Modelling the impact of aging in the early visual system (Master 2 project)
The Aging in Vision and Action Lab is interested in how ageing impacts visual perception and spatial cognition. The current project deals with visual system modeling using artificial spiking neural networks. We designed a biologically detailed model of the cat early visual system including the retina, the lateral geniculate nucleus (LGN), and the primary visual cortex (V1). The model can process natural images and videos, and has self-learning abilities. We study how neurons of the different processing stages process and transmit visual information through temporal coding. Our last results suggest a critical role for saccades and microsaccades: these eye movements appear to be necessary and sufficient to have coding by synchrony, a scheme in which neurons strongly stimulated tend to spike together, and patterns of synchronous spikes signal visual features such as edges.
We now want to understand how ageing impacts this scheme. Some experiments suggest that the precision of spike timing decreases with ageing, in particular in the LGN and V1. This could be due to retinal cell degeneration. Furthermore, it appears that eye movements’ frequency, amplitude, and precision also decrease with age, which could also impact visual processing. Finally, we also have reasons to believe that plasticity decreases with age, in particular in V1.
The successful applicant will:
(1) review the above-mentioned experimental literature;
(2) implement the ageing effects in the model and test their effects on visual processing.
Candidates should have solid quantitative skills and a keen interest in neural models of brain functions. Programming experience is also desirable, but not mandatory.
For further inquiries, please contact Angelo.Arleo@inserm.fr or Timothee.Masquelier@inserm.fr.