January 12th – Matt Jones (University of Bristol)
E204, Chemistry
Working memory: some stuff on my mind
Working memory (WM) reflects the temporary storage and manipulation of information pertinent to online behaviour. Sustained activity in recurrently interconnected prefrontal cortical neurons is commonly credited as central to WM. However, recent recordings of prefrontal neural dynamics portray an increasingly complex picture involving mixed selectivity, trajectories through state space, spectral signatures and other daunting words.
Could implanting arrays of microelectrodes and optrodes into and around rat prefrontal cortex help? Maybe – I will try to set some Jones Lab findings in the context of this knotty neurodynamical nebula.
http://www.bristol.ac.uk/phys-pharm/people/matt-w-jones/index.html
January 19th – Giovanni Diana (King’s College London)
E204, Chemistry
Inference of functional subtypes and spontaneous neuronal assemblies in the zebrafish optic tectum
Understanding the functional diversity of the nervous system is one of the most prominent questions in neuroscience. Neurons are often classified based on their tuning curves under selected stimuli, however describing how neural computation is orchestrated within a complex network requires a paradigm shift from single neurons to neuronal ensembles emerging in the brain. By applying functional imaging of calcium activity in the zebrafish optic tectum we could address these questions directly. In the presence of visual stimulation we identified neuronal sub-types involved in prey-capture and escape response. In the absence of visual input, tectal neurons display synchronous activity events. We will discuss the inference methodologies that can be reliably employed to detect these spontaneous ensembles as well as the clustering techniques used to classify evoked responses.
https://devneuro.org/cdn/people-detail.php?personID=1387
January 26th – Rasmus Petersen (University of Manchester)
SM3, Maths
Sensation and locomotion in behaving mice: Investigation using ‘Neuropixel’ probes
Despite the fact that almost all regions of the brain are richly interconnected, the standardview of sensory processing is that the subcortical, ascending sensory pathway simply encodes sensory stimuli and relays them onto the cerebral cortex. However, this view is derived from experiments on anaesthetised animals where motor and cognitive circuits are disengaged, and sensation can only be studied in a reduced, passive form. Using the whisker system as a model, we have developed a paradigm for studying active sensation in behaving mice that permits us to record the activity of neurons at the same time as measuring the mechanical input to the whiskers. In our most recent work, in collaboration with Karel Svoboda, we have used high-density Neuropixel silicon probes to record the spiking activity of neurons across >10 brain regions (including Ventro Posterior Medial thalamus, Posterior Medial thalamus and ZonaIncerta) whilst at the same time measuring sensory input to the whiskers and locomotor velocity. Surprisingly, even in relay nuclei conventionally considered simply to be sensory relays, we observed modulation of neural activity by locomotion velocity. Indeed, we observed such modulation in all recorded areas, including the hippocampal formation. Our data suggest that sensory processing, right from the earliest stages of the Central Nervous System, is substantially dependent on motor control context.
https://www.research.manchester.ac.uk/portal/en/researchers/rasmus-petersen(a96ce8ab-ae04-401d-844b-82fed7bbbc0e).html
February 2nd – Melanie Stefan (Edinburgh Medical School)
SM3, Maths
It’s not what it looks like: Counter-intuitive effects in synaptic biochemistry
A lot of what we know about Biochemistry comes from tightly controlled experiments with large numbers of molecules in well-mixed test-tubes. Real biological systems such as dendritic spines are often very different from that: Absolute numbers of any one molecular species can be very small, their spatial distribution is often highly non-uniform, and they compete for ligands that are in short supply. This can give rise to behaviours that are difficult to predict and often counter-intuitive. In such cases, computational modelling can uncover and explain system behaviours that are otherwise difficult to understand. In this seminar, I will talk about a few examples from our past and ongoing research on Calcium-regulated synaptic proteins.
http://melaniestefan.net/
February 9th – Nelson Totah (Max Planck Institute)
E204, Chemistry
The locus coeruleus is a complex and differentiated neuromodulatory system
Understanding forebrain neuromodulation by the noradrenergic locus coeruleus (LC) is fundamental for cognitive and systems neuroscience. The diffuse projections of individual LC neurons and their presumably synchronous spiking have long been perceived as features of the global nature of noradrenergic neuromodulation. Yet, the commonly referenced “synchrony” underlying global neuromodulation, has never been assessed in a large population, nor has it been related to projection target specificity. We recorded up to 52 single units simultaneously (3164 unit pairs in total) in rat LC and characterized projections by stimulating 15 forebrain sites. Spike count correlations were low and, surprisingly, only 13% of pairwise spike trains had synchronized spontaneous discharge. Notably, even noxious sensory stimulation did not activate the entire population. We also identified a novel firing pattern: infra-slow (0.01-1 Hz) fluctuations of LC unit spiking, which were also asynchronous across the population. A minority of LC neurons, synchronized possibly by gap junctions, was biased toward restricted (non-global) forebrain projection patterns. Finally, we characterized two types of LC single units differing by waveform shape, propensity for synchronization, and interactions with cortex. These cell types formed finely-structured ensembles. Our findings suggest that the LC conveys a highly complex, differentiated, and potentially target-specific neuromodulatory signal.
http://www.kyb.tuebingen.mpg.de/nc/employee/details/ntotah.html
February 16th – John Grogan (University of Bristol)
SM3, Maths
Dopamine and Parkinson’s disease’s effects on working memory: behaviour, pharmacology and genetics
Parkinson’s disease (PD) causes cognitive as well as motor impairments, some of which are improved by the dopaminergic medication prescribed for motor symptoms, while others are exacerbated by it. Working memory (WM) is one such function that shows both improvements and impairments caused by dopaminergic medication in healthy people and people with PD.
We have run several studies into different aspects of WM: using the digit span in PD patients and healthy older adults given levodopa; examining the influence of genetic polymorphisms in dopamine-related genes on digit span and n-back tasks; and looking at how dopaminergic medication affects spatial WM in PD patients. We use a simple model to analyse the spatial WM task which allows us to separate different sources of errors in participants’ responses, so we can examine the precision of memory, and the mis-binding of items and locations.
These diverse approaches allow us to combine large-sample, high-powered analyses with smaller more sensitive tasks to zero in on the range of WM deficits caused by PD, and how dopamine affects these deficits. This can help us understand dopamine’s function in WM in healthy brains.
http://www.bris.ac.uk/clinical-sciences/people/john-p-grogan/index.html
February 23rd – Friedemann Zenke (University of Oxford)
E204, Chemistry
Making Cell Assemblies: What can we learn about plasticity from spiking neural network models?
Long-term synaptic changes are thought to underlie learning and memory. Hebbian plasticity and homeostatic plasticity work in concert to combine neurons into functional cell assemblies. This is the story you know. In this talk, I will tell a different tale. In the first part, starting from the iconic notion of the Hebbian cell assembly, I will show the difficulties that synaptic plasticity has to overcome to form and maintain memories stored as cell assemblies in a network model of spiking neurons. Teetering on the brink of disaster, a diversity of synaptic plasticity mechanisms must work in symphony to avoid exploding network activity and catastrophic memory loss – in order to fulfill our preconception of how memories are formed and maintained in biological neural networks. I will introduce the notion of Rapid Compensatory Processes, explain why they have to work on shorter timescales than currently known forms of homeostatic plasticity, and motivate why it is useful to derive synaptic learning rules from a cost function approach. Cost functions will also serve as the motivation for the second part of my talk in which I will focus on the issue of spatial credit assignment. Plastic synapses encounter this issue when they are part of a network in which information is processed sequentially over several layers. I will introduce several recent conceptual advances in the field that have lead to algorithms which can train spiking neural network models capable of solving complex tasks. Finally, I will show that such algorithms can be mapped to voltage-dependent three-factor Hebbian plasticity rules and discuss their biological plausibility.
https://fzenke.net/
March 2nd – Mark Olenik (University of Bristol)
SM3, Maths
Modelling struggling tadpoles: Interactions of strong and weak synaptic coupling for burst rhythm generation
Xenopus hatchling tadpoles “struggle” when held at their tail, rhythmically bending their body in an attempt to free themselves. During struggling, neurons on opposite sides of the tadpole spinal cord fire bursts of spikes in alternation, resulting in motor activity and the rhythmic bends of the whole body. A central question of interest in tadpole struggling is the role of the excitatory and inhibitory coupling for the rhythm generation. In this talk we will therefore study a reduced biophysical model of tadpole struggling that contains weakly excitatory neurons and strongly inhibitory neurons, where the latter are also subject to short-term synaptic depression. We will discuss approaches to understanding the dynamics of such a network, focusing on reducing the complex model equations to a more tractable, and intuitive, quantity. For the reduction we will employ two useful mathematical techniques: geometric singular perturbation theory and phase response curves.
http://www.bristol.ac.uk/biology/people/mark-olenik/index.html
March 9th – Ramin Azodi Avval (University of Tübingen)
E204, Chemistry
Characterization of cortico-subthalamic networks during deep brain stimulation surgery in Parkinson’s disease
Deep brain stimulation (DBS) of the subthalamic nucleus (STN) is a well-established symptomatic treatment for Parkinson’s diseases (PD). However, knowledge on local electrophysiological biomarkers within the STN and their cortical connectivity profile is still scarce. Such information would be necessary for optimal positioning of the DBS leads based on PD network pathophysiology. This talk describes the introduction and exploration of a novel technique for electrophysiological measurements during DBS surgery. Combined electroencephalography (EEG) with stepwise local field potentials recordings during insertion of the DBS lead was performed intraoperatively, thereby, allowing to capture local STN and cortico-subthalamic physiology with high speactral and spatial specificity. Our results revealed that strong beta oscillatory activity in the STN was located more dorsally than the STN-ipsilateral motor network phase coupling; the respective frequency bands were in the low and high beta-band, respectively. Moreover, the spot within the STN, where this STN-cortical phase coupling occurred, correlated highly with the STN spot where the phase of beta oscillations modulated the amplitude of high-frequency oscillations. This STN location was furthermore, characterized by information flowed from the ipsilateral motor cortex to the STN in the high beta-band suggesting a pathologically synchronized network with a direct STN-motor cortex connection via the hyperdirect pathway. Interestingly, the very same STN spot showed a resonance like responses to electrical stimulation suggesting a decoupling of pathologically synchronized STN-motor cortex connectivity during therapeutic DBS. In conclusion, we provide first evidence that macroelectrode recordings with the chronic electrode concurrent with EEG recordings are a reliable method for STN localization during DBS surgery. Additionally, combining LFP and EEG recordings during mapping of STN offered a new way of DBS targeting on the basis of pathological local biomarkers and network activity.
http://razodia.com/
March 16th – Tara Keck (UCL)
SM3, Maths
Interactions between different Hebbian and homeostatic plasticity
Homeostatic plasticity is thought to be mediated by mechanistic changes, such as synaptic scaling and shifts in the excitation/inhibition balance. These mechanisms are thought to be distinct from the Bienenstock, Cooper, Munro (BCM) learning rule, where the threshold for the induction of long-term potentiation and long-term depression slides in response to changes in activity levels; however, both mechanisms produce a homeostatic response of a relative increase (or decrease) in strength of excitatory synapses in response to overall activity-level changes. I will present a model that suggests how these two mechanisms may interact to facilitate firing-rate homeostasis, while maintaining functional properties of neurons.
http://www.kecklab.com/
March 23rd – Petroula Laiou (University of Exeter)
E204, Chemistry
Evaluation of epilepsy surgery using mathematical modelling
Surgery is one option for treating pharmacoresistant epilepsy patients. In this treatment, brain tissue is resected to render a patient free from seizures. However, epilepsy surgery is often ineffective. A basic reason for this is our insufficient understanding of the generation of seizures in the brain network. Mathematical modelling can be used to study normal and abnormal dynamical states of the brain. Moreover, it can elucidate the effects of brain surgery by quantifying the contribution of different brain regions (nodes in the brain network) to the generation of seizures. In this talk, I will present a mathematical framework which aims to identify the optimal set of nodes that has to be removed from a brain network in order to obtain seizure-free patients. The ultimate goal of this approach is to provide a support tool for epilepsy surgery to help in the presurgical evaluation.
http://emps.exeter.ac.uk/mathematics/staff/pl380
Xenopus hatchling tadpoles “struggle” when held at their tail, rhythmically bending their body in an attempt to free themselves. During struggling, neurons on opposite sides of the tadpole spinal cord fire bursts of spikes in alternation, resulting in motor activity and the rhythmic bends of the whole body. A central question of interest in tadpole struggling is the role of the excitatory and inhibitory coupling for the rhythm generation. In this talk we will therefore study a reduced biophysical model of tadpole struggling that contains weakly excitatory neurons and strongly inhibitory neurons, where the latter are also subject to short-term synaptic depression. We will discuss approaches to understanding the dynamics of such a network, focusing on reducing the complex model equations to a more tractable, and intuitive, quantity. For the reduction we will employ two useful mathematical techniques: geometric singular perturbation theory and phase response curves.