Anterior cingulate input tunes striatal dynamics for context-dependent social preference in a mouse model of autism

Social preference in autism spectrum disorder varies across contexts, but the underlying neural mechanisms remain unclear. Here we show that Shank3 knockout (KO) mice, an autism mouse model, exhibit reduced preference for social over nonsocial stimuli while retaining a preference for females over males. Calcium imaging reveals context-dependent differences between KO and wild-type (WT) mice in the organization of neuronal population dynamics in the dorsomedial striatum (DMS), corresponding to their behavioural preferences.

Functional imaging and molecular analyses reveal reduced functional connectivity between the anterior cingulate cortex (ACC) and DMS and molecular alterations more pronounced in the striatal matrix. ACC inhibition eliminates genotype-dependent differences in DMS dynamics. Inhibiting the ACC–DMS pathway impairs social preference, whereas its activation in KO mice shifts DMS dynamics towards WT patterns and improves social preference.

These findings identify a cortical pathway that shapes striatal population dynamics and context-dependent social preference in an autism mouse model. We thank Kushal Kolar (University of Bergen) for the patient support and guidance on two-photon data processing. We thank Yixin Li and Jiahao Hu (Fudan University), Mackenzie Mathis (Swiss Federal Institute of Technology in Lausanne), Martin Kim (Insitro) and Can Ergen (University of California, Berkeley) for their technical assistance and help.

We thank Beijing Transcend Vivoscope Bio-Technology Co., Ltd, and Nanjing Jingruikang Molecular Medicine Technology Co., Ltd for the support on two-photon data acquisition, and the Analysis & Testing Laboratory for Life Sciences and Medicine of the Fourth Military Medical University for support with confocal imaging. We also thank Lanting Huang, Shihan Deng, Xiaofan Guo, and other colleagues at Leads Bio-Technology Co., Ltd for the support with fUS imaging and analysis. Schematic diagrams were created on BioRender (https://biorender.com).

The language of the manuscript was edited by the Springer Nature Scientific Editing Team (No. This study was supported by Brain Science and Brain-like Intelligence Technology-National Science and Technology Major Project (2025ZD0216200 to B.G.), the National Natural Science Foundation of China (82422030 to B.G., 32394032 and 82221001 to S.W., 32471069 and 82201699 to B.G.), STI2030-Major Project (2021ZD0201005 to S.W.). These authors contributed equally: Yongsheng Zhou, Haoxiang Xiao.

Department of Neurobiology, School of Basic Medicine, Fourth Military Medical University, Xi’an, China Yongsheng Zhou (周永生), Haoxiang Xiao (肖昊翔), Dayun Feng, Kaifang Wang, Kaiwen Xi, Wenting Wang, Shengxi Wu (武胜昔) & Baolin Guo (郭保霖) The Shaanxi Province Key Laboratory of Brain Function Analysis and Modulation, Xi’an, China Yongsheng Zhou (周永生), Shengxi Wu (武胜昔) & Baolin Guo (郭保霖) Air Force Hospital of Eastern Theater Command, Nanjing, China Tangdu Hospital, Fourth Military Medical University, Xi’an, China Institute of Cellular Physiology, National Autonomous University of Mexico (UNAM), Mexico City, Mexico Stanley Center for Psychiatric Research, Broad Institute of Massachusetts Institute of Technology and Harvard, Cambridge, MA, USA Faculty of Psychology, National Autonomous University of Mexico (UNAM), Mexico City, Mexico School of Biomedical Engineering and Technology, Tianjin Medical University, Tianjin, China Correspondence to Shengxi Wu (武胜昔) or Baolin Guo (郭保霖). The authors declare no competing interests. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.