Olivier Marre

Motion tracking is a challenge the visual system has to solve by reading out the retinal population. It is still unclear how the information from different neurons can be combined together to estimate the position of an object. Here we recorded a large population of ganglion cells in a dense patch of salamander and guinea pig retinas while displaying a bar moving diffusively. We show that the bar's position can be reconstructed from retinal activity with a precision in the hyperacuity regime using a linear decoder acting on 100+ cells. We then took advantage of this unprecedented precision to explore the spatial structure of the retina's population code. The classical view would have suggested that the firing rates of the cells form a moving hill of activity tracking the bar's position. Instead, we found that most ganglion cells in the salamander fired sparsely and idiosyncratically, so that their neural image did not track the bar. Furthermore, ganglion cell activity sp...

Recent experimental results based on multielectrode and imaging techniques have reinvigorated the idea that large neural networks operate near a critical point, between order and disorder. However, evidence for criticality has relied on the definition of arbitrary order parameters, or on models that do not address the dynamical nature of network activity. Here we introduce a novel approach to assess criticality that overcomes these limitations, while encompassing and generalizing previous criteria. We find a simple model to describe the global activity of large populations of ganglion cells in the rat retina, and show that their statistics are poised near a critical point. Taking into account the temporal dynamics of the activity greatly enhances the evidence for criticality, revealing it where previous methods would not. The approach is general and could be used in other biological networks.

In vivo intracellular electrophysiology offers the unique possibility of listening to the "synaptic rumor " of the cortical network, captured by a recording electrode in a single V1 cell. It allows one to reconstruct the distribution of input sources in space and time, i.e. the effective network dynamics. We have used a reverse engineering method to demonstrate the propagation of visually evoked activity through lateral (and feedback) connectivity in the primary cortex of higher mammals. This approach, based on synaptic echography, is compared here with a real-time brain imaging technique based on voltage-sensitive dye imaging. The former method gives access to the microscopic convergence processes of single neurons, whereas the latter describes the macroscopic divergence process on the neuronal map. A combination of the two techniques can be used to elucidate the cortical origin of low-level (non attentive) binding processes participating in the emergence of Gestalt perce...

Neurons in the primary visual cortex receive subliminal information originating from the periphery of their receptive fields (RF) through a variety of cortical connections. In the cat primary visual cortex, long-range horizontal axons have been reported to preferentially bind to distant columns of similar orientation preferences, whereas feedback connections from higher visual areas provide a more diverse functional input. To understand the role of these lateral interactions, it is crucial to characterize their effective functional connectivity and tuning properties. However, the overall functional impact of cortical lateral connections, whatever their anatomical origin, is unknown since it has never been directly characterized. Using direct measurements of postsynaptic integration in cat areas 17 and 18, we performed multi-scale assessments of the functional impact of visually driven lateral networks. Voltage-sensitive dye imaging showed that local oriented stimuli evoke an orienta...

Retinal prostheses aim at restoring vision in patients blind from photoreceptor degeneration by electrically stimulating the residual retinal tissue. Currently, the most efficient implants are either inserted in the subretinal space or on the vitreal side of the retina (epi-retinal). Although the residual tissue can partly degenerate, it was shown that acute stimulation of residual neurones can induce visual percepts. Recently, a clinical trial with the epiretinal Argus2 device (60 electrodes) from the company 2nd Sight enabled most patients to orient and find light targets, some even reading words. This device has received a CE mark. Surprisingly, when the subretinal implant from the company Retina Implant AG displaying many more electrodes (1500 electrodes) was evaluated in clinical trials, the patient visual performances were fairly similar. The restored visual performances of the patients demonstrate that blind patients can recover some visual function when their residual retina is properly stimulated. However, the resolution is not yet sufficient to perform complex tasks such as autonomous locomotion, face identification or text reading. Several challenges remain to generate an increase in pixel density corresponding to the increase in electrode number and density. These challenges include the stimulation modality, the tissue/implant interface design, the electrode materials, and the visual information encoder. This review will discuss these great challenges after introducing the major clinical results.

Retinal prostheses aim at restoring vision in patients blind from photoreceptor degeneration by electrically stimulating the residual retinal tissue. Currently, the most efficient implants are either inserted in the subretinal space or on the vitreal side of the retina (epi-retinal). Although the residual tissue can partly degenerate, it was shown that acute stimulation of residual neurones can induce visual percepts. Recently, a clinical trial with the epiretinal Argus2 device (60 electrodes) from the company 2nd Sight enabled most patients to orient and find light targets, some even reading words. This device has received a CE mark. Surprisingly, when the subretinal implant from the company Retina Implant AG displaying many more electrodes (1500 electrodes) was evaluated in clinical trials, the patient visual performances were fairly similar. The restored visual performances of the patients demonstrate that blind patients can recover some visual function when their residual retina is properly stimulated. However, the resolution is not yet sufficient to perform complex tasks such as autonomous locomotion, face identification or text reading. Several challenges remain to generate an increase in pixel density corresponding to the increase in electrode number and density. These challenges include the stimulation modality, the tissue/implant interface design, the electrode materials, and the visual information encoder. This review will discuss these great challenges after introducing the major clinical results.

ABSTRACT The number of possible activity patterns in a population of neurons grows exponentially with the size of the population. Typical experiments explore only a tiny fraction of the large space of possible activity patterns in the case of populations with more than 10 or 20 neurons. It is thus impossible, in this undersampled regime, to estimate the probabilities with which most of the activity patterns occur. As a result, the corresponding entropy—which is a measure of the computational power of the neural population—cannot be estimated directly. We propose a simple scheme for estimating the entropy in the undersampled regime, which bounds its value from both below and above. The lower bound is the usual 'naive' entropy of the experimental frequencies. The upper bound results from a hybrid approximation of the entropy which makes use of the naive estimate, a maximum entropy fit, and a coverage adjustment. We apply our simple scheme to artificial data, in order to check their accuracy; we also compare its performance to those of several previously defined entropy estimators. We then apply it to actual measurements of neural activity in populations with up to 100 cells. Finally, we discuss the similarities and differences between the proposed simple estimation scheme and various earlier methods.

ABSTRACT Purpose This aims to develop and improve the use of retinal prosthesis in animal model For that purpose we measured retinal and cortical response to direct subretinal electrical stimulation to understand how the patterns of stimuli can be adapted to improve stimulation to get closer to the response evoked by natural visual stimuli Methods In a rat model, a comparative analysis of the functional impact of similar stimulation of a subretinal implant is done at two levels (1)in the retina in vitro by multi electrodes array (2)in vivo in the primary visual cortex by optical imaging recordings Optical imaging permits a functional mapping of the cortex, using light reflection and absorption changes depending on the rate of blood oxygenation Diverse parameters were investigated: stimulus shape and polarity,intensity,size and location Results At the cortical level we have quantified the size, position and intensity of the point-spread function in response to the various electrical stimulations and compared them to those generated by calibrated light stimuli. The point-spread function was much larger for electrical stimulations compared to visual stimulation We have performed retinal recordings of ganglion cells, while the prosthesis is on the photoreceptor side The cortical response and the recruitment of ganglion cells respond according to a logarithmic equation The ability to evaluate in vitro within the retina and in vivo the cortical responses induced by the prosthesis allowed us refining the patterns of electrical stimulation to get closer to a natural activation Conclusion These results offer interesting prospect for improving the design of prostheses as well as their patterns of stimulation for a medical application