Back

Interactions between the medial prefrontal cortex, dorsomedial striatum, and dorsal hippocampus that support rat category learning

Broschard, M. B.; Kim, J.; Halverson, H. E.; Farley, S. J.; Freeman, J. H.

2025-04-22 neuroscience
10.1101/2025.04.21.649785 bioRxiv
Show abstract

Categorization creates memory representations that are efficient, generalizable, and robust to noise. Multiple brain regions have been implicated in categorization, including the prefrontal cortex, striatum, and hippocampus; however, few studies have examined how these regions interact during category learning. We recorded neural activity in the medial prefrontal cortex (PFC), dorsomedial striatum (DMS), and dorsal hippocampus (HPC) while rats learned to categorize distributions of visual stimuli. We found a learning-related shift in contributions from the PFC (with DMS[->]PFC theta (4-10Hz) interactions) to the HPC (with HPC[->]DMS and bidirectional PFC-HPC theta interactions). Decision-making depended on DMS and HPC spiking, as well as the PFC[->]HPC[->]DMS pathway. Our results provide a framework that characterizes how the PFC-DMS-HPC network interacts during category learning. This is informative for multiple neurological disorders that affect category learning, including Parkinsons Disease, autism, and dementia. Highlights- Rats learned to categorize distributions of visual stimuli. - Early training sessions relied on PFC spiking and DMS[->]PFC theta interactions. - After learning, contributions shifted to HPC spiking, HPC[->]DMS interactions, and bidirectional PFC-HPC interactions. - Decision-making depended on DMS and HPC spiking and PFC[->]HPC[->]DMS interactions. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=180 HEIGHT=200 SRC="FIGDIR/small/649785v1_ufig1.gif" ALT="Figure 1"> View larger version (57K): org.highwire.dtl.DTLVardef@ba3342org.highwire.dtl.DTLVardef@7b62a4org.highwire.dtl.DTLVardef@890676org.highwire.dtl.DTLVardef@75f84_HPS_FORMAT_FIGEXP M_FIG C_FIG

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.