Introduction: Animal models play a central role in the experimental study of cognitive dysfunction. The development of specific behavioral tools allows for detailed analysis of the mechanisms involved in spatial learning and memory. The three-tower maze (3T) was designed as an original device for sequential analysis of orientation, decision-making, and memory abilities. Valproic acid, a molecule widely used in neurology, is suspected of impairing cognitive functions through neurotoxic mechanisms. Objective: Experimentally validate the 3T maze as a tool for detecting memory deficits after subchronic exposure to valproic acid in Wistar rats. Methods: Forty-eight Wistar rats of both sexes were randomly divided into three experimental groups: a control group and two groups treated with valproic acid at doses of 200 and 400 mg/kg. The drug was administered orally once a day for 30 days. At the end of the protocol, the animals were subjected to the three-tower maze test. Performance was quantified using parameters combining spatial orientation, cognitive efficiency, decision latency, route errors, and locomotor activity. Results: Animals exposed to valproic acid showed a progressive deterioration in learning and memory abilities, reflected in a significant increase in path errors, longer reward access times, and decreased overall cognitive efficiency. In addition, an increase in avoidance and vigilance behaviors was recorded, suggesting an emotional impact associated with cognitive deficits. Conclusion: The three-tower maze appears to be a relevant experimental tool for the integrated study of memory and behavioral disorders induced by valproic acid. It offers a complementary approach to conventional cognitive exploration tests in rats.