Summary of the project. This project aims to better understand the retina’s response to complex visual stimuli and unravel the role played by the lateral inter-neurons network (amacrine cells) in this response. For this, we will adopt an experimental methodology using real-time control and feedback loop to adapt, in real-time, visual stimulations to recorded retinal cell responses. The experiments will be done in Valparaiso (Chile). The Biovision team will develop the control software and extrapolate results at Inria Sophia-Antipolis based on its recent theoretical advances in retina modelling. This is, therefore, a transdisciplinary and international project at the interface between biology, computer sciences and mathematical neurosciences. Beyond a better understanding of its network structure’s role in shaping the retina’s response to complex spatiotemporal stimuli, this project could potentially impact methods for diagnosing neurodegenerative diseases like Alzheimer’s.
Project objective. In collaboration with the experimental research team of A. Palacios (Universidad de Valparaíso), we want to develop a protocol utilising a feedback stimulation-control loop to precisely stimulate retinal ganglion cells (RGCs) in a determined region of space, intersecting its receptive field, with a stimulus varying in time. This will enable us to scan the space-time scales of this response and identify potential resonances corresponding to preferred time or space frequencies. These resonances can result either from the intrinsic properties of the cell or from the fact that they are embedded in a network essentially made of a feed-forward structure (from photoreceptors to RGC via bipolar cells) connected laterally via inter-neurons: horizontal and amacrine cells.
Our mathematical analysis allows us to disentangle these two origins and quantify the effects induced by the interneurons network structure [Kartsaki et al, 2023], particularly the role of lateral amacrine cells connectivity. As we have shown in previous theoretical work, the combination of resonances induced by the conjunction of the retinal network dynamics and moving stimuli with the proper size and speed can induce a retinal wave of activity ahead of the stimulus, helping the brain not only to decipher but also anticipate its trajectory, a fundamental feature of the visual system |Berry & al, 1999; Souihel & Cessac, 2019; Benvenutti et al, 2020]. There is, therefore, a close relation between the amacrine network structure, the presence of space-time resonances, and the dynamics of anticipatory waves. Although these anticipation effects have been reported experimentally [Menz et al, 2020], a systematic experimental study of this aspect, guided by theoretical results, is still missing. To fill this gap, we want to answer the following questions in a combination between experiments, modelling, and online data processing
- Using the feedback stimulation-control loop evoked above and discussed in more detail below, do
we observe space-time resonances as expected from the theoretical analysis? - A theoretical consequence of such resonances is that preferred speeds exist for moving objects where the anticipation effect is maximum. Can we confirm this experimentally and explain these preferred speeds from the observed resonances?
- What do we learn about the retinal structure and dynamics from observing these resonances?
- From this knowledge, can we elaborate more complex, optimal, spatio-temporal stimuli that would allow us to extract the maximal amount of information about the retinal network in a fast, online process based on the feedback stimulation-control loop?
Despite the importance of understanding the dynamics of the retinal network in response to spatio-temporal stimuli, this area of research has not been systematically addressed. This is mainly due to the lack of a theoretical framework within the neurobiology community to anticipate the dynamic effect of the retinal network. Although some teams are using feedback stimulation-control loop methods, such as the A. Palacios’ lab, the current algorithms driving the control process do not incorporate network effects. Our interdisciplinary project aims to address this gap by proposing a new data processing approach to reveal retinal network features previously hidden and essential aspects in response to spatio-temporal stimuli [Cessac et al, 2019].
Description of the expected impact
Our project will impact fundamental research in neuroscience with a better understanding of retinal network processing, data acquisition, and visual information coding. New algorithms inspired by the retinal network’s dynamic can also potentially enrich the field of computer vision. Importantly, this proposal is a milestone of a more ambitious project. In the long term, a significant outcome would be the characterization of retinal ganglion cells in healthy retinas and genetically modified mice exhibiting symptoms of Alzheimer’s disease. In these mice, as shown by A. Palacios lab, the accumulation of beta- amyloid in the retina leads to a sporadic increase in activity, visible on spike trains [Araya et al. 2022]. We hypothesise that this will also affect the spatio-temporal response of retinal ganglion cells and that the spatio-temporal stimuli we aim to design will enable the diagnosis of early symptoms of Alzheimer’s disease from retinal responses.
Partners
Adrian Palacios, Facultad de Ciencias, Universidad de Valparaíso
Rodrigo Cofré, Universidad de Valparaíso and Institut des Neurosciences, Paris-Saclay – NeuroPSI CNRS
Ludovic Sacchelli, MacTao team, Inria Sophia Antipolis
Luc Pronzatto, I3S, Sophia Antipolis.
