January 27th – Robert Schmidt (University of Sheffield)
F40, Biomedical Sciences Building
Sensorimotor processing in the basal ganglia leads to transient beta oscillations during behaviour
The basal ganglia have been implicated in motor control including the initiation and suppression of actions. In Parkinson’s disease neural processing in the basal ganglia is impaired, and patients suffer from severe motor symptoms. Signatures of the impaired neural processing include oscillations in the beta band (~13-30 Hz), but beta oscillations have actually also been found in the basal ganglia of healthy animals and may reflect the utilisation of sensory cues for behaviour. To see whether healthy and pathological beta oscillations share neural mechanisms we combined computational models for beta oscillations in Parkinson’s disease with input patterns derived from single unit recordings in healthy rats. We found that movement-related increases in striatal activity lead to transient beta oscillations in the model with a time course that closely matches the experimentally measured oscillations. In addition, our model can account for further unintuitive aspects of beta modulation including beta phase resets following sensory cues and correlations with reaction time. Overall, our model can explain how sensorimotor neural activity in the basal ganglia leads to transient beta oscillations during behaviour.
https://www.shef.ac.uk/psychology/staff/academic/robert-schmidt
http://www.schmidt-lab.net/
February 10th – Rosalyn Moran (University of Bristol)
SM4, Maths
Neural Dynamics in Dynamic Causal Modeling
Dynamic Causal Modeling or DCM for electrophysiological responses uses neural mass models and mean field approaches to mimic the population activity observed in local field potentials, and their non-invasive source localised EEG and MEG correlates. In this talk I will describe the generative models underlying DCM for M/EEG and describe the inversion procedures which allows DCMs to be fit to subject specific data. I will outline recent validation studies in autoimmune NMDA encephalitis. Specifically I will demonstrate how DCM’s connectivity estimates, which are based explicitly on ion-channel dynamics, can be recovered to show particular deficits in NMDA-mediated connectivity in these interesting patients. I will finally show some examples of DCM applied to understanding neural hierarchical processing changes with age.
http://www.bristol.ac.uk/engineering/people/rosalyn-j-moran/index.html
February 17th – Aleks Domanski (University of Bristol)
SM4, Maths
Decoding Hippocampal-Frontal Cortical cell assemblies during spatial working memory
Functionally specialised brain regions such as the hippocampus (HP) and medial prefrontal cortex (mPFC) communicate under cognitive demand (such as during spatial working memory) to share information required for the successful completion of a behavioural task. The circuit mechanisms underlying this process remain poorly resolved, however population coding by “cell assemblies” – transiently synchronised groups of neurons – is favoured as a biologically-plausible candidate.
Here, using rat neurophysiology I will introduce a novel method for detecting cell assemblies from parallel single-unit recordings and present evidence that the activity of inter-structural cell assemblies provides superior and broader information encoding of task-related variables compared to individual single units. Moreover, cell assembly activity is stable in rest periods flanking the behavioural task in expert-performing rats. Together, our work argues that HP-mPFC cell assemblies provide a robust channel binding the specialised information encoded by each brain area.
http://research-information.bristol.ac.uk/en/persons/aleks-p-f-domanski(4dd1f5f5-8b94-4448-b5e5-4692cbc15794).html
February 24th – Tony Pickering (University of Bristol)
SM3, Maths
Noradrenaline / Norepinephrine : Facts / Alternate Facts
It is known that noradrenaline is an important neuromodulator in the brain; implicated in processes ranging from alerting, arousal and salience detection through novelty learning and emotional recall and including sensory discrimination, motor control and pain regulation. Much of this appears to be attributable to the actions of a small cluster of neurones in the pons known as the locus coeruleus (LC). We have been focussing on the role of the LC in the regulation of pain and have developed some novel tools and approaches to interventionally assess its function both in rodents and in man. In so doing we have stumbled into the often apparently contradictory world of LC neurobiology and Nor-adrenaline/epinephrine (the clue is in the name(s)). High profile papers are currently being published on these topics. I will present the facts / the alternate facts (and their cognate models) and you can be the experts / members of the press who can pull me up on their interpretation. It’s gonna be so great. (byo wall).
The contrasting views include:
- LC is a primitive relic / important determinant of complex behaviour in vertebrates.
- Acts via Noradrenaline / Dopamine release
- Each neuron projects to the whole brain / to a discrete territory.
- Volume transmission vs focussed synaptic release.
- Signalling saliency to the cortex / preventing salient events from reaching the cortex.
- Whether a modular organising principle can help account for these apparently opposing characteristics.
http://www.bris.ac.uk/phys-pharm/people/tony-e-pickering/index.html
March 17th – Nina Kazanina (University of Bristol)
SM3, Maths
http://www.bristol.ac.uk/expsych/people/nina-kazanina/index.html
March 23rd – Krasimira Tsaneva Atanasova (University of Exeter)
SM3, Maths
Socio-motor biomarkers in schizophrenia
In an effort to establish reliable indicators of schizophrenia we have developed a test that could detect deficits in movement and social interactions, both characteristics of the disorder. We asked people to perform movements alone, and to mirror the movements of a computer avatar or a humanoid robot. Using statistical learning techniques we were able to distinguish people with schizophrenia from healthy participants with accuracy and specificity slightly better than clinical interviews and comparable to tests based on much more expensive neuroimaging methods. This methodology could help with diagnosis of schizophrenia and to monitor patients’ responses to treatment, but needs to be tested further before being potentially widely applied in clincal practice.
http://emps.exeter.ac.uk/mathematics/staff/kt298
March 31st – Julijana Gjorgjieva (The Max Planck Institute for Brain Research)
SM3, Maths
Optimal coding in sensory populations
At the output of the retina, some ganglion cell classes consist of paired cell types that have very similar spatiotemporal response properties, except that one type is excited by light increments (ON), the other type by light decrements (OFF). Besides the retina, the ON-OFF dichotomy is also observed in other sensory modalities, which include invertebrate motion vision, thermosensation, chemosensation, audition, and electrolocaton in electric fish. The commonality in the splitting of sensory signals across neural systems and species suggests a strong evolutionary pressure driving the emergence of opposite response types to positive and negative stimuli. Why have the ON and OFF channels dierged in different species and sensory modalities? I will discuss how ON-OFF organization at the level of sensory neuronal populations emerges from an efficient coding strategy to maximize information transfer of incoming stimuli. Our work underscores the significance of external stimulus statistics and neural noise in governing how populations organize to efficiently encode and decode sensory information.
http://brain.mpg.de/research/computation-in-neural-circuits-group.html
April 7th – Tom Jahans-Price (University of Oxford)
SM3, Maths
Glutamatergic dysfunction leads to a hyper-dopaminergic phenotype: a possible cause of aberrant salience
Current thinking suggests that psychosis is a disorder of aberrant salience. This describes when a stimulus continues to grab inappropriately high levels of attention, and it is thought to be mediated via elevated dopamine (DA) levels, which have been robustly demonstrated in schizophrenia. However, the causes of this DA dysregulation are generally unspecified. Recent large scale GWAS meta-analyses have established genome-wide significant association to schizophrenia for the Gria1 locus which codes for the GluA1 subunit of the AMPA glutamate receptor. GluA1 KO mice have previously been studied in relation to schizophrenia but, notably, striatal whole tissue levels of dopamine and its metabolites appear normal in these animals. However, we might not expect to see changes in dopamine activity in anaesthetised animals, or in a home-cage environment. Indeed, changes in phasic DA responses are likely to be both behaviour-dependent and stimulus-specific.
To test this possibility we have recorded phasic DA signals with high temporal resolution, in freely moving, behaving wild-type and GluA1 KO mice, using fast-scan cyclic voltammetry (FSCV). I will present data showing that phasic dopamine signals in response to neutral light stimuli fail to habituate in Gria1-/- mice, resulting in a behaviourally relevant, hyper-dopaminergic phenotype in these animals. This parallels previous behavioural data from these mice. In addition, phasic dopamine responses to unsignalled rewards were also significantly enhanced in the knockout mice. This provides evidence for behaviourally-relevant hyper-dopaminergic responses in a genetically modified mouse model of glutamatergic dysfunction relevant to schizophrenia.
April 28th – Cian O’Donnell (University of Bristol)
SM3, Maths
Memory processing by molecular signals in neurons
The dominant models of long-term memory formation are based on Hebbian plasticity rules. These rules are usually given solely in terms of electrical spiking activity of pre- and post-synaptic neurons. However, in real cells, expression and consolidation of synaptic plasticity involves a complicated cascade of biochemical signals, protein synthesis, and gene transcription. Surprisingly little is understood about the computational function of these molecular signals for memory. In this talk I will present our recent work on how spatial patterning of protein synthesis in dendrites allows brains to choose what to remember and what to forget. Based on these results, we also propose a novel model for memory generalisation during sleep.
http://www.bris.ac.uk/engineering/people/cian-odonnell/index.html
May 5th – Angela Roberts (University of Cambridge)
SM3, Maths
A multidimensional approach to the study of positive and negative emotion regulation in the marmoset prefrontal cortex
Dysregulated emotions are a core feature of many neuropsychiatric disorders involving altered activity in limbic emotional circuitry that includes the amygdala, hippocampus and prefrontal cortex (PFC). Front line treatments include drugs that target the serotonin system and more recently the glutamate system, but how they work, and in which patient, is poorly understood. Moreover, at least 40% of patients are resistant to such treatments emphasizing the need to understand better the neural circuits and cognitive and behavioural processes that underlie the regulation of positive and negative emotion. The prefrontal cortex is at its most highly developed in primates and so to further our understanding of its regulation of amygdala-dependent emotional learning and expression we have developed models of negative and positive emotional learning and expression in a new world primate, the common marmoset. Since emotional states are composed of both physiological and behavioural components we use an automated telemetry system to allow the simultaneous measurement of behavioural and cardiovascular emotional responses e.g. heart rate and blood pressure, in freely moving marmosets. We employ three main experimental strategies. The first, to determine the effects of localized prefrontal manipulations on emotional states, their impact on activity in downstream targets using fluorodeoxyglucose microPET and their sensitivity to drugs targeting the glutamate and serotonin system. Second, we have initiated a neuroimaging program to characterize the development of prefrontal circuits across childhood and adolescence in the marmoset since the majority of anxiety and mood disorders have their onset during these critical periods of development in humans. The third, to study the impact of known behavioural and genetic risk factors for mood and anxiety disorders, i.e. trait anxiety and a polymorphism in the upstream promotor region of the serotonin transporter gene, on these prefrontal circuits using microPET, structural mri, microdialysis and post mortem mRNA analysis. In this presentation I will illustrate these approaches in the study of anxiety and anhedonia.
Recent review: Shiba et al (2016) Frontiers in Systems Neuroscience 10:12.
http://www.neuroscience.cam.ac.uk/directory/profile.php?acroberts
