Temporal memory refers to the ability to encode, store, and retrieve information about the sequence and timing of events. It is particularly impaired in the early stages of Alzheimer's disease (AD), constituting one of its main clinical markers. Studying these alterations in animal models is crucial for understanding underlying mechanisms. Behavioral experiments require precise control of data streams. The goal of this project was to design and develop an automated experimental arena to assess temporal learning and memory in wildtype and triple-transgenic (3xTg) mice, focusing on their ability to recognize temporal sequences. To achieve this, a circular chamber was designed, equipped with six lights, a video camera, and a reward dispenser, controlled by an Arduino board and integrated through Bonsai, a reactive programming framework. This novel protocol consisted of sequences of light stimuli paired with rewards. Mice were first trained to associate rewards with timely responses to light cues, after which sequence length was progressively increased. Position tracking was also performed using open-source software, DeepLabCut. Custom algorithms developed in our laboratory were applied to analyze the movement patterns, reaction times, and errors in both groups. This setup provides a robust and fully automated platform for investigating sequence learning and temporal memory in mice. It provides precise behavioral measurements critical for understanding cognitive impairments.