Modelling individual ampullary afferents in two species of gymnotiform fish using simulation-based inference

Abstract

Ampullary electroreceptors are widespread across aquatic vertebrates. The purpose of sensing exogeneous electric fields is conserved across species but the implementations differ and the encoding mechanisms remain incompletely understood. We compared baseline and stimulus-driven response properties of ampullary electroreceptor afferents in the weakly electric fish Apteronotus leptorhynchus and Eigenmannia virescens. We find that their activity is very well captured by an extended leaky integrate-and-fire model that generalizes across both species. The model shares similarities to a previous model of the tuberous electroreceptor afferents but further incorporates a low-pass pre-filtering and additional noise sources to reproduce the observed spectral response characteristics. The low-pass is essential to shape stimulus encoding in the high-frequency range. Accurate prediction of low-frequency stimulus encoding further requires two distinct noise sources: stimulus-independent white current noise and activity-dependent noise in the adaptation current, which is shaped by the adaptation time constant to yield pink noise dynamics. Using simulation-based inference, we trained a neural network to map model parameters to neuronal response features. This approach enables the generation of heterogeneous, biologically plausible model populations that may serve as a realistic input layer for studying neuronal processing on the next level. With this, we provide a unified and mechanistic model of ampullary electroreceptor encoding in these species and possibly beyond. The proposed model is another step towards a full model of the electrosensory periphery in these animals. Author summaryThe ability to sense exogenous electric fields, i.e. passive electroreception, is widespread among aquatic animals. It plays a central role in prey detection and, in some species, also contributes to communication. To understand higher-order brain function, we also need to grasp the sensory periphery and ideally have models that provide naturalistic peripheral responses. Using simulation-based inference (SBI), we here develop a mechanistic model of passive electroreception that is valid for at least two species of electric fish. Through detailed analyses, we identify model components, such as a source of pink noise, that are essential for matching the models spectral response properties to those of the recorded cells. The main results of this work are the model itself and the trained inference network (here called the SBI network), which can now be used to create artificial but biologically plausible populations of sensory neurons that may serve as a realistic input layer for studying higher-order neuronal processing. Our work complements the existing models of the active electric sense and is a big step towards a full model of the electrosensory periphery in these animals.