Abstract: Objective To investigate the effect of lutein on improving cognitive dysfunction in diabetic mice and related mechanisms. Methods A diabetic model of mice was established by injecting streptozotocin (STZ). Experimental animals were divided into three groups: a control group, a diabetes group, and a lutein group. Mice in the lutein group were intragastrically administered with lutein at 0.2 mg/kg for four consecutive weeks. During the experiments, their body weight and blood glucose changes were monitored. Meanwhile, behavioral indicators were collected from the Morris water maze, open field, and forced swimming tests. Hippocampal tissue from the mice was also collected for immunoblotting analysis. Results Compared with the diabetes group, the lutein group showed increased body weight and decreased blood glucose (P<0.001). In the Morris water maze test, the lutein group showed reduced escape latency and increased distance and time spent in the target quadrant (P<0.001). In the open field test, the lutein group displayed increased walking distance and exploration frequency in the center area (P<0.001). In the forced swimming test, the lutein group showed increased swimming distance (P<0.001) and swimming time (P<0.05). In the hippocampal region of the lutein group, immunoblotting results indicated that the expression of p62 and Bax decreased (P<0.001), while LC3B-Ⅱ/Ⅰ (P<0.05) and Bcl-2 expression (P<0.001) increased. Conclusions Lutein can protect hippocampal neurons in diabetic mice from damage by enhancing autophagy and inhibiting apoptosis, thereby improving cognitive dysfunction in diabetic mice.