Back

Integration of ethologically defined anxiety-related behaviors in the ventral hippocampus

Li, K.; Kyprou, M.; Ciocchi, S.

2026-01-25 neuroscience
10.64898/2026.01.24.701488 bioRxiv
Show abstract

Anxiety is a complex emotional state that unfolds as structured sequences of risk assessment and exploratory behaviors enabling animals to evaluate potential threats in the environment. The ventral hippocampus (vH) regulates anxiety responses, yet the neuronal substrates orchestrating specific ethologically defined anxiety-related behaviors remain unclear. Here, we combined high-resolution 3D behavioral tracking with cell-type specific in vivo calcium imaging and optogenetic manipulations to dissect vH microcircuit dynamics during anxiety. We identified protected and unprotected stretch-attend postures (pSAP and uSAP), along with head dipping, as core anxiety-related risk-assessment behaviors organized along a spatial gradient in the elevated plus maze (EPM). Ventral hippocampal (vH) pyramidal neurons were broadly engaged across all risk-assessment behaviors, consistent with a generalized role in encoding anxiety-related information. In contrast, interneuronal subclasses exhibited striking functional specialization: parvalbumin (PV) interneurons were selectively recruited during uSAP and head dipping, behaviors associated with direct threat exposure, whereas somatostatin (Sst) interneurons were preferentially activated during pSAP, which reflect approach-avoidance conflicts and decision-making processes. Collectively, these findings establish a microcircuit-level framework in which distinct vH neuronal subclasses differentially gate risk-assessment strategies, enabling flexible transitions between avoidance and exploratory behaviors during anxiety.

Matching journals

The top 6 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.