A hippocampal digital Twin reveals latent structure in freely flying bats
- 1University of California, Berkeley
Abstract
Concurrent advances in artificial intelligence (AI), large-scale neuronal recording, and behavioral tracking have enabled increasingly powerful predictive models of brain-behavior relationships. Yet predictive accuracy does not automatically yield interpretable principles of neural organization, particularly during natural behavior, where the relevant variables computed are often unknown in advance. Here, we developed a functional digital twin of the hippocampus, trained jointly on large-scale neural activity and self-directed three-dimensional flight in freely flying bats. A multimodal, multitask transformer accurately predicted both neural activity and behavior, outperformed standard approaches, and generalized across recording sessions. Interrogating the model's representations suggested a form of hippocampal generalization across self-selected trajectories that was not apparent in physical coordinates, but was plainly visible in the model's low-dimensional latent space. We confirmed this organization directly in the recorded neural activity: neurons fired at corresponding stages of spatially distinct paths connecting the same origin and destination, preserving their relative firing order even as their firing fields shifted with the path. Analysis of neuron-specific readouts further identified functional relationships among neurons that generalized from flight to waking rest and sleep, brain states never used for training, and differed systematically between the hippocampal subregions CA1 and CA3. Together, these results show how AI-based functional digital twins can complement conventional analysis by generating testable hypotheses about neural organization during natural behavior.
Related
- A fully immersive Virtual Reality setup using DeepLabCut reveals species-specific behavioural responses of freely moving coral reef fishes to conspecifics and predators
- Recurrent latent dynamics in spinal circuits enable a digital twin of autonomic state transitions
- Phylogeny of Cortical‐Hippocampal Projections Reveals Selective Elimination of Input From Sensory Regions
- Hippocampal CA3 connectomics reveals a gradient of mossy fiber inputs and selective feedforward inhibition onto pyramidal cells
- Multi-omics analysis reveals chromatin and transcriptomic remodeling in hippocampal CA1 following adolescent social isolation
BibTeX
@misc{lowet2026hippocampal,
title = {{A hippocampal digital Twin reveals latent structure in freely flying bats}},
author = {A. S. Lowet and Y. Qiu and A. Forli and K. C. Kasuba and Y. Lin and A. S. Tolias and M. M. Yartsev},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.09.758148},
url = {https://www.biorxiv.org/content/10.64898/2026.10.09.758148v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - A hippocampal digital Twin reveals latent structure in freely flying bats
AU - A. S. Lowet
AU - Y. Qiu
AU - A. Forli
AU - K. C. Kasuba
AU - Y. Lin
AU - A. S. Tolias
AU - M. M. Yartsev
PY - 2026
DA - 2026/10/09
PB - bioRxiv
DO - 10.64898/2026.10.09.758148
UR - https://www.biorxiv.org/content/10.64898/2026.10.09.758148v1
AB - Concurrent advances in artificial intelligence (AI), large-scale neuronal recording, and behavioral tracking have enabled increasingly powerful predictive models of brain-behavior relationships. Yet predictive accuracy does not automatically yield interpretable principles of neural organization, particularly during natural behavior, where the relevant variables computed are often unknown in advance. Here, we developed a functional digital twin of the hippocampus, trained jointly on large-scale neural activity and self-directed three-dimensional flight in freely flying bats. A multimodal, multitask transformer accurately predicted both neural activity and behavior, outperformed standard approaches, and generalized across recording sessions. Interrogating the model's representations suggested a form of hippocampal generalization across self-selected trajectories that was not apparent in physical coordinates, but was plainly visible in the model's low-dimensional latent space. We confirmed this organization directly in the recorded neural activity: neurons fired at corresponding stages of spatially distinct paths connecting the same origin and destination, preserving their relative firing order even as their firing fields shifted with the path. Analysis of neuron-specific readouts further identified functional relationships among neurons that generalized from flight to waking rest and sleep, brain states never used for training, and differed systematically between the hippocampal subregions CA1 and CA3. Together, these results show how AI-based functional digital twins can complement conventional analysis by generating testable hypotheses about neural organization during natural behavior.
ER -