Biology subjects

Masquelier, T.

Publications and source records attributed to Masquelier, T..

5 recordsLinked to original sources

bioRxiv2025

Temporal recurrence as a general mechanism toexplain neural responses in the auditory system

Computational models of neural processing in the auditory cortex usually ignore that neurons have an internal memory: they characterize their responses from simple convolutions with a finite temporal window of arbitrary duration. To circumvent this limitation, we propose here a new, simple and fully recurrent neural network (RNN) architecture incorporating cutting-edge computational blocks from the deep learning community and constituting the first attempt to model auditory responses with deep RNNs. We evaluated the ability of this approach to fit neural responses from 8 publicly available datasets, spanning 3 animal species and 6 auditory brain areas, representing the largest compilation of this kind. Our recurrent models significantly out-perform previous methods and a new Transformer-based architecture of our design on this task, suggesting that temporal recurrence is the key to explain auditory responses. Finally, we developed a novel interpretation technique to reverse-engineer any pretrained model, regardless of their stateful or stateless nature. Largely inspired by works from explainable artificial intelligence (xAI), our method suggests that auditory neurons have much longer memory (several seconds) than indicated by current STRF techniques. Together, these results highly motivate the use of deep RNNs within computational models of sensory neurons, as protean building blocks capable of assuming any function.

neuroscience↗

bioRxiv2024

A general theoretical framework unifying the adaptive, transient and sustained properties of ON and OFF auditory neural responses

Sounds are temporal stimuli decomposed into numerous elementary components by the auditory nervous system. For instance, a temporal to spectro-temporal transformation modelling the frequency decomposition performed by the cochlea is a widely adopted first processing step in todays computational models of auditory neural responses. Similarly, increments and decrements in sound intensity (i.e., of the raw waveform itself or of its spectral bands) constitute critical features of the neural code, with high behavioural significance. However, despite the growing attention of the scientific community on auditory OFF responses, their relationship with transient ON, sustained responses and adaptation remains unclear. In this context, we propose a new general model, based on a pair of linear filters, named AdapTrans, that captures both sustained and transient ON and OFF responses into a unifying and easy to expand framework. We demonstrate that filtering audio cochleagrams with AdapTrans permits to accurately render known properties of neural responses measured in different mammal species such as the dependence of OFF responses on the stimulus fall time and on the preceding sound duration. Furthermore, by integrating our framework into gold standard and state-of-the-art machine learning models that predict neural responses from audio stimuli, following a supervised training on a large compilation of electrophysiology datasets (ready-to-deploy PyTorch models and pre-processed datasets shared publicly), we show that AdapTrans systematically improves the prediction accuracy of estimated responses within different cortical areas of the rat and ferret auditory brain. Together, these results motivate the use of our framework for computational and systems neuroscientists willing to increase the plausibility and performances of their models of audition.

neuroscience↗

bioRxiv2019

Sub-optimality of the early visual system explained through biologically plausible plasticity

The early visual cortex is the site of crucial pre-processing for more complex, biologically relevant computations that drive perception and, ultimately, behaviour. This pre-processing is often viewed as an optimisation which enables the most efficient representation of visual input. However, measurements in monkey and cat suggest that receptive fields in the primary visual cortex are often noisy, blobby, and symmetrical, making them sub-optimal for operations such as edge-detection. We propose that this suboptimality occurs because the receptive fields do not emerge through a global minimisation of the generative error, but through locally operating biological mechanisms such as spike-timing dependent plasticity. Using an orientation discrimination paradigm, we show that while sub-optimal, such models offer a much better description of biology at multiple levels: single-cell, population coding, and perception. Taken together, our results underline the need to carefully consider the distinction between information-theoretic and biological notions of optimality in early sensorial populations.

systems biology↗

bioRxiv2017

Emergence of binocular disparity selectivity through Hebbian learning

Neural selectivity in the early visual cortex strongly reflects the statistics of our environment (Barlow, 2001; Geisler, 2008). Although this has been described extensively in literature through various encoding hypotheses (Barlow and Foldiak, 1989; Atick and Redlich, 1992; Olshausen and Field, 1996), an explanation as to how the cortex develops the structures to support these encoding schemes remains elusive. Here, using the more realistic example of binocular vision as opposed to monocular luminance-field images, we show how a simple Hebbian coincidence-detector is capable of accounting for the emergence of binocular, disparity selective, receptive fields. We propose a model based on spike-timing dependent plasticity (STDP) which not only converges to realistic single-cell and population characteristics, but also demonstrates how known biases in natural statistics may influence population encoding and downstream correlates of behaviour. Furthermore, we show that the receptive fields we obtain are closer in structure to electrophysiological data than those predicted by normative encoding schemes (Ringach, 2002). We also demonstrate the robustness of our model to the input dataset, noise at various processing stages, and internal parameter variation. Taken together, our modelling results suggest that Hebbian coincidence-detection is an important computational principle and could provide a biologically plausible mechanism for the emergence of selectivity to natural statistics in the early sensory cortex.

neuroscience↗

bioRxiv2017

Convis: A Toolbox To Fit and Simulate Filter-based Models of Early Visual Processing

1We developed Convis, a Python simulation toolbox for large scale neural populations which offers arbitrary receptive fields by 3D convolutions executed on a graphics card. The resulting software proves to be flexible and easily extensible in Python, while building on the PyTorch library [32], which was previously used successfully in deep learning applications, for just-in-time optimization and compilation of the model onto CPU or GPU architectures. An alternative implementation based on Theano [33] is also available, although not fully supported.\n\nThrough automatic differentiation, any parameter of a specified model can be optimized to approach a desired output which is a significant improvement over e.g. Monte Carlo or particle optimizations without gradients. We show that a number of models including even complex non-linearities such as contrast gain control and spiking mechanisms can be implemented easily.\n\nWe show in this paper that we can in particular recreate the simulation results of a popular retina simulation software VirtualRetina [35], with the added benefit of providing (1) arbitrary linear filters instead of the product of Gaussian and exponential filters and (2) optimization routines utilizing the gradients of the model. We demonstrate the utility of 3d convolution filters with a simple direction selective filter. Also we show that it is possible to optimize the input for a certain goal, rather than the parameters, which can aid the design of experiments as well as closed-loop online stimulus generation. Yet, Convis is more than a retina simulator. For instance it can also predict the response of V1 orientation selective cells.\n\nConvis is open source under the GPL-3.0 license and available from https://github.com/jahuth/convis/ with documentation at https://jahuth.github.io/convis/.

neuroscience↗