Online Object Detection (OOD) requires learning novel object categories from a stream of images, similarly to an agent exploring new environments. In this context, the widely used Faster R-CNN architecture faces catastrophic forgetting: acquiring new knowledge leads to the loss of previously learned information. In this paper, we investigate the learning and forgetting mechanisms of the Faster R-CNN in OOD through three main contributions. Firstly, We show that the Faster R-CNN’s forgetting curves reflect human memory cognitive processes as developed by Hermann Ebbinghaus: knowledge is lost exponentially over time and recalls enhance knowledge retention. Secondly, we introduce a new methodology for analysing the Faster R-CNN architecture and quantifying forgetting across the Faster R-CNN components. We show that forgetting is mainly localized in the Softmax classification layer. Lastly, we propose a new training strategy for OOD called Configurable Recall (CR). CR performs recalls on old data using images stored in a memory buffer with variable frequency and recall length to ensure efficient learning. CR also masks the logits of old objects in the Softmax classification layer to mitigate forgetting. We evaluate our strategy against state-of-the-art methods across three OOD benchmarks. We analyze the effectiveness of different recall types in mitigating forgetting and show that CR outperforms existing methods