Keynote speakers
Computations on the Neuronal Manifold
Univ.-Prof. Dr.-Ing. Moritz Grosse-Wentrup
Faculty of Computer Science, University of Vienna
In computational neuroscience, the design of handcrafted models of
neuronal circuits has been highly fruitful in elucidating how neuronal
computations are realized in small model systems. Recent developments in
neuronal imaging techniques, such as calcium imaging, have expanded the
scope of study to larger neuronal populations and complex behaviors,
overwhelming traditional analysis methods. As a result, machine learning
and AI models are increasingly adopted to analyze the relation between
neuronal dynamics and behaviors. However, it remains uncertain whether
these techniques can provide the same mechanistic insights as
traditional methods in small models or what new advancements they offer
in cognitive neuroscience. In this talk, I present our efforts to
develop AI algorithms that infer the algorithms implemented by neuronal
dynamics from neuronal data. While an algorithmic description of a
neuronal system does not per se provide mechanistic insights into how a
neuronal circuit realizes its computations, I argue that the algorithmic
level provides valuable insights into how neuronal dynamics give rise to
cognition and its disorders. I showcase our results on calcium imaging
data recorded in the nematode C. elegans.
Experience Dependent Neuroplasticity – lessons from studies of childhood hemiplegia
Professor Dido Green, Department of Rehabilitation,
Jönköping University, Jönköping, Sweden
Childhood hemiplegia (CH), characterised by predominant unilateral motor impairment, is the most common type of cerebral palsy but may also be acquired later in childhood. Typically it occurs as a result of unilateral bias of brain malformation, periventricular haemorrhage or peri-ventricular leukomalacia, posthaemorrhagic porencephaly or middle cerebral artery infarct. Type, extent and timing of cerebral insult seriously influences brain development, impacting on sensori-motor areas, corpus callosum and corticospinal tract integrity with large variability demonstrated on brain imaging findings, degree of hand function impairment and developmental outcomes. The identification of the ‘best responders’ to treatment from the research evidence remains elusive.
There is inconclusive evidence that neuroplastic changes reflecting ipsi-lateral CST connectivity to the affected hand, measured by magnetic resonance imaging (MRI) or transcranial magnetic stimulation (TMS) prior to intensive motor interventions, predict response to intensive unimanual therapies. Less is known about the interaction between movement and behaviour and the neuroplasticity and mechanisms of change surrounding bimanual control in CH in response to bimanual intervention. In this lecture I will take a multimodal approach to the understanding of neuroplastic changes in childhood movement disorders and implications for motor learning, demonstrated through our research. I will propose models by which Machine Learning may contribute to the understanding of experience dependent neuroplasticity. Consideration will be given to the theoretical as well as empirical evidence of the mechanisms by which the interventions (should) work to guide future research and clinical practice.
When Physics Meets Neuroscience: A Dynamical Systems Approach to (Loss of) Motor Adaptability During Aging
Jean-Jacques Temprado1 and Julia Jakubowska2
1. PhD, Professor, Aix-Marseille University – Institute of Movement Science
2. PhD student, Translational Research Network in Motor Disorder Rehabilitation (TReND).
In recent years, the use of theoretical physics and advanced mathematical modeling with aging biology, collectively known as „Gerophysics,” has provided new insights into the mechanisms underlying aging and longevity. Physics-based frameworks, complex systems science, and quantitative mathematical models, initially introduced in the 1980s to study the mind-brain-body system, have also a great potential to elucidate behavioral adaptability and its decline during aging.
This lecture aims to showcase how the frameworks of Coordination Dynamics and Loss of Complexity offer new perspectives on aging within the neurocognitive and behavioral system. After presenting a historical overview of these frameworks’ introduction to cognitive neuroscience and human movement science, we will show how they can explain the control of complex coordination movements and the critical roles played by sensory and cognitive mechanisms in behavioral adaptability (J.J. Temprado). Then, we will consider how the relationship between motor variability — a potential marker of age-related loss of complexity in the neuro-behavioral system — and adaptive capacities in coordinated movements can be experimentally investigated and modelled in older adults (Julia Jakubowska).