The neurons in the mammalian brain can be classified into two broad categories: excitatory and inhibitory neurons. The former has been historically associated to information processing whereas the latter has been linked to network homeostasis. More recently, inhibitory neurons have been related to several computational roles such as the gating of signal propagation, mediation of network competition, or learning. However, the ways by which excitation and inhibition can regulate learning have not been exhaustively explored. Here we explore several model systems to investigate the role of excitation and inhibition in learning and memory formation. Additionally, we investigate the effect that third factors such as neuromodulators and network state exert over this process. Firstly, we explore the effect of neuromodulators onto excitatory neurons and excitatory plasticity. Next, we investigate the plasticity rules governing excitatory connections while the neural network oscillates in a sleep-like cycle, shifting between Up and Down states. We observe that this plasticity rule depends on the state of the network. To study the role of inhibitory neurons in learning, we then investigate the mechanisms underlying place field emergence and consolidation. Our simulations suggest that dendrite-targeting interneurons play an important role in both promoting the emergence of new place fields and in ensuring place field stabilization. Soma-targeting interneurons, on the other hand, are suggested to be related to quick, context-specific changes in the assignment of place and silent cells. We next investigate the mechanisms underlying the plasticity of synaptic connections from specific types of interneurons. Our experiments suggest that different types of interneurons undergo different synaptic plasticity rules. Using a computational model, we implement these plasticity rules in a simplified network. Our simulations indicate that the interaction between the different forms of plasticity account for the development of stable place fields across multiple environments. Moreover, these plasticity rules seems to be gated by the postsynaptic membrane voltage. Inspired by these findings, we propose a voltage-based inhibitory synaptic plasticity rule. As a consequence of this rule, the network activity is kept controlled by the imposition of a maximum pyramidal cell firing rate. Remarkably, this rule does not constrain the postsynaptic firing rate to a narrow range. Overall, through multiple stages of interactions between experiments and computational simulations, we investigate the effect of excitation and inhibition in learning. We propose mechanistic explanations for experimental data, and suggest possible functional implications of experimental findings. Finally, we proposed a voltage-based inhibitory synaptic plasticity model as a mechanism for flexible network homeostasis.