A long-held hypothesis in neuroscience holds that learning and memory mechanisms involve lasting changes in synaptic weights. Multiple mechanisms for producing such changes exist, of which NMDA-receptor–dependent long-term potentiation (LTP) is the most widely studied. Curiously, the relatively simple hypothesis that LTP plays a role in learning and memory has proven difficult to test. A current experimental strategy is to generate genetically altered mice with mutations in genes thought to be involved in LTP and assess the effects of these mutations both on LTP and animal behavior[1,2]. A difficulty associated with these approaches has been that they are not temporally or spatially refined. To alleviate this problem, Dr. Isabelle Mansuy and colleagues used an inducible and reversible transgene expression system in which transgene expression could be controlled on a week-to-week timescale to assess the effects of genetic reduction of the activity of a protein phosphatase known as calcineurin or PP2B in adult mouse forebrain[3,4].