Objective To investigate the effects of exposure of the head of rats to different doses of ionizing radiation, to establish an appropriate animal model of cognitive function damage caused by such radiation, and to provide a research basis for studying the mechanism of cognitive function damage due to ionizing radiation and the development of radiation-protective agents. Methods Twenty male SPF rats were randomly and equally divided into a control group (C group), a 10 Gy irradiation group (10 Gy IR group), a 20 Gy irradiation group (20 Gy IR group), and a 30 Gy irradiation group (30 Gy IR group). The group C was left untreated, while the other three groups were irradiated with 10,20, and 30 Gy electron beams, focused on the heads of the SD rats. The mortality of the animals was calculated using survival rates at 30 days after the irradiation and spatial memory was assessed using the Morris water maze. The concentrations of glutathione (GSH) and malondialdehyde (MDA) in rat brain cortical homogenate were determined by chemical colorimetry. An enzyme-linked immunosorbent assay (ELISA) was used to detect 8-hydroxydeoxyguanosine (8-OHdG). Brain pathological sections were analyzed by HE staining. Results The rats in the 30 Gy IR group exhibited irregular back hair, excessive salivation, and other symptoms, along with severe hair loss on the head and neck. At 30 days after the irradiation, no animal died in group C. The mortality rates of both the 10 and 20 Gy IR groups were 20%, while that of the 30 Gy IR group was 40%. The results regarding positioning and navigation in the Morris water maze experiment showed that the latency of the 30 Gy IR group in the 1-5-day training period was significantly higher than in the other three groups ( P <0. 05). The spatial search experiment showed that the residence time in the original platform quadrants of the 30 Gy IR group was significantly shorter than in the other groups ( P <0. 05); the number of platform crosses in this group was also significantly reduced ( P <0. 05). The swimming distances in the original platform for the 10, 20, and 30 Gy IR groups were significantly shorter than for the group C ( P < 0. 05). Moreover, the results of oxidative stress-related indicators showed that the concentrations of GSH, 8-OHdG, and MDA in the 30 Gy IR group were significantly different from those in group C ( P <0. 05). The difference of MDA concentration between the 20 Gy IR group and group C was statistically significant ( P <0. 05). Pathological examination also showed that the tissue damage of the 20 and 30 Gy IR groups was more severe than that of the 10 Gy IR group, and there were significant pathological changes such as cell apoptosis and tissue necrosis. Conclusions This experiment shows that a rat model of cognitive dysfunction can be created by a single 30 Gy electron beam irradiation to the head.