From falcons spotting prey to humans recognizing faces, the ability to rapidly process visual information depends on a foveated retinal organization that provides high-acuity central vision while preserving low-resolution peripheral vision. This organization is conserved along early visual pathways, yet remains under-explored in machine learning. Here, we examine the impact of embedding a foveated retinotopic transformation as a preprocessing layer on convolutional neural networks (CNNs) for image classification. By applying a log-polar mapping to off-the-shelf models and retraining them, we achieve comparable accuracy while improving robustness to scale and rotation. We demonstrate that this architecture is highly sensitive to shifts in the fixation point and that this sensitivity provides an effective proxy for defining saliency maps that facilitate object localization. Our results demonstrate that foveated retinotopy encodes prior geometric knowledge, providing a solution for visual searches and a meaningful classification robustness and localization trade-off. These findings provides a proof of concept in order to connect principles of biological vision with artificial networks, suggesting new, robust and efficient approaches for computer vision systems.

This study investigates the use of Riemannian geometry to detect and monitor physiological states such as mental workload (MWL) from an EEG dataset collected in an aeronautical context. The analysis, based on EEG data recorded from 14 participants performing a Simon’s task after inducing MWL by the Multi Attribute Task Battery-II (MATB), aimed to differentiate low and high workload conditions while tracking MWL effect over time. Using covariance matrices and a Minimum Distance to Mean classifier, Temporal Generalization Method (TGM) was used to assess stable decoding performance, indicating a consistent neural signature of MWL throughout the trials. Results demonstrate spatial effects of mental workload irrespective of the investigated time domain: spatial information is distributed evenly across all explored timescale.

🧠 Excited to share our latest research led by Alexandre Lainé and presented this summer at AREADNE 2026!

“Population decoding of visual motion direction”

Our work explores how populations of neurons in the primary visual cortex (V1) of marmoset monkeys encode visual motion direction, with a particular focus on understanding how uncertainty influences this neural decoding process. Key highlights: