The reliable integration of an ever-changing multitude of inputs from the environment is fundamental for learning and memory and shapes behaviour. Our research seeks to understand how sensory inputs are represented and integrated to uncover key principals of neuronal computation underlying memory formation in a state-dependent manner. To achieve this, we combine state of the art circuit neuroscience tools like opto- and pharmacogenetics with single and two-photon imaging techniques as well as AI to reveal frameworks of neuronal coding and plasticity in deep brain areas. Our goal is to understand how these mechanisms support memory formation and recall in healthy brains, as well as how they malfunction in the context of disease.