NOW 2025: Nature-Neurotechnology-Olfactory-Wellbeing Workshop

• Dr. Mihoko Otake-Matsuura, Cognitive Behavioral Assistive Technology Team (CBAT), RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Tokyo, Japan
• Prof. Marek Jeziński, Institute of Information and Communication Research, Nicolaus Copernicus University (NCU), Toruń, Poland

This workshop will explore the multifaceted intersection of technology, neuroscience, and aging, with a focus on promoting healthy aging and mitigating age-related neurodegenerative conditions. The discussions will span cutting-edge research in artificial intelligence, its applications in healthcare, and innovative approaches to understanding and influencing brain health. Key areas of focus include:

• AI Theory & Applications: Delving into the theoretical underpinnings of AI and exploring novel methodologies with the latest applications.
• AI in Healthcare: Examining the transformative potential of AI in revolutionizing healthcare practices and improving patient outcomes.
• The Role of Smell and Taste in Healthy Aging: Investigating the impact of olfactory and gustatory systems on cognitive function and overall well-being in older adults.
• Sleep, Circadian Rhythms, Respiration, and Meditation for Healthy Aging: Analyzing the effects of these key factors on brain health and exploring interventions to promote healthy aging.
• Music for Healthy Aging and Dementia Prevention: Exploring the therapeutic potential of music in maintaining cognitive function and preventing dementia.
• Neurotechnology and Multisensory Stimulation for Healthy Aging: Investigating the latest advancements in neurotechnology and multisensory stimulation techniques for promoting healthy aging and intervention in dementia.

• Prof. Tomasz Piotrowski, Institute of Engineering and Technology, Nicolaus Copernicus University (NCU), Toruń, Poland
• Dr. Minh Ha Quang, RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Tokyo, Japan

• AI Theory & Applications: Theory and Methods
• AI in Healthcare: Pioneering the Future of Healthcare with AI

Abstract: Current source reconstruction is a tool of fundamental importance for EEG analysis. Unfortunately, the multitude of methods and reconstruction parameters pose great challenges to the proper application of reconstruction. In this presentation, I will discuss reconstruction methods from physical fundamentals and mathematical methods to practical applications. Reconstructions for synthetic sources and for real MEEG recordings will be discussed. The importance of method selection and reconstruction parameter choices will be analyzed.

Bio: Tomasz Górski is a computational neuroscientist. His research involves modeling spiking neural networks, including modeling slow-wave sleep, modeling neurons with neuronal adaptation, and modeling dendritic activity. On the EEG signal analysis side, his research focuses on current source reconstructions. He received his PhD in theoretical physics working at the Center for Theoretical Physics of the Polish Academy of Sciences. He has worked at research centers in France (Unit of Neuroscience, Information and Complexity, CNRS and at the European Institute for Theoretical Neuroscience in Paris) and in Switzerland (University of Bern). He is currently working at the Department of Physics, Astronomy, and Informatics at the Nicolaus Copernicus University in Toruń.

“Developing Japanese Vision Language Models” by Ryuichiro Hataya, SB Intuitions, Tokyo, Japan

Abstract: As large language models and vision language models gain widespread adoption, their cultural and linguistic alignment becomes increasingly critical. This talk highlights SB Intuitions Corp's development of Japanese vision language models and explores the key differences between academic and industrial research based on the speaker's recent experience.

Bio: Dr. Ryuichiro Hataya is a senior research scientist at SB Intuitions, a company developing Japanese Large Language Models and Vision Language Models. His research focuses on efficient deep learning through differentiable programming, meta-learning, and the development of multimodal language models. Prior to his current position, he was a postdoctoral researcher at RIKEN AIP and RIKEN ADSP after receiving his Ph.D. from the University of Tokyo in 2022.

“Joint Empirical Risk Minimization for Positive-unlabeled Data” by Wojciech Rejchel, Nicolaus Copernicus University, Toruń, Poland

Abstract: Learning from positive and unlabeled data (PU learning) is to train a binary classifier based on a data set containing a part of positives which are labeled, and unlabeled instances. An unlabeled set includes the remaining part of positives and all negative observations. An important element in PU learning is modeling a labeling mechanism, i.e. assigning labels to positive observations. Unlike in many prior works, we consider a realistic setting for which probability of label assignment, i.e. a propensity score, is instance-dependent. We investigate a minimizer of a ,,joint’’ risk which depends on both posterior probability of a positive class and the propensity score. This non-convex empirical risk is alternately optimised with respect to parameters of both functions. The theoretical and practical properties of the algorithm will be provided.

Bio: Wojciech Rejchel is an Associate Professor at the Faculty of Mathematics and Computer Science of Nicolaus Copernicus University in Torun (Poland). There he received the his Ph.D. in mathematics in 2011 and his habilitation in 2022. His research interests focus on statistics and machine learning, in particular high-dimensional statistics, penalized methods, model selection, partially observed data, Monte Carlo methods and empirical processes.

“Mapping Brain Activity: Spectral Fingerprinting Reveals Frequency Gradients” by Michał Lemańczyk and B-NICER Team, Nicolaus Copernicus University, Toruń, Poland

Abstract: Brain oscillations are crucial for neural communication, as frequency modulation enables interactions between different functional brain regions. Recognizing distinct spectral patterns is essential for understanding functional connectivity. Spectral Fingerprinting (SF) has been shown to be an effective method for identifying region-specific activity within the frequency domain. We analyzed eyes-open resting-state MEG (rMEG) data from 89 participants in the Human Connectome Project. Source activity was estimated using Linearly Constrained Minimum Variance (LCMV) beamforming on 1-second epochs. Spectral estimates for the 1.5–34.5 Hz range were computed for each ROI and epoch, followed by k-means clustering to identify consistent frequency patterns across individuals. To evaluate pairwise spectral similarity, we employed Earth Mover’s Distance (EMD), generating a similarity matrix. Finally, diffusion embedding was used to extract spectral gradients across ROIs. We identified four gradients that accounted for 67% of the variance. Notably, the first gradient followed an anterior-posterior spatial pattern, while the second spanned an axis from sensorimotor to visual areas, with associative regions in between. Our findings align with previous reports indicating that sensory areas primarily engage lower frequencies, whereas associative regions tend to shift toward higher frequencies. Moreover, the spatial patterns of the gradients resemble those observed in functional connectivity studies using fMRI. The study was supported by the National Science Centre Poland (UMO-2016/20/W/NZ4/00354) grant.

Bio: Michał Lemańczyk is a mathematician specializing in high-dimensional probability, concentration of measure phenomena, machine learning, information theory, ergodic theory, and thermodynamics. He is an assistant professor at IS faculty of Nicolaus Copernicus University in Toruń. Recently, as a member of B-NiCER team, he pursues problems connected with EEG signals mainly focusing on a method called spectral fingerprinting.

Coffee Break

“MVP-NAI Framework for E/MEG Inverse Problem Using MNE-Python” by Julia Jurkowska, and Tomasz Piotrowski, Faculty of Physics, University of Warsaw, Poland, and Nicolaus Copernicus University, Toruń, Poland

Abstract: We introduce two novel families of reduced-rank MVP-based neural activity indices (MVP-NAIs), designed to provide unbiased source localization even when signal and noise covariance matrices are derived from finite sample data. These newly proposed MVP-NAIs offer provably higher spatial resolution than existing state-of-the-art neural activity indices in well-defined scenarios, bridging theoretical advancements with practical applications. Moreover, they establish a unified reduced-rank framework that subsumes previously known reduced-rank NAIs. To facilitate adoption, we integrate MVP-NAIs alongside MV-PURE spatial filters into a comprehensive reduced-rank localization and reconstruction pipeline, implemented as an extension of MNE-Python. This package includes user-friendly tutorials and detailed case studies, demonstrating the effectiveness of our methods, e.g., on both simulated data and real-world EEG recordings, such as oddball experiment paradigms. By making these tools accessible to the broader neuroscience and machine learning communities, we aim to advance high-resolution EEG source imaging and promote reproducible research.

Bio: Julia Jurkowska is a student in Applications of Physics in Biology and Medicine, specializing in Neuroinformatics, at the Faculty of Physics, University of Warsaw. A recipient of the DeepMind Scholarship Programme, conducting a master’s thesis investigating whether machine learning methods can be used to differentiate individuals diagnosed with dyslexia from typical readers based on EEG signals. Her research focuses on advanced signal processing techniques applied to EEG data.

“The Most Stable Changing Structures throughout Life in Healthy Aging: Brain Age Prediction from Morphometric MRI Data using Machine Learning” by Maria Waligórska and Tomasz Wolak, Faculty of Physics, University of Warsaw, Poland, and Bioimaging Research Center, Institute for Physiology and Pathology of Hearing, Kajetany, Poland

Abstract: Accurately predicting brain age from structural MRI data is valuable for understanding the healthy aging process. In this talk, I will present the results from machine learning models trained to estimate brain age using morphometric data extracted from MRI images. We focused on exploring different preprocessing strategies and identifying the brain structures that contribute most to accurate age prediction. In contrast to many studies that focus on markers of dementia, our research highlights brain structures that exhibit the most consistent and systematic changes over the lifespan.

Bio: Maria Waligórska is a master's student of Neuroinformatics at the Faculty of Physics, University of Warsaw. After completing my Bachelors degree in Neuroinformatics I began working at the Bioimaging Research Center at the Institute for Physiology and Pathology of Hearing in Kajetany, Poland. My research interests lie at the intersection of neuroscience and machine learning, with a focus on developing practical solutions for real-world applications and utilizing computational techniques to understand the human brain.

“Riemannian Geometry of Covariance Matrices: from Information Geometry to Optimal Transport” by Minh Ha Quang, RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Tokyo, Japan

Abstract: Symmetric positive definite (SPD) matrices, in particular covariance matrices, play important roles in many areas of mathematics and statistics, with numerous applications in various different fields, including machine learning, computer vision, brain imaging, and brain computer interfaces. The set of SPD matrices possesses rich geometrical structures, which have been extensively studied mathematically and exploited practically. In this talk, we will present an overview of the Riemannian geometrical structures of SPD matrices that are commonly used in practice. These include in particular Fisher-Rao geometry (affine-invariant metric) from Information Geometry and Wasserstein geometry from Optimal Transport. The theoretical formulations will be illustrated by examples of their practical applications.

Bio: Dr. Minh Ha Quang current research interests focus on machine learning and statistical methodologies using theories and techniques from functional analysis and related mathematical fields. In particular, He has been working on theories and methods involving reproducing kernel Hilbert spaces (RKHS), Riemannian geometry, matrix and operator theory, information geometry, and optimal transport, especially in the infinite-dimensional setting. received his PhD in mathematics from Brown University (Providence, RI, USA) and wrote a dissertation under the supervision of Stephen Smale. Before joining RIKEN, Dr Minh was a researcher at the Pattern Analysis and Computer Vision group at the Italian Institute of Technology (Istituto Italiano di Tecnologia) in Genoa (Genova), Italy. Prior to Italy, I was a postdoctoral researcher at the University of Vienna, Austria, and the Humboldt University of Berlin, Germany.

1. “Passive BCI for Dementia Prediction Using Path Signature and Riemannian Geometry Classifier” by Tomasz M. Rutkowski and Minh Ha Quang, RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Tokyo, Japan
2. “Sleep in Aging: Dementia Prediction and Intervention” by Stanisław Narębski, Nicolaus Copernicus University (NCU), Toruń, Poland
3. “I am aware of my symptoms. They don’t bother me anymore. Effects of mindfulness training on well-being of psychosomatic disorder patients - preliminary study” by Marek Chełstowski, Julia Tyszkiewicz, Natalia Krymow, Kamila Łaszewska, Monika Lewandowska, and Rafał Milner, Nicolaus Copernicus University, Toruń
4. “Olfactory Neurotechnology for Neurobiomarker Discovery of Mild Cognitive Impairment and Dementia” by Hubert Kasprzak, Academia Copernicana, Nicolaus Copernicus University (NCU), Toruń, Poland
5. “Polish adaptation of Cognitive Reserve Scale. Relationship of cognitive reserve with personality and subjective / objective cognitive functioning” by Jakub Słupczewski, Bartłomiej Kiljanek, Maria Dolores Roldan-Tapia, Juan García-García, Martyna Olszewska, Ewa Ratajczak, Nicolaus Copernicus University, Toruń, Poland; and University of Almeria, Almeria, Spain
6. “Retinal-Based Eye Tracking Biomarkers in Parkinson’s Disease: Preliminary Evidence” by Krzysztof Tołpa, Bogna Bylicka, Marta K. Skrok, Valentyna Pryhodiuk, Robert Konklewski-Pilewicz, and Maciej Szkulmowski, Nicolaus Copernicus University (NCU), Toruń, Poland; Inoko Vision sp. z o.o, Toruń, Poland
7. “Spatial Auditory Soundscapes for Developing Digital Neurobiomarkers or Cognitive Interventions in Early-onset Dementia Based on EEG and fNIRS Machine-learning Analysis” by Simon Kojima, Reiko Shiba, Yota Morimoto, Kiyoshi Furukawa, Shin'ichiro Kanoh, Tomasz Komendziński, Mihoko Otake-Matsuura, and Tomasz M. Rutkowski, RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Tokyo, Japan; Shibaura Institute of Technology, Tokyo, Japan; Waseda University, Tokyo, Japan; Tokyo University of the Arts, Tokyo, Japan; and Nicolaus Copernicus University (NCU), Toruń, Poland
8. “Cortical brain volume is related to neuropsychological tests and spatial navigation task results in the elderly with SCI” by Natalia Anna Pawlaczyk, Rafał Milner, Ewa Ratajczak, Monika Lewandowska, Institute of Psychology, Faculty of Philosophy and Social Sciences, Nicolaus Copernicus University in Torun, Torun, Poland

Copernican Integration Centre

• Dr. Mihoko Otake-Matsuura, Cognitive Behavioral Assistive Technology Team (CBAT), RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Tokyo, Japan
• Prof. Arkadiusz Gut, Department of Cognitive Science, Institute of Information and Communication Research, Nicolaus Copernicus University (NCU), Toruń, Poland

• The Role of Smell and Taste in Healthy Aging
• Sleep, Circadian Rhythms, Respiration, and Meditation for Healthy Aging
• Music for Healthy Aging and Dementia Prevention
• Neurotechnology and Multisensory Stimulation for Healthy Aging

“A retinal eye tracker combined with a DPI and P-CR eye trackers for high-accuracy and high-precision registration of all kinds of eye movements” by Maciej Szkulmowski, Nicolaus Copernicus University (NCU), Toruń, Poland

Abstract: One of the most prominent features of the eye is its constant motion. Understanding this phenomenon is of utmost importance to the scientific communities in neurology, psychology, ophthalmology, cognitive and brain sciences, as well as medicine. The purpose of this study is to simultaneously register the images from the retina and the front of the eye during fixations, saccades, and smooth pursuit, to compare the differences in the obtained eye motion traces.

Bio: Dr. hab. Maciej Szkulmowski, Professor at Nicolaus Copernicus University (UMK), has dedicated over 20 years to developing methods for imaging the morphology and function of the human retina using various optical technologies, including multiple variants of optical coherence tomography (OCT) and scanning laser ophthalmoscopy. He is also advancing techniques for measuring blood flow in the eye and developing ultra-fast, high-precision eye-tracking methods. Currently, he serves as the Head of the Department of Biophotonics and Optical Engineering, whose members contribute to the university’s OptoPhoto Centre of Excellence under the UMK Research University Initiative.

“Neural Markers of Rhythm, Timing, and Audio-motor Interactions in Parkinson’s Disease.” by Victor Manuel Pando Naude, Department of Clinical Medicine, Center for Music In the Brain, Aarhus University, Denmark

Abstract: Movement disorders such as Parkinson’s disease (PD) expose a fundamental link between rhythm, prediction, and pleasure in the human brain. In this talk, I will trace a translational arc from basic music-cognition research to rhythm-based therapeutics. First, meta-analytic work establishing how top-down/bottom-up processing reshapes neural circuits will be reviewed. I will then present behavioural and fMRI evidence showing that the “pleasurable urge to move to music” (PLUMM) depends on dopaminergic signalling in the basal ganglia and how this mechanism is attenuated in PD. These findings explain both reduced groove sensation and impaired timing in patients, and they motivate targeted interventions.

Bio: Victor Pando-Naude, MD, PhD, is an Assistant Professor of Cognitive Neuroscience at the Center for Music in the Brain, Aarhus University. His work on rhythm, timing and music’s clinical applications has earned the 2023 ICMPC Young Researcher Award and >580 citations, informing Cochrane guidelines and shaping new therapies for Parkinson’s disease, chronic pain and sleep disorders.

“Impact of Deep Brain Stimulation on Eye Movement in Patients with Parkinson's Disease” by Marta K. Skrok, Nicolaus Copernicus University (NCU), Toruń, Poland

Abstract: The analysis of eye movements facilitates a rapid and non-invasive evaluation of brain activity. However, video-based eye-trackers, widely used in eye movement research, require subject-dependent calibration and are prone to errors caused by the dynamics of pupil and lens. Our research focuses on developing an innovative calibration-free retinal eye-tracker, that allow for high precise and high accurate assessment of patients with neurodegenerative diseases. During the clinical pilot investigation, we implemented our device to assess the eye movements of subjects with Parkinson's disease who were being treated with Deep Brain Stimulation. This presentation will show the impact of activating and deactivating Deep Brain Stimulation on eye movements in response to different stimulus-driven visual tasks.

Bio: Marta K. Skrok received the B.Eng. degree in Optics, the M.Eng. degree in Optometry and Ph.D. degree in Physics from the Wroclaw University of Science and Technology, Wroclaw, Poland, in 2016, 2017 and 2023 respectively. She is currently a postdoctoral researcher at the Faculty of Physics, Astronomy and Informatics of Nicolaus Copernicus University, Torun, Poland. Her research focusses on possibility of using a prototype retinal eye-tracker in the diagnosis of neurodegenerative diseases. She is also investigating the clinical application of optical coherence tomography and optical elastography, especially in the process of cancer diagnosis and treatment.

“What Can Recurrence Quantitative Analysis Tell us about Pathological Abnormalities in Eye Movements?” by Bogna Bylicka, Nicolaus Copernicus University (NCU), Toruń, Poland

Abstract: Analysing eye movements provides a non-invasive route to assess brain health. We investigated differences in eye tracking patterns between clinical and healthy populations to identify potential biomarkers for neurodegenerative conditions. This talk introduces a framework employing Recurrence Quantification Analysis (RQA) to characterise the complex dynamics of eye movements and reveal subtle abnormalities. We illustrate this approach using a Parkinson's disease study. This application demonstrates the method's ability to detect disease-related alterations, offering interpretable insights into the underlying movement abnormalities revealed by RQA, which is of paramount importance for its translation into medical applications.

Bio: Bogna Bylicka is a doctor of physics, an assistant professor at Nicolaus Copernicus University in Toruń. She leverages her background in quantum information and artificial intelligence to explore cognitive science questions. Her main areas of interest are: the brain as a dynamical system, mind-body relation, especially in the context of autoimmune issues and neurodegenerative diseases.

Coffee Break

“Sleep and Smell Capital for Healthy Aging and Dementia Prevention - NCU Perspective” by Tomasz Komendziński, Nicolaus Copernicus University (NCU), Toruń, Poland

Abstract: Sleep quality and olfactory function are increasingly recognized as critical for cognitive health. This presentation introduces the concepts of “sleep capital” and “smell capital” – the cumulative benefits of healthy sleep and olfaction – as potential protective factors against age-related cognitive decline and dementia. From the Nicolaus Copernicus University (NCU) perspective, this talk will discuss the theoretical basis, synthesize literature on the bidirectional links between sleep, olfaction, and neurodegeneration, and outline future NCU-based research directions. Understanding and enhancing these “capitals” may offer novel strategies for healthy aging and dementia risk reduction within the Polish population and beyond.

Bio: Tomasz Komendziński is a lecturer in the Department of Cognitive Science at Nicolaus Copernicus University in Toruń. Dr. Komendziński founded and coordinates InteRDoCTor (International-interdisciplinary Research for Disorders of Consciousness in Torun), a group focused on advancing research in this field. He holds memberships in numerous national and international scientific societies, including the Cognitive Science Society and the Society for Social Neuroscience, and serves on the editorial boards of journals such as “Phenomenology and Cognitive Sciences” and “Frontiers in Neuroscience – Neural Technology.” As a member of the Emerging Field – Perception, Cognition and Language (PeCoLa) within the IDUB initiative, he received funding to establish a diagnostic unit for patients with consciousness disorders. He co-organizes the international “Science Bridges” seminar. Dr. Komendziński's research focuses on disorders of consciousness, interdisciplinary and transdisciplinary studies, and the integration of research methods, particularly in brain and mind research. He develops theories of communication as sharing, embodied cognitive science, and the application of cognitive science in medicine, with recent emphasis on deep medicine, new medical competencies related to advanced technologies and AI (especially in consciousness disorders), and the challenges of Society 5.0, AI for Social Good, and Inclusive Society. Currently, he is involved in projects concerning transcutaneous vagus nerve stimulation for seniors' wellbeing and children's visual competence, and participates in a project on art perception using modern brain research methods. He has led and participated in many interdisciplinary projects and is a member of a Polish team in an international cross-cultural study on compassion, social bonds, and resilience during the Covid-19 pandemic. He has authored over 100 publications and organizes scientific conferences. He is also the academic advisor of the Toruń Cognitive Science Association.

“A Few Attempts to Use EEG Signals for Diagnosis” by Włodzisław Duch, Nicolaus Copernicus University (NCU), Toruń, Poland

Abstract: Despite numerous sophisticated signal processing approaches, the identification of reliable clinical EEG biomarkers for brain disorders remains a significant challenge. We summarize our recent approaches aimed at discovering parsimonious yet informative representations of complex EEGsignals, based on techniques such as Recurrence Quantitative Analysis (RQA), Microstates, asymptotic spatiotemporal averaging of power, similarity of transition probability matrices, and homological features from topological data analysis. We also explore different approaches to EEG channel and band selection. Using such methods we are analyzing the Default Mode and Executive Networks, Cortical-Thalamic Information Transfer and Theta-Gamma and Alpha-Beta Coupling Dynamics, highlighting temporal patterns in resting-state EEG using autoencoder latent spaces and multivariate angular distance analysis. Preliminary results demonstrating the potential of these approaches for schizophrenia and Mild Cognitive Impairment diagnosis will be presented. This work was done in collaboration with Łukasz Furman, Krzysztof Tołpa, Ewa Ratajczak, Luis Alexandre, Simone Poetto, and Ludovico Minati.

Bio: Wlodzislaw Duch is the head of the Neurocognitive Laboratory in the Center of Modern Interdisciplinary Technologies, and the Neuroinformatics and Artificial Intelligence group in the University Centre of Excellence Dynamics, Mathematical Analysis and Artificial Intelligence at the Nicolaus Copernicus University, Toruń, Poland. PhD in theoretical physics/quantum chemistry (1980), postdoc at the Univ. of Southern California, Los Angeles (1980-82), D.Sc. in applied math (1987). President of the European Neural Networks Society executive committee (2006-2008-2011), Fellow of the International Neural Network Society (2013), Asia-Pacific Artificial Intelligence Association (2022), and the International Artificial Intelligence Industry Alliance (2024). Expert of the European Union science programs, member of the high-level expert group of European Institute of Innovation & Technology (EIT). Worked in the School of Computer Engineering, Nanyang Technological University (2003-07, 2010-12 as the Nanyang Visiting Professor). Visiting professor at the University of Florida; Max-Planck-Institute, Munich, Germany, Kyushu Institute of Technology, Meiji and Rikkyo University in Japan, and several other institutions. He is/was on the editorial board of IEEE TNN, CPC, NIP-LR, Cognitive Neurodynamics, Journal of Mind and Behavior, and 16 other journals; published over 380 peer-reviewed scientific papers, has written or co-authored 6 books and co-edited 21 books, and published about 300 conference abstracts and popular articles on diverse subjects. In 2014-15 he served as a deputy minister for science and higher education in Poland. His DuchSoft company has made in 1990 the GhostMiner data mining software package, for many years marketed by Fujitsu. With a wide background in many branches of science and understanding of different cultures he bridges many scientific communities. To unwind he plays electronic wind instruments and dives with whale sharks. To find more information about his activities (CV, Academic Genealogy, plans) type his full name in the address bar of your browser, or check his full CV here.

“A Robot Moderating Group Conversation for Cognitive Intervention to Healthy Older Adults” by Mihoko Otake-Matsuura, RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Tokyo, Japan

Abstract: Embodied intelligence is demonstrated in conversation since thoughts and feelings are exchanged through visual, auditoryand motor functions. Multiple cognitive functions, such as memory, comprehension, reasoning, and executive functions, are utilized in conjunction with bodily functions. Based on these characteristics of conversation, we are developing technology to intervene in and help maintaining and improving the cognitive functions of older adults through conversation. In this talk, we demonstrate a robot moderating group conversation and describe the effects of group conversations among humans using a moderator robot on human cognitive functions.

Bio: Mihoko Otake-Matsuura is a Team Director (PI) of Cognitive Behavioral Assistive Technology Team, Center for Advanced Intelligence Project, RIKEN, Japan. She is concurrently the founding director of the NPO Fonobono Research Institute, which is a platform for citizen science, and Adjunct Professor of Tokyo University of Agriculture and Technology. She is one of the pioneers in the field of soft robotics and her thesis on “Gel Robots” was published as a monograph in Springer Tracts in Advanced Robotics series. Her current research focuses on technology for cognitive health in super aged society where she successfully collected the world first evidence of cognitive intervention via conversation supported by robots. Her research interests include but not limited to: communication support system, human-robot interaction, brain-computer interface, AI and robotics for super aged society.

Lunch break on your own

“Biomarkers of Age-Related Decline in Spatial Orientation” by Bibianna Bałaj, Nicolaus Copernicus University (NCU), Toruń, Poland

Abstract: Spatial orientation—the ability to perceive and navigate within one’s environment—often declines with aging, impacting autonomy and safety. Recent research highlights eye movements as sensitive and non-invasive indicators of spatial orientation abilities. Age-related changes in oculomotor behavior, such as reduced saccadic accuracy and increased fixation duration, are linked to diminished spatial awareness and slower navigational decision-making. These alterations may reflect underlying deficits in visuospatial attention, working memory, and the integrity of neural circuits involving the parietal and frontal eye fields. Monitoring eye movement dynamics offers a promising approach for the early detection of spatial disorientation and cognitive impairment in aging.

Bio: Bibianna Bałaj is a doctor of psychology, deputy director of the Institute of Psychology at the Faculty of Philosophy and Social Sciences of the Nicolaus Copernicus University. She researches spatial orientation, visual imagery, cognitive control, and memory. Her interests focus on mental rotation and perspective-taking, which are revealed by eye movements. Scan paths and saliency models when viewing art are exciting to her. She is also interested in gaze-based interaction in computer-aided communication for users with physical disabilities.

“ASME-BCI: An Auditory Brain-Computer Interface Leveraging Auditory Stream Segregation” by Simon Kojima, Shibaura Institute of Technology, Tokyo, Japan

Abstract: The auditory brain-computer interface (BCI) has the advantage of not occupying the visual field and being suitable for everyday use. However, its performance is generally lower than that of visual BCIs. In this talk, an auditory BCI (ASME-BCI) is introduced, which detects the user’s selective attention to multiple auditory streams by leveraging the perceptual processing, known as auditory stream segregation. In addition, recent research on an auditory BCI speller utilizing the ASME paradigm is also presented.

Bio: Dr. Simon Kojima is a postdoctoral researcher at the Inria Centre at the University of Bordeaux, France. His research interests lie in brain-computer interfaces (BCI), neural engineering, neuroscience, and machine learning. He received his Ph.D. from Shibaura Institute of Technology in Tokyo, Japan, under the supervision of Prof. Shin’ichiro Kanoh. His doctoral research focused on developing auditory BCI systems by leveraging auditory scene analysis and exploring how complex sound environments can enhance BCI performance. Before joining Inria, he conducted research as an intern at the RIKEN Center for Advanced Intelligence Project (AIP) in Tokyo and as a visiting researcher at the Donders Institute for Brain, Cognition, and Behaviour at Radboud University in Nijmegen, the Netherlands.

Coffee Break

“Passive BCI for Dementia Prediction Using Information Geometry Tools” by Tomasz M. Rutkowski, RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Tokyo, Japan; and Nicolaus Copernicus University (NCU), Toruń, Poland

Abstract: This talk will discuss the use of passive brain-computer interface (pBCI) for assessing brain health and detecting neurodegenerative processes. It will focus on a methodology using path signatures and Riemannian geometry to analyze noisy EEG data, targeting MCI, and present preliminary results that demonstrate the potential for creating digital biomarkers.

Bio: Dr. Tomasz M. Rutkowski's career is marked by a passion for understanding the brain and developing cutting-edge technologies to improve human health. After earning his M.Sc. and Ph.D. from Wroclaw University of Technology, he held research positions at Kyoto University and RIKEN Brain Science Institute in Japan. His drive to translate research into practical applications led him to a role at an AI startup in Tokyo. Currently, he is a research scientist at the RIKEN Center for Advanced Intelligence Project (AIP) and a research fellow at The University of Tokyo and Nicolaus Copernicus University. Dr. Rutkowski's research focuses on brain-computer interfacing (BCI), computational modeling of evoked brain processes and awareness, and the application of AI for elucidating dementia biomarkers. His achievements include The BCI Annual Research Award in 2014 and a First Degree Scientific Award from Nicolaus Copernicus University in 2024. He is also dedicated to promoting diversity in science as a jury member for the Maria Sklodowska-Curie Prize for Young Female Scientists in Japan. A senior IEEE member, his work is detailed at http://tomek.bci-lab.info/.

“The Promises and Pitfalls of Passive Neuroart Brain-Computer-Interfaces for Health and Well-being” by Marc Welter, Inria, LaBRI, Bordeaux, France

Abstract: A growing amount of publications claim strong evidence for beneficial health and well-being effects of art engagement. It seems plausible that Neuroart Brain-Computer-Interfaces (BCIs) could optimize beneficial neural and mental states during art experience, e.g. by presenting user-specific artworks that reduce neuromarkers of stress for this user. This talk will present the state-of-the-art in passive Neuroart BCIs, and discuss the major challenges towards health and well-being applications outside the lab.

Bio: Marc Welter is a doctoral researcher in Computer Science supervised by Fabien Lotte at the Inria Centre at the University of Bordeaux, France. His research interests lie in brain-computer interfaces (BCI), neuroaesthetics, machine learning and philosophy. His doctoral research focused on developing a hybrid passive BCI to improve user experience in virtual art museums. Before joining Inria, he received a B.A. in Philosophy and a M.S. in Embodied Cognitive Science and has worked as a research assistant in cognitive science (University of Potsdam), computer vision (Charité Berlin), natural language processing (Karlsruhe Institute of Technology) and psychology (University of Freiburg).

“Intermedia and Wholistic Music: from Audiovisual Piano to VR” by Jarosław Kapuściński, Stanford University, CA, USA

Abstract: Composer Jarosław Kapuściński specializes in creating interactive audiovisual works that are primarily meant to engage audiences in art contexts but they also have the potential to enhance well-being and support healthy aging. His compositions often involve musicians, particularly pianists, who control visual content, or more recently, general audiences who can explore and create within immersive VR environments. This unique interplay of music, visuals, and audience participation may foster cognitive stimulation and emotional engagement—key elements in promoting overall well-being. In this presentation, Kapuściński will discuss his work, highlighting his recent VR collaboration with the OpenEndedGroup, Point Line Piano, a project that reimagines the composition, performance, and reception of piano music. Participants engage with the work by drawing lines in 3D space, which trigger musical notes and create immersive audiovisual landscapes. This allows for a full-body experience of creativity and playfulness. The project's interactive nature offers possibilities for applications in workshops, therapy, and community engagement.

Info and trailer: https://www.jaroslawkapuscinski.com/works/point-line-piano/

Bio: Jarosław Kapuściński is an Associate Professor of Music at Stanford University, where he is affiliated also with the Department of East Asian Languages and Cultures. His research focuses on intermedia composition, performance, and Japanese traditional aesthetics. Kapuściński has received grants and commissions from numerous international organizations, including the National Endowment for the Arts, the Governor General of Canada, and Institut National de l’Audiovisuel (INA) in France. His works have been awarded prizes at festivals in Canada, France, Switzerland, and the United States, and have been presented at venues such as New York MOMA, Spoleto USA, EMPAC NY, Logan Center in Chicago, ZKM in Karlsruhe, Reina Sophia Museum in Madrid, Media Biennale Wroclaw, Warsaw Autumn Festival, Creative Media Center in Hong Kong, Benz Arena in Shanghai, and National Art Centre in Ottawa. In addition to his artistic work, Kapuściński has collaborated on scholarly websites about Japanese Gagaku music (gagaku.stanford.edu) and Noh Theater (noh.stanford.edu). https://jaroslawkapuscinski.com/

“Smart Sleep Science: AI and Wearables Transforming Sleep Staging and Therapy” by Manuel Schabus, University of Salzburg, Austria

Abstract: The rise of wearable devices has sparked interest in quantifying sleep, though concerns about their accuracy persist, potentially leading to negative daytime effects. I will review the respective literature and introduce an automated, low-cost, deep learning algorithm for sleep staging, offering reliable classification into four stages (Wake, Light, Deep, REM) using only inter-beat-interval (IBI) heart rate variability (HRV) data. Our multi-resolution convolutional neural network (MCNN) was trained on IBIs from more than 9000 polysomnographies (PSG) and tested for accuracy using two consumer wearables: a PPG-based heart rate sensor (Verity Sense) and an ECG chest strap (H10) from Polar©. In a study of 136 self-reported poor sleepers, the model demonstrated high accuracy (ECG: 86.3%, κ = 0.79; H10: 84.4%, κ = 0.76; PPG: 84.2%, κ = 0.75) and strong correlations with PSG on key sleep parameters (e.g., Sleep Onset Latency, Sleep Efficiency, Total Sleep Time, REM sleep). Importantly, the model remained accurate even for participants using heart or psychoactive medications, highlighting its clinical potential. In addition, brand-new data from a multi-sensor study will show how the most well known wearables on the market (Oura Gen 3, Apple Wach 9, Whoop, Fitbit, Garmin) perform as compared to gold-standard PSG. A recent randomized-controlled trial (RCT) using the sleep² algorithm showed both subjective and objective sleep improvements when combining daily sleep feedback with digital cognitive behavioral therapy for insomnia (dCBT-I). These findings support the use of cost-effective wearables for continuous, reliable sleep stage monitoring, with implications for both research and clinical practice. Conflict of Interest: MS is co-founder of the Nukkuaa GmbH and CSO for the sleep² App which developed the MCNN model for IBI-based sleep classification

Bio: Prof. Manuel Schabus is a renown sleep and brain researcher at the Centre for Cognitive Neuroscience in the Department of Psychology at the University of Salzburg. He is the founder and head of the “Laboratory for Sleep, Cognition, and Consciousness” at the university, where he has made significant contributions to the understanding of sleep, including both healthy and disordered sleep, as well as the role of sleep in learning processes. In addition to his scientific work, Prof. Schabus is a trained person-centered psychotherapist. He has published over 100 international research papers and has garnered more than 12,000 citations, reflecting his influence in the field (h-index > 50). In recognition of his groundbreaking work in brain research, Prof. Schabus was awarded the prestigious Austrian FWF START Prize in 2014 which also led to his appointment in 2015 as Chair of “Cognition and Consciousness” at the University of Salzburg (Austria).

Co-chair PL: Dr. Tomasz Komendziński, Department of Cognitive Science, Institute of Information and Communication Research, Nicolaus Copernicus University (NCU), Toruń, Poland

Co-chair JP: Dr. Tomasz M. Rutkowski, RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Tokyo, Japan

OC Members:

• Prof. Włodzisław Duch, Nicolaus Copernicus University (NCU), Toruń, Poland
• Prof. Marta Podhorecka, Department of Geriatrics (NCU), Nicolaus Copernicus University (NCU), Toruń, Poland
• Prof. Agnieszka Hamerlińska, Institute of Educational Sciences, Nicolaus Copernicus University (NCU), Toruń, Poland
• Dr. Bibianna Bałaj, Institute of Psychology, Nicolaus Copernicus, Toruń, Poland
• Dr. Minh Ha Quang, RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Tokyo, Japan
• Jan Nikadon, Department of Cognitive Science, Institute of Information and Communication Research, Nicolaus Copernicus University (NCU), Toruń, Poland
• Michał Joachimiak, Institute of Engineering and Technology, Nicolaus Copernicus University (NCU), Toruń, Poland
• Simon Kojima, RIKEN Center for Advanced Intelligence Project (RIKEN AIP) & Shibaura Institute of Technology (SIT), Tokyo, Japan
• Hubert Kasprzak, Academia Copernicana, Nicolaus Copernicus University (NCU), Toruń, Poland

• This workshop is offered free of charge and is open to all members of the academic community.
• We encourage faculty, researchers, students, and staff to participate.
• We're inviting our online community in Japan to register via DoorKeeper for the Zoom event details [link].

Instytut Psychologii UMK
ul. Gagarina 39
87-100 Toruń
Poland
Tel. +48566114171
Plus Code: 2HCH+2Q Toruń, Poland

• Nicolaus Copernicus University, Toruń, Poland
• RIKEN Center Center for Advanced Intelligence Project (AIP), Tokyo, Japan
• Department of Cognitive Science and Emerging Field - Culture, Development & Wellbeing, Nicolaus Copernicus University, Toruń, Poland

• Under the Auspices of the Rector Magnificus, Professor Andrzej Tretyn, Nicolaus Copernicus University, Toruń, Poland
• Miasto Toruń, Poland