show how we have moved beyond one-size-fits-all approaches towards AI-guided personalised intervention. By integrating behavioural, neural, and computational data, this work reveals why individuals differ in their response and how those differences can be harnessed to optimise human performance. I will argue that the next frontier is intelligent intervention: building models that not only explain cognition, but predict it, personalise it, and demonstrate that it can be improved.

Bio: Professor Roi Cohen Kadosh is a cognitive neuroscientist and academic leader working at the intersection of cognitive science, computational modelling, and artificial intelligence. He is Professor of Cognitive Neuroscience and Head of the School of Psychology at the University of Surrey, and Founder and Director of Cognite Neurotechnology.

His research focuses on the mechanisms underlying learning, attention, and cognitive performance across the lifespan. Integrating cognitive neuroscience with computational approaches, he has pioneered AI-driven personalisation of cognitive interventions, including the development of Bayesian optimisation methods for individualised brain stimulation and fully remote, home-based neurotechnology systems.

Professor Cohen Kadosh’s work bridges fundamental science and real-world application, spanning education, neurodevelopmental conditions such as ADHD, and human performance. His research has contributed to theoretical advances in understanding brain functions and has informed clinical innovation, policy, and regulation of neurotechnology. He has published over 200 research outputs in leading journals and has received international recognition for both scientific excellence and translational impact.

Dr Ilyena Hirskyj-Douglas

Title: Opportunities and pitfalls in animal-computing for cognitive science 

Abstract: The tools we use to study animal cognition have largely been inherited from human-computer interactions, and yet how animals interact with computers is fundamentally different from human participants. When animals fail to comply, we tend to exclude them, preprocess the data, or accept the limitations rather than question whether the technology itself is the problem. In this talk, I will put forward that animal-centred computing offers cognitive science a different approach to designing technology around animals' own behaviour, physiology, and ecology. Drawing across my work, from lemurs to parrots, I will show how iterative, animal-centred design can improve research validity, but also surfaces new questions about cognition that top-down methods obscure. This offers us, as those who develop technology, a new lens and tools for understanding and interpreting non-human minds. 

Bio: Dr Hirskyj-Douglas is an Assistant Professor at the University of Glasgow and runs the Animal-Computer Interaction Group,  developing technology that gives animals choice and control over their own lives. The recipient of an ERC Starter Grant (FUTUREFAUNA), she builds computer systems ranging from animals video-calling their owners and each other to non-human primates managing their own environment, using technology as a new window into animal cognition and wellbeing. Her work has been featured in the New York Times and a Netflix documentary, and challenges the assumption that technology is exclusively human territory.

Dr Letizia Gionfrida

Title: Human-in-the-Loop Optimization for HRI 

Abstract: The next generation of assistive and wearable robotic systems requires a shift from task-driven automation towards human-in-the-loop intelligent interaction. In this talk, I will present recent work from the Vision in Human Robotics Lab on bio-inspired robotic systems that integrate multimodal human sensing, biomechanical modelling, predictive digital twins, and adaptive actuation to enable personalized and responsive human-machine interaction.

By learning from visual perception and multimodal sensing, and through musculoskeletal models, our work aims to continuously capture human movement to understand physical behaviour, and individual adaptation in real-world environments. A key aspect of this research is the development of holistic predictive frameworks that study human-robot interaction not only from a control perspective, but also through the physical and biological dynamics underlying human motion and assistance.

I will discuss how these approaches can support adaptive control architectures in which robotic behaviour emerges from biological principles and evolves in response to the user’s state, intent, and functional needs. Overall, this work highlights the role of intelligent perception and bio-inspired design in creating more personalized, intuitive, and human-centred robotic systems.

Bio: Dr Letizia Gionfrida is an Assistant Professor in Computer and Robot Vision in the Department of Informatics at King’s College London, where she leads the Vision in Human Robotics Lab. She received her PhD in Bioengineering from Imperial College London and subsequently worked as a Postdoctoral Research Fellow at Harvard University. Her research focuses on bio-inspired robotics and human-centred AI, developing intelligent systems that integrate continuous human sensing, biomechanical modelling, digital twins, and adaptive actuation for next-generation human-robot interaction. Her work investigates predictive and holistic models of human-robot interaction, where robotic behaviour emerges from biological principles and physical human dynamics.

Her research outputs have been published in leading robotics journals and international conferences, including Science Robotics, Nature Communications, ICORR, and BioRob, and have attracted international media coverage. Her work has been supported by major funding bodies including the Royal Academy of Engineering, EPSRC, the Leverhulme Trust, and charitable foundations, with approximately £1 million secured as Principal Investigator within the first years of her appointment. Dr Gionfrida is the recipient of the Royal Academy of Engineering Research Fellowship and the Philip Leverhulme Prize, alongside recognitions including the Italy Made Me Award from the Italian Embassy. Since 2023, she has served as Associate Co-Chair of the IEEE Robotics and Automation Society Technical Committee on Computer and Robot Vision. She is also actively involved in initiatives supporting diversity and inclusion in robotics, including Black in Robotics.

Prof Tony Prescott

Title: A neurorobotics perspective on the sense of self

Abstract: What is the self? The human condition is defined by our awareness that we are distinct from the world, that we are the same person from day to day, even though our bodies change, and that other people are also selves. But we still do not really know what we are. As William James explained more than a century ago, the dual nature of the self lies at the heart of the mystery—the self is this most unusual thing, in that it is both the perceiver of itself and the content of what it perceives. In this talk I will propose that we can better understand both of these aspects of self through embodied (robotic) modelling of related brain systems; this is the neurorobotics perspective. I will argue that the self begins with the brain’s discovery of the body, of its ability to control it, and of the distinctive dynamics of interoceptive compared to exteroceptive sensory signals. These processes give rise to a minimal sense of self as a bounded agent, upon which layers of reflective self-processes are constructed that extend the self in space and time and that allow the recognition of other selves. To illustrate the talk, I will discuss ongoing attempts to understand the sense of self using robots from my own lab and others.

Bio: Tony Prescott is Professor of Cognitive Robotics at the University of Sheffield in the UK. His background mixes psychology, neuroethology and brain theory with robotics and AI, and his research aims at answering questions about natural intelligence by creating synthetic entities with capacities such as perception, memory, emotion and sense of self. He has co-founded the International Living Machines conference series, and two UK companies developing robotic platforms and software. He has worked extensively on brain-inspired cognitive architectures for both mammal-like and humanoid robots, and with his collaborators he has created a number of novel animal-like robot platforms including the commercial biomimetic robot MiRo-e. Tony has co-authored over 250 journal articles and conference papers and the edited volumes Scholarpedia of Touch and the Handbook of Living Machines. His popular science book The Psychology of Artificial Intelligence, published in 2024, explores the similarities and differences between human and artificial intelligence. Tony’s research is currently supported by the EIC Pathfinder project CAVAA which is concerned with possibility of creating artificial awareness in machines. His research has been covered by major news and scientific media including the BBC, CNN, Discovery Channel, Science Magazine, France 24 and New Scientist.

Dr Max Birk

Title: Interdisciplinary Challenges for Game-based Technologies

Abstract: Video games are the defining medium of our time. More than 3 billion people play regularly; design strategies have become so powerful that they attract regulatory scrutiny, and in some cases governments restrict gameplay because it is too engaging. At the same time, game-based health applications show promising results under controlled conditions but often fall short in real-world settings. In this talk, Max explores the interdisciplinary opportunities and pitfalls of game technologies: how to design for engagement, opportunities of measuring cognition through gameplay, and how scalable approaches to characterizing gameplay can provide new tools to harness the power of games responsibly and effectively.

Bio: Max Birk is an Associate Professor in the Human-Technology Interaction group at Eindhoven University of Technology. With an interdisciplinary background, Max draws from psychology, interaction design, data science, and game design, to investigate the effects of game-based design strategies on mental processes and design-induced behaviour change. His research contributes to games user research, digital health, and motivational interface design. He is interested in projects contributing to a healthy society, improving entertainment experiences (e.g., eSports), and developing tools and methods for researching interactive experiences. In his ERC funded project, GAMECHAR, Max is building AI-driven approaches for at-scale gameplay characterization. 

Lewis Hotchkiss

Title: Building Trustworthy AI for Dementia to Enable Clinical Translation

Abstract: The Dementias Platform UK Data Portal is a Trusted Research Environment which provides secure data access to over 3.5 million participants worth of multi-modal data for health data research. Over the past few years, DPUK along with many other TREs have seen a significant increase in project applications for AI model development which has meant we have had to develop new governance and infrastructure to be able to enable this type of research in a safe and secure way. Through DARE UK funded initiatives, we have established governance frameworks and processes to ensure the responsible development and deployment of AI models into clinical practice. This talk will showcase the experiences of the DPUK Data Portal in developing guidelines for responsible AI models, within the context of neuroimaging data for dementia, and how we are enabling the clinical translation of models into practice.  

Bio: Lewis Hotchkiss is the Senior Research Officer at the Dementias Platform UK Data Portal where he leads research on Responsible AI, Neuroimaging, and Federation within Trusted Research Environments. His research bridges the gap between governance and technology to streamline data access and AI innovation. He has received several grants from DARE UK to lead the AI Risk Evaluation Group and the Synthetic Data Community Group which focus on developing governance and tools to support the development of AI in sensitive data research, and he also now works for SeRP, leading the scientific use cases for data federation across multiple Trusted Research Environments.