Logo Logo
Hilfe
Kontakt
Switch language to English
Characterization of response properties and connectivity in mouse visual thalamus and cortex
Characterization of response properties and connectivity in mouse visual thalamus and cortex
How neuronal activity is shaped by circuit connectivity between neuronal populations is a central question in visual neuroscience. Combined with experimental data, computational models allow causal investigation and prediction of both how connectivity influences activity and how activity constrains connectivity. In order to develop and refine these computational models of the visual system, thorough characterization of neuronal response patterns is required. In this thesis, I first present an approach to infer connectivity from in vivo stimulus responses in mouse visual cortex, revealing underlying principles of connectivity between excitatory and inhibitory neurons. Second, I investigate suppressed-by-contrast neurons, which, while known since the 1960s, still remain to be included in standard models of visual function. I present a characterization of intrinsic firing properties and stimulus responses that expands the knowledge about this obscure neuron type. Inferring the neuronal connectome from neural activity is a major objective of computational connectomics. Complementary to direct experimental investigation of connectivity, inference approaches combine simultaneous activity data of individual neurons with methods ranging from statistical considerations of similarity to large-scale simulations of neuronal networks. However, due to the mathematically ill-defined nature of inferring connectivity from in vivo activity, most approaches have to constrain the inference procedure using experimental findings that are not part of the neural activity data set at hand. Combining the stabilized-supralinear network model with response data from the visual thalamus and cortex of mice, my collaborators and I have found a way to infer connectivity from in vivo data alone. Leveraging a property of neural responses known as contrast-invariance of orientation tuning, our inference approach reveals a consistent order of connection strengths between cortical neuron populations as well as tuning differences between thalamic inputs and cortex. Throughout the history of visual neuroscience, neurons that respond to a visual stimulus with an increase in firing have been at the center of attention. A different response type that decreases its activity in response to visual stimuli, however, has been only sparsely investigated. Consequently, these suppressed-by-contrast neurons, while recently receiving renewed attention from researchers, have not been characterized in depth. Together with my collaborators, I have conducted a survey of SbC properties covering firing reliability, cortical location, and tuning to stimulus orientation. We find SbC neurons to fire less regularly than expected, be located in the lower parts of cortex, and show significant tuning to oriented gratings.
visual neuroscience, computational connectomics
Renner, Simon
2022
Englisch
Universitätsbibliothek der Ludwig-Maximilians-Universität München
Renner, Simon (2022): Characterization of response properties and connectivity in mouse visual thalamus and cortex. Dissertation, LMU München: Graduate School of Systemic Neurosciences (GSN)
[thumbnail of Renner_Simon.pdf]
Vorschau
PDF
Renner_Simon.pdf

9MB

Abstract

How neuronal activity is shaped by circuit connectivity between neuronal populations is a central question in visual neuroscience. Combined with experimental data, computational models allow causal investigation and prediction of both how connectivity influences activity and how activity constrains connectivity. In order to develop and refine these computational models of the visual system, thorough characterization of neuronal response patterns is required. In this thesis, I first present an approach to infer connectivity from in vivo stimulus responses in mouse visual cortex, revealing underlying principles of connectivity between excitatory and inhibitory neurons. Second, I investigate suppressed-by-contrast neurons, which, while known since the 1960s, still remain to be included in standard models of visual function. I present a characterization of intrinsic firing properties and stimulus responses that expands the knowledge about this obscure neuron type. Inferring the neuronal connectome from neural activity is a major objective of computational connectomics. Complementary to direct experimental investigation of connectivity, inference approaches combine simultaneous activity data of individual neurons with methods ranging from statistical considerations of similarity to large-scale simulations of neuronal networks. However, due to the mathematically ill-defined nature of inferring connectivity from in vivo activity, most approaches have to constrain the inference procedure using experimental findings that are not part of the neural activity data set at hand. Combining the stabilized-supralinear network model with response data from the visual thalamus and cortex of mice, my collaborators and I have found a way to infer connectivity from in vivo data alone. Leveraging a property of neural responses known as contrast-invariance of orientation tuning, our inference approach reveals a consistent order of connection strengths between cortical neuron populations as well as tuning differences between thalamic inputs and cortex. Throughout the history of visual neuroscience, neurons that respond to a visual stimulus with an increase in firing have been at the center of attention. A different response type that decreases its activity in response to visual stimuli, however, has been only sparsely investigated. Consequently, these suppressed-by-contrast neurons, while recently receiving renewed attention from researchers, have not been characterized in depth. Together with my collaborators, I have conducted a survey of SbC properties covering firing reliability, cortical location, and tuning to stimulus orientation. We find SbC neurons to fire less regularly than expected, be located in the lower parts of cortex, and show significant tuning to oriented gratings.