Home
 /
Papers
 /
Neuroscience
 /
Precision editing of an aggression-encoding network relay suppresses violent action
Preprint
Open access
V1
Top 1% · Validated by QED

Precision editing of an aggression-encoding network relay suppresses violent action

Yael S. Grossman, Austin Talbot, Neil M. Gallagher +6 more
March 2026
 ·
bioRxiv
 ·
DOI 
10.1101/2022.12.07.519272
94
%
Reproducible
QED Impressions
This paper passed independent validation.
QED's automated review checked the evidence supporting each claim, the statistical methods, and whether the results are reproducible from the data provided.

It ranks in the top 1% of preprints assessed in Neuroscience this quarter.
Run full QED analysis
Read paper
Save

Why this is interesting

By decoding aggression as a distributed oscillatory brain state rather than a localized circuit, this work enables real-time closed-loop stimulation and chronic circuit editing that selectively eliminates violent behavior while fully preserving pro-social interactions.

QED claims analysis

Summary

Aggressive behavior in mice is encoded by a distributed brain network (EN-AggINH) that actively inhibits aggression through synchronized theta and beta oscillations relaying through medial prefrontal cortex, and precision editing of a key medial prefrontal cortex-to-nucleus accumbens circuit within this network chronically suppresses violent action while preserving non-aggressive social behaviors.

See full analysis
Main Claim 1 –

Multi-region brain network decodes aggression across mice and contexts

Supporting evidence

  • Direct mPFC stimulation broadly suppresses social behaviors, not aggression-specific
  • Few regions encode aggression specifically; most encode social engagement
  • Single-region aggression signatures fail to generalize across contexts
  • EN-AggINH features theta/beta oscillations with directional network flow
  • EN-AggINH activity correlates with neuronal firing across regions
Main Claim 2 –

Isolation EN-AggINH activity predicts individual trait aggressiveness

Supporting evidence

  • EN-AggINH generalizes across genotypes and chemogenetic aggression induction
Main Claim 3 –

EN-AggINH mediates mPFC stimulation–driven suppression of aggression

Supporting evidence

  • Real-time EN-AggINH estimation enables closed-loop network manipulation
Main Claim 4 –

Closed-loop mPFC stimulation reduces aggression while sparing social behavior

Supporting evidence

  • Aggression suppression requires stimulation synchronized to EN-AggINH timing

Abstract

Aggression is a psychological state characterized by intent to harm oneself or others, and often manifests as hostile, harmful, or violent action. Social aggression can aid an organism in securing access to resources, or it can impair group function and survival in behavioral pathology. Since many brain regions contribute to multiple social behaviors, expanded knowledge of how the brain distinguishes between social states would enable the development of interventions that suppress violent action, while leaving other social behaviors intact. Here we showed that a murine aggressive internal state is encoded by a widespread network. This network is organized by prominent and synchronized theta (4-11 Hz) and beta (14-30 Hz) oscillations that relay through the medial prefrontal cortex, and couples to widespread cellular firing. Strikingly, network activity during social isolation was found to encode the trait aggressiveness of mice, and causal cellular manipulations known to impact aggression were found to modulate the network’s activity. We next established that this network mediates aggression using closed-loop stimulation of medial prefrontal cortex and causal mediation analysis. Finally, we deployed ‘Long-term integration of Circuits using connexins’ (LinCx) to selectively edit a key circuit within the network from the medial prefrontal cortex to nucleus accumbens. Editing this circuit chronically suppressed violent action while leaving non-aggressive social behavior intact.

Related papers

How the QED Score works

QED Score is a validated AI-based quality metric that evaluates life science papers on two dimensions: originality and validity. Each manuscript is anonymized, broken down into its claims and supporting evidence, and examined by specialized AI agents for statistical rigor, consistency across figures, and fit with the existing literature. Scores are percentile ranks, so they're comparable across fields. In blinded tests, expert professors sided with QED Score over journal rank in 75% of decisive cases.
Read the methodology

Free access for academic researchers

Create your free QED account to validate your research, strengthen grant proposals, and uncover scientific insights.
We've sent you an access link.
Please check your inbox.

Didn't get your email? Check your spam folder or reach out to info@qedscience.com

Oops! Something went wrong while submitting the form.
Looking for QED for pharma, biotech or life science organizations?