2026 ECNP Congress

PS03-3255

Ventral pallidum activity links aggression experience to social-rank change

Jonathan Reinwald — N. Danai, W. Weber-Fahr, W. Kelsch · University Medical Center Mainz, Department of Psychiatry and Psychotherapy, Mainz, Germany

2026 ECNP Congress

Introduction

Threat-triggered aggression can perturb social hierarchies, but behavioural strategies and neural circuits predicting social-rank change remain poorly understood. Winning history influences stress susceptibility [1], suggesting that aggression experience may interact with pre-existing social structure. The ventral pallidum (VP) is embedded in motivational pallido-thalamo-cortical circuits relevant to action selection and defensive responding [2,3], but its role in linking aggression experience to social outcome is unclear.

Aims

We identified behavioural and whole-brain neural predictors of social-rank change after brief aggression experience. Based on the VP-centred signature, we tested whether acute chemogenetic perturbation of VP output during aggression and subsequent social-odour exposure paradigm alters aggression-related behaviour, network responses, and rank dynamics.

Methods

Groups of 11-12 male C57BL/6J mice lived for several weeks in semi-naturalistic habitats (Non-invasive Sensor-rich Maze; NoSeMaze), where automatically tracked incidental tube competitions quantified social rank. In a first, unmanipulated predictor cohort, 22 mice underwent brief aggression experience with a CD1-strain aggressor outside the home environment. Rank change was defined as post-aggression minus pre-aggression rank, each quantified over 10-day periods. Aggression encounters comprised three short direct confrontations, each followed by a protected safety-fence phase, during which median interpersonal distance from the aggressor was quantified. Mice then underwent awake 9.4 T fMRI during sensory exposure to social odours from the familiar aggressor and from an unfamiliar CD1-strain mouse. Blood oxygenation level-dependent responses were analysed with cluster-level family-wise error correction. Task functional connectivity was estimated by beta-series correlation, together with local network metrics. In a second, manipulation cohort, 36 mice expressed an inhibitory pharmacologically selective actuator module in VP neurons [4]. Varenicline or saline was administered before aggression exposure and imaging (18 per group) to perturb VP output during aggression-related behavioural and imaging phases.

Results

Shorter interpersonal distance during the safety-fence phase predicted upward rank shifts (r=0.54, p=0.01, n=22). Whole-brain analysis of aggression-related social-odour responses identified activation in the VP and in parts of the mediodorsal thalamus correlating with later rank change, with the VP signal showing the strongest association (r=-0.794, p<0.001; family-wise error-corrected p<0.05). This response was selective for the familiar aggressor odour compared with unfamiliar CD1-strain odour, remained stable after controlling for pre-encounter rank, was robust at a stringent threshold (cluster-defining p<0.001), and was mainly driven by wins rather than losses. Network analyses supported a hub-like VP role with its betweenness centrality predicting rank change (r=-0.65, p<0.001). Chemogenetic VP perturbation increased task-evoked activation in downstream thalamic and periaqueductal grey regions, but reduced VP-thalamus coupling. Behaviourally, varenicline-treated mice showed shorter interpersonal distance than saline controls (estimate=-4.43, standard error=1.26, t=-3.51, p=0.0005) and more active defence. They also showed stronger upward rank shifts after aggression, although this effect was attenuated after correction for initial rank and remained trend-level.

Conclusions

Aggression-related social-odour re-exposure revealed a VP-centred brain signature that predicted later social-rank change. Acute VP perturbation altered aggression-related behaviour and downstream network responses, with a trend-level effect on subsequent rank dynamics. Together, these findings support VP involvement in linking aggression experience, defensive state control, and later social trajectory.

References

  1. LeClair KB, Chan KL, Kaster MP, et al. eLife. 2021;10:e71401.
  2. Leung BK, Balleine BW, Corbit LH. Current Biology. 2024.
  3. Wolff M, Halassa MM. Neuron. 2024;112:893–908.
  4. Magnus CJ, Lee PH, Atasoy D, et al. Science. 2019;364:eaav5282.