Workshop description
Neural populations continuously generate spatiotemporal activity patterns at multiple scales. The study of how different brain states emerge and impact function is key to understand the organizing principles of brain activity. Although brain state changes were historically attributed to sleep, akin phenomena were recently observed during wakefulness, allowing the study of their impact on behavior. Moreover, current recording techniques allow to track the neural activity at unprecedented spatial and temporal resolutions going from the micro-circuit to brain-wide activity and from milliseconds to months.
During the workshop, we will discuss current theoretical frameworks to study the emergence of different states in neural networks and their impact on information processing. Furthermore, we will discuss the experimental findings relating different brain states to animal behavior during complex cognitive tasks.
Registration:
www.cnsorg.org/cns-2021
During the workshop, we will discuss current theoretical frameworks to study the emergence of different states in neural networks and their impact on information processing. Furthermore, we will discuss the experimental findings relating different brain states to animal behavior during complex cognitive tasks.
Registration:
www.cnsorg.org/cns-2021
SpeakersTatiana Engel Cold Spring Harbor Laboratory, US
Anna Levina University of Tübingen, Germany
Enzo Tagliazucchi Buenos Aires University, Argentina
Valentin Dragoi University of Texas, US
|
|
Schedule July 6th 2021 (Paris Time)
15:00-15:05 Opening
15:05:15:45 Tatiana Engel A mesoscale connectome defines spatiotemporal dynamics of neural activity across the mouse cortex.
15:45-16:25 Anna Levina Local brain states and their specialization captured by autocorrelation timescales.
16:25-16:30 Break
16:30-17:10 Enzo Tagliazucchi Perturbational in silico approaches to understand states of consciousness.
17:10-17:50 Valentin Dragoi Synchrony in visual cortical populations during wakefulness: the good and the bad.
17:50-18:15 General discussion
ABSTRACTS
A mesoscale connectome defines spatiotemporal dynamics of neural activity across the mouse cortex
Tatiana Engel
Recently available large-scale recordings reveal the rich heterogeneity of neural dynamics on the brain-wide scale. These global neural dynamics define internal states of the brain, which influence behavior with spatial and temporal specificity. Yet, how global activity arises from anatomical connectivity to drive animal behaviors is unknown. We developed a computational framework for modeling global neural dynamics, which utilizes connectivity and predicts rich behavioral outputs on single trials. By integrating large-scale datasets of neural activity and connectivity, we constructed a model of mesoscopic functional dynamics across the mouse cortex. We found that global activity is restricted to a low-dimensional subspace spanned by a few cortical areas, exploring different parts of this subspace in different behavioral contexts. Rich patterns of animal movements can be decoded from the activity generated by our low-dimensional model. The model generalizes across individual animals and behaviors, indicating that the connectome fundamentally defines the cortex-wide neural dynamics on single trials.
Local brain states and their specialization captured by autocorrelation timescales.
Anna Levina
Cortical neurons process information on multiple timescales related to the dynamics of intrinsic activity fluctuations. While areal differences in intrinsic timescales reflect functional specialization of cortical areas, it is unknown whether the timescales can vary within a single area and adjust rapidly and selectively to the demands of a cognitive task. Here I use an example of local spiking activity within columns of monkey area V4 to show how timescales of local activity can be precisely determined. I demonstrate that the ongoing spiking activity unfolds across at least two distinct timescales - fast and slow - and the slow timescale increases when monkeys attended to the location of the receptive field. From the theoretical point of view, long timescales allow for better integration of temporally extended signals. However, it can also lead to the increased sensitivity to noise. I discuss when long timescales can be beneficial or detrimental for neuronal computation.
Perturbational in silico approaches to understand states of consciousness.
Enzo Tagliazucchi
Consciousness transiently fades away during deep sleep, more stably under anesthesia, and sometimes permanently due to brain injury. The development of an index to quantify the level of consciousness across these states is regarded as a key problem both in basic and clinical neuroscience. Here we argue that this problem is ill-defined, since such index would not exhaust all the relevant information about a given state of consciousness. While the level of consciousness can be taken to describe the actual state, a complete characterization should also include the potential behavior against external perturbations. We developed and analyzed whole-brain computational models to show that the stability of conscious states provided information complementary to their similarity to conscious wakefulness. Our work leads to a new methodological framework to sort out different states by their stability and reversibility, showing its usefulness to dissociate between physiological (sleep), pathological (patients), and pharmacologically-induced (anesthesia) states of consciousness.
Synchrony in visual cortical populations during wakefulness: the good and the bad
Valentin Dragoi
Brain activity during wakefulness is characterized by fluctuations in neuronal responses at a variety of time scales. Whether these fluctuations play any role in modulating neural coding and the accuracy of behavioral responses is poorly understood. Using multiple-electrode recording techniques, we show that slow changes in population synchrony in monkey visual cortex impair the coding of sensory information and perceptual performance. These changes also occur in executive areas, such as prefrontal cortex, while monkeys freely explore their environment during foraging. However, while population synchrony is detrimental for encoding sensory information at longer time scales, it plays a beneficial role at shorter time scales. Indeed, by simultaneously recording visual cortical populations in multiple areas we discovered that the precise temporal coordination between the spikes of three of more neurons carries information about perceptual reports in the absence of firing rate modulation. These results demonstrate the differential impact of population synchrony on sensory coding and perception depending on the time scale at which it manifests.
15:05:15:45 Tatiana Engel A mesoscale connectome defines spatiotemporal dynamics of neural activity across the mouse cortex.
15:45-16:25 Anna Levina Local brain states and their specialization captured by autocorrelation timescales.
16:25-16:30 Break
16:30-17:10 Enzo Tagliazucchi Perturbational in silico approaches to understand states of consciousness.
17:10-17:50 Valentin Dragoi Synchrony in visual cortical populations during wakefulness: the good and the bad.
17:50-18:15 General discussion
ABSTRACTS
A mesoscale connectome defines spatiotemporal dynamics of neural activity across the mouse cortex
Tatiana Engel
Recently available large-scale recordings reveal the rich heterogeneity of neural dynamics on the brain-wide scale. These global neural dynamics define internal states of the brain, which influence behavior with spatial and temporal specificity. Yet, how global activity arises from anatomical connectivity to drive animal behaviors is unknown. We developed a computational framework for modeling global neural dynamics, which utilizes connectivity and predicts rich behavioral outputs on single trials. By integrating large-scale datasets of neural activity and connectivity, we constructed a model of mesoscopic functional dynamics across the mouse cortex. We found that global activity is restricted to a low-dimensional subspace spanned by a few cortical areas, exploring different parts of this subspace in different behavioral contexts. Rich patterns of animal movements can be decoded from the activity generated by our low-dimensional model. The model generalizes across individual animals and behaviors, indicating that the connectome fundamentally defines the cortex-wide neural dynamics on single trials.
Local brain states and their specialization captured by autocorrelation timescales.
Anna Levina
Cortical neurons process information on multiple timescales related to the dynamics of intrinsic activity fluctuations. While areal differences in intrinsic timescales reflect functional specialization of cortical areas, it is unknown whether the timescales can vary within a single area and adjust rapidly and selectively to the demands of a cognitive task. Here I use an example of local spiking activity within columns of monkey area V4 to show how timescales of local activity can be precisely determined. I demonstrate that the ongoing spiking activity unfolds across at least two distinct timescales - fast and slow - and the slow timescale increases when monkeys attended to the location of the receptive field. From the theoretical point of view, long timescales allow for better integration of temporally extended signals. However, it can also lead to the increased sensitivity to noise. I discuss when long timescales can be beneficial or detrimental for neuronal computation.
Perturbational in silico approaches to understand states of consciousness.
Enzo Tagliazucchi
Consciousness transiently fades away during deep sleep, more stably under anesthesia, and sometimes permanently due to brain injury. The development of an index to quantify the level of consciousness across these states is regarded as a key problem both in basic and clinical neuroscience. Here we argue that this problem is ill-defined, since such index would not exhaust all the relevant information about a given state of consciousness. While the level of consciousness can be taken to describe the actual state, a complete characterization should also include the potential behavior against external perturbations. We developed and analyzed whole-brain computational models to show that the stability of conscious states provided information complementary to their similarity to conscious wakefulness. Our work leads to a new methodological framework to sort out different states by their stability and reversibility, showing its usefulness to dissociate between physiological (sleep), pathological (patients), and pharmacologically-induced (anesthesia) states of consciousness.
Synchrony in visual cortical populations during wakefulness: the good and the bad
Valentin Dragoi
Brain activity during wakefulness is characterized by fluctuations in neuronal responses at a variety of time scales. Whether these fluctuations play any role in modulating neural coding and the accuracy of behavioral responses is poorly understood. Using multiple-electrode recording techniques, we show that slow changes in population synchrony in monkey visual cortex impair the coding of sensory information and perceptual performance. These changes also occur in executive areas, such as prefrontal cortex, while monkeys freely explore their environment during foraging. However, while population synchrony is detrimental for encoding sensory information at longer time scales, it plays a beneficial role at shorter time scales. Indeed, by simultaneously recording visual cortical populations in multiple areas we discovered that the precise temporal coordination between the spikes of three of more neurons carries information about perceptual reports in the absence of firing rate modulation. These results demonstrate the differential impact of population synchrony on sensory coding and perception depending on the time scale at which it manifests.
Organizers |
Gabriela Mochol
Centre de Recerca Matemàtica Computational Neuroscience Unit Universitat Autònoma de Barcelona, Spain [email protected] Adrián Ponce-Alvarez Center for Brain and Cognition Computational Neuroscience Group Pompeu Fabra University Barcelona, Spain [email protected] |
