eliezyer
(((Dr Hannah Wirtshafter))) 🔬🚨 New Preprint 🚨 Ever wondered how the HPC supports learning across different contexts? This study, w @SaraASolla & @DisterhoftLab, uncovers a 'universal' memory code in the HPC—consistent across animals and environments! 🧵👇
1️⃣ Contextual generalization is crucial for learning and navigating the world. We focused on how the hippocampus represents task-related information across different environments, even while place cell encoding remaps.
2️⃣ Using calcium imaging, we tracked hippocampal neuron activity as rats performed a conditioning task in multiple contexts. Despite remapping of place cells, task-related neural representations stayed consistent across environments:
3️⃣ Advanced dimensionality reduction and machine learning techniques (shoutout to @cebraAI !!) revealed that neural manifolds representing the task maintained similar geometries across contexts (even as these same cells remapped spatially...
4️⃣ .... showing that task encoding in the hippocampus is resilient to context changes!
5️⃣ Most surprising: this stable task representation wasn't just observed within the same animal, but also across different animals! This points to a standardized neural code or 'neural syntax' shared across individuals.
6️⃣ This discovery bridges the gap between memory and navigation research, showing how the hippocampus preserves cognitive consistency across varying spatial environments—an essential aspect of learning and memory.
7️⃣ Of course there is way more in the paper!! Feel free to hit me up with any comments/criticisms/questions! And many thanks to coauthors/mentors @SaraASolla and @DisterhoftLab, as well as @NU_BSA , the Brain Initiative k99/r00, among others