A key issue for drug discovery in the post genomic era is target validation. This is particularly important when considering the CNS, where currently the majority of drug targets are neurotransmitter receptors that are known to exist as multi-gene families. The GABAergic system, which is the major inhibitory neurotransmitter system in the CNS, is no exception in that respect. The GABAA receptors, which are the site of action of a number of clinically used drugs such as benzodiazepines and barbiturates, exist in a large gene family. Existing drugs mediating their effects through the GABAA receptor are generally non-selective, i.e. will act at several subtypes of that receptor family. Thus, if we are both to refine existing therapeutic approaches, and develop novel approaches, a key question is to define which subtype(s) of the GABAA receptor family we should target; which will mediate the beneficial effects of a drug, and which could be responsible for unwanted side effects? One of the tools, which has been developed over the last decade to elucidate the function of a given gene, is the generation and analysis of gene-targeted mice. This review will summarize progress on identifying individual GABAA receptor subtypes as potential drug targets by using genetically modified mice.