Abstract: Objective　To observe the change of congnitive function and the expression level of hippocampal insulin receptors (InsR) in 1 month and 3 months of diabetic rats, explore the effect of InsR on diabetic rats and investigate that whether insulin can be applied to prevent diabetic cognitive impairment.　Methods　36 healthy adult male SD (Sprague-Dawley) rats, weight 180-240 g, were divided into 3 groups randomly: diabetic group (n=12), insulin treatment group (n=12) and control group (n=12), and then each group was divided into 2 groups equal in amount randomly: 1 month group and 3 months group. Diabetic group and insulin treatment group rats were induced by streptozotocin injection. Fixed time to measure blood glucose of all rats twice a week. Insulin treatment group rats were injected insulin daily, and we adjusted insulin dosage according to the blood glucose level to ensure blood glucose in normal range. All the rats underwent the test of Morris water maze in 1 month and 3 months. After the assessment on cognitive function of the 1 month and 3 months rats that experienced the Morris water maze experiment, the expression of InsR in brain were observed.　Results　Compared with the control group, 1 month group of diabetic rats appeared cognitive impairment and their expression of InsR reduced, which has statistical difference (PPPPP<0.05). Compared with 3 months group of diabetic rats, there are no statistical difference.　Conclusion　Diabetic cognitive impairment is closely related to InsR, the reduced InsR expression may contribute to diabetic cognitive impairment. Early intervention of insulin has protective effect on the brain,then delay cognitive impairment. However, as the extension of the duration, the insulin treatment group rats still show cognitive impairment, which suggests that long-term application of insulin can not prevent the occurrence of cognitive impairment. Mechanisms need to be further explored.