According to statistics from the International Atomic Energy Agency (IAEA), as of December 31, 2023, the IAEA’s incident and trafficking database has confirmed a total of 4,243 cases related to the theft of nuclear materials and other radioactive materials. When a radiation source is stolen or lost in a suspicious area, radiation surveys are typically conducted to locate the anomalous radiation source. To achieve autonomous searching for radiation sources with high precision in a short time, this paper proposes a method based on reinforcement learning with spiking response model (RL-SRM). By leveraging the characteristics of place cells in the hippocampus and the associated information transmission pathways, a spiking neural network model hypothesizing the action cells from hippocampal place cells to the prefrontal cortex is constructed. The state space and action space are represented by place cells and action cells, respectively. The model utilizes an algorithm that combines reinforcement learning with the spiking response model (SRM) for autonomous radiation source searching. Simulation results indicate that, compared to conventional gradient search algorithms and uniform search algorithms, the reinforcement learning based on the spiking response model can accomplish the radiation source search task, reducing the search time by at least 46%.