Deep‐learning models of the ascending proprioceptive pathway are subject to illusions

• Published in: Experimental physiology
• Main Authors: Perez Rotondo, Adriana, Simos, Merkourios, David, Florian, Pigeon, Sebastian, Blanke, Olaf, Mathis, Alexander
• Format: Journal Article
• Language: English
• Published: 04.06.2025
• ISSN: 0958-0670; 1469-445X; 1469-445X
• DOI: 10.1113/EP092313

Proprioception is essential for perception and action. Like any other sense, proprioception is also subject to illusions. In this study, we model classic proprioceptive illusions in which tendon vibrations lead to biases in estimating the state of the body. We investigate these illusions with task‐driven models that have been trained to infer the state of the body from distributed sensory muscle spindle inputs (primary and secondary afferents). Recent work has shown that such models exhibit representations similar to the neural code along the ascending proprioceptive pathway. Importantly, we did not train the models on illusion experiments and simulated muscle–tendon vibrations by considering their effect on primary afferents. Our results demonstrate that task‐driven models are indeed susceptible to proprioceptive illusions, with the magnitude of the illusion depending on the vibration frequency. This work illustrates that primary afferents alone are sufficient to account for these classic illusions and provides a foundation for future theory‐driven experiments. What is the central question of this study? Are deep‐learning models of the ascending proprioceptive pathway subject to illusions like those induced by muscle tendon vibrations? What is the main finding and its importance? Task‐driven models trained on arm localization from proprioceptive input exhibit vibration‐induced illusory behaviour. This finding implies the sufficiency of simple feedforward processing of Ia afferents in explaining classic kinaesthetic illusions and enables future theory‐driven experiments.