Memory formation is one of the most sophisticated and fascinating functions of the nervous system. Our brains remain plastic throughout life and enable us to store and access an amount of information that no computer could cope with. While the physiological mechanisms underlying learning have been studied extensively using in vitro preparations, much less is known about how memories are stored at the level of local neuronal circuits in the intact animal. This is an essential question since the dynamic interaction of neurons in their networks is absolutely critical to any form of higher brain function. This proposal outlines a strategy to elucidate how memory formation and extinction alters the activation patterns of local, identified neuronal networks. We will apply a novel and multi-disciplinary approach which combines the recently devised technique of in vivo two-photon calcium imaging using GRIN lenses with established molecular (transgenic mouse lines) and behavioural (fear conditioning) approaches. This will enable us to optically record the activity of many neurons simultaneously and to discern the two major neuronal populations, principal neurons and interneurons, which have opposing roles in the brain. This strategy will be used to study how fear memories manifest themselves in the mouse amygdala. In addition, we will characterise the mechanisms of fear generalisation and extinction. These phenomena are implicated in several human anxiety disorders which afflict an estimated 13% of the population in the EU. Our results will help to better understand the underlying causes of these conditions and potentially reveal novel strategies for pharmaceutical intervention. In addition, this research will define novel concepts on memory storage at the network level, which will be of great interest to the scientific community as well as society at large.