AIM: To investigate the effects and possible mechanisms of liraglutide on postoperative cognitive function in aged mice. METHODS: C57BL/6J male-aged mice were randomly divided into 4 groups: control group (N group), liraglutide group (L group), model group (M group), and model+liraglutide group (ML group). There were 12 mice in each group. The model was constructed by using sevoflurane anesthesia combined with dissecting the abdominal cavity in M and ML groups. In L and M groups, liraglutide was injected into the peritoneal cavity at 300 μg/kg, once per day, for 14 days. Postoperatively, the cognitive function of mice was detected by using the open field test, the Y maze experiment, and the conditioned fear experiment. Western blotting and ELISA were used to detect the expression of hippocampal glucagon-like peptide-1 receptor (GLP1R), β-arrestin2 (β-arrestin2), stimulator of interferon genes (STING), TANK-binding kinase 1 (TBK1), IL-1β, and IL-6; immunofluorescence was used to observe the Iba1-positive microglial cell quantity. RESULTS: Compared with N group, M group showed a decrease in the postoperative spontaneous alternation rate, percentage of freezing time, and GLP1R expression in hippocampal tissue (P<0.05) and an increase in the expression of β-arrestin2, STING, P-TBK1, IL-6, IL-1β, and the number of Iba1-positive cells (P<0.05). Compared with M group, postoperative spontaneous alternation rate, percentage of freezing time, and GLP1R expression were increased in ML group (P<0.05), and the number of β-arrestin2, P-TBK1, IL-6, IL-1β, and Iba1-positive cells was significantly decreased in ML group (P<0.05). CONCLUSION: Liraglutide may ameliorate postoperative cognitive impairment in aged mice by inhibiting the β-arrestin2/STING/TBK1 pathway.