The recent development of recombinant viral vectors that are capable of transducing postmitotic cells may provide a powerful new tool for studying brain function, as well as ameliorative strategies in models of neurological disease. Some of these vectors have recently demonstrated direct expression of biologically relevant levels of protein expression (Mandel et al., 1997 , Mandel 1998 ; Leff et al., 1999 ; Bl�mer et al., 1998 ) for varying periods of time of up to 1 yr after direct intracerebral injection (Mandel et al., 1997 , Mandel et al., 1998 ; Bl�mer et al., 1997 ; Naldini et al., 1996a ). Moreover, vectors injected into the nervous system have been shown, not only to be capable of expressing de novo transgenes, but also to be able to regionally suppress gene expression by delivery of antisense oligonucleotides (Xiao et al., 1997 ), by expression of a dominant-negative protein, by local knock-out of genes in transgenic animals carrying a transgene surrounded by lox-P sites (Donello et al., 1998 ), or by the production of molecules capable of inhibiting specific mRNAs, called ribozymes (Lewin et al., 1998 ). Thus, the use of viral vectors has great potential, not only for the study of gene therapy (GT), but also for the study of the molecular basis of certain brain functions.