Parametric copula models reveal neuronal and behavioral time-dependent relationships in primary visual cortex

Nina Kudryashova, Theoklitos Amvrosiadis, Nathalie Dupuy, Nathalie Rochefort, Arno Onken

Research output: Contribution to conferencePosterpeer-review

Abstract / Description of output

Recent research in neuroscience shows that neural responses to stimuli are strongly modulated by behavioral context even as early as in primary sensory cortical areas [1, 2]. Many research groups reported alteration in the activity of the primary visual cortex in response to the running speed [3, 4], whisking [5, 6], arousal [7, 8], and even to more subtle and spontaneous face movements [2]. Such mixing of behavioral and sensory data appears to be beneficial for choosing the best action in a given situation and might explain the amazing versatility of sensory processing. Nonetheless, the precise functional interactions underlying behavioral modulation of sensory inputs remain unknown.
The unbiased analysis of this modulation proves to be challenging due to the different statistics (e.g. discrete licks or neuronal spikes vs. continuous velocity or calcium signals) and different timescales of the observed variables. In order to overcome these hurdles, we propose parametric copula models with time-varying parameters, which separate the statistics of the variables from their dependence structure. For model fitting, we use Gaussian Process latent variable inference schemes, which are naturally suited to take into account different timescales. Therefore, the doubly stochastic copula approach with underlying temporal profiles provides a complete probabilistic model of the involved joint variables.
We applied our model to two-photon calcium imaging data of neuronal populations in the primary visual cortex of awake behaving mice performing a spatial navigation task in a virtual reality setup. We have found pairs of variables showing strong non-trivial changes in the dependence structure subject to the position in the virtual reality and related visual information (see Figure 1). Our preliminary findings agree with previous studies, showing that beyond independent responses, functional connectivity carries additional visual information [9-11]. However, contrary to other commonly applied methods, our parametric copula-based approach makes stochastic relationships explicit and generates interpretable models of dependencies between neural responses, visual stimuli, and behavior.
Original languageEnglish
Number of pages1
Publication statusPublished - 19 Sept 2019
EventBernstein Conference 2019 - Berlin, Germany
Duration: 18 Oct 201820 Oct 2018


ConferenceBernstein Conference 2019
Abbreviated titleBC19
Internet address


Dive into the research topics of 'Parametric copula models reveal neuronal and behavioral time-dependent relationships in primary visual cortex'. Together they form a unique fingerprint.

Cite this