Skip to content


Open Access

Rapid neural coding in the mouse retina with the first wave of spikes

  • Geoffrey Portelli1Email author,
  • John Barrett2,
  • Evelyne Sernagor2,
  • Timothée Masquelier3, 4 and
  • Pierre Kornprobst1
BMC Neuroscience201415(Suppl 1):P120

Published: 21 July 2014


Spatial FrequencyDiscrimination TaskRetinal Ganglion CellLateral Geniculate NucleusRelative Latency

For flashed stimuli presentations, it is known that the latency of the first spike of retinal ganglion cells (RGC) encodes information about the stimulus. This was shown at the level of individual neurons but also at the population level by considering the relative latency between certain pairs of RGC [1]. In this work, we further investigated this population code on mouse retinas using a 60-channel MEA (469 RGC pooled from 5 retinas) in response to gratings of varying phase, spatial frequency.

Interestingly, due to the presence of a high spontaneous activity, we did not find any RGC pair showing a clear relation between the relative latencies and stimuli as in [1]. So we extended this analysis to the whole population instead of looking at individual pairs, by considering the relative order of all spike latencies, i.e. the shape of the first wave of spikes (FWS) after stimulus onset.

We first showed that the FWS is specific to each stimuli. To do so, we defined a distance between the FWS of a reference grating and the gratings with similar spatial frequency and varying phases. This distance was correlated to the phase difference between gratings.

Then, to estimate quantitatively the coding efficiency of the FWS, we performed a discrimination task where the aim was to identify the phase among gratings of identical spatial frequency. We compared the performance (fraction of correct predictions, FCP) of the FWS under classical Bayesian decoders to independent response latency of each recorded RGC. Results showed the FWS decoder which is based on the relative rank of latencies only, was as efficient as the pure latency decoder based on absolute latency values (~73% of FCP for both).

Finally, as the spikes from the output of the retina are conveyed and processed by higher neural structure such as the Lateral Geniculate Nucleus (LGN), we investigated the possible effects of an a posteriori processing stage on the neural code. We fed a simulated LGN-like layer [2] with the spikes obtained from our recordings. We then analyzed the output spikes from the simulated LGN using the same discrimination task. As the number of RGC increased, the FWS decoder rapidly outperforms the latency decoder. Considering all RGC, we compared the performance obtained before and after the LGN. Results showed the latency decoder discrimination performance decreased (from 73% to 40% of FCP) although the FWS decoder discrimination performance remained more stable (from 73% to 70% of FCP).



The research received financial support from the 7th Framework Programme for Research of the European Commission, under Grant agreement no 600847: RENVISION project of the Future and Emerging Technologies (FET) programme (Neuro-bio-inspired systems (NBIS) FET-Proactive Initiative).

Authors’ Affiliations

Neuromathcomp, INRIA, Sophia Antipolis, France
Institute of Neuroscience, Medical School, Newcastle University, Newcastle, UK
Institut de la Vision, UPMC Université Paris 06, Paris, France
CNRS, UMR 7210, Paris, France


  1. Gollisch T, Meister M: Rapid neural coding in the retina with relative spike latencies. Science. 2008, 319: 1108-1111. DOI: 10.1126/science.1149639View ArticlePubMedGoogle Scholar
  2. Masquelier T: Relative spike time coding and STDP-based orientation selectivity in the early visual system in natural continuous and saccadic vision : a computational model. J Comput Neurosci. 2012, 32: 425-441. 10.1007/s10827-011-0361-9.View ArticlePubMedGoogle Scholar


© Portelli et al; licensee BioMed Central Ltd. 2014

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.