Despite decades of laboratory and clinical research, Alzheimer's disease (AD) is still the leading cause of dementia in adults and there are no curative therapies currently available for this disease. This may be due to the pathological features of AD, which include extensive extracellular amyloid plaques and intracellular neurofibrillary tangles, as well as subsequent neuronal and synaptic loss, which begin to appear several years prior to memory loss and the damge is already irreversible and extensive at the time of clinical diagnosis. The poor therapeutic effects of current treatments necessitate the introduction of experimental models able to replicate AD pathology, particularly in the pre-symptomatic stage, and then to explore preventive and therapeutic strategies. In response to this necessity, various experimental models reproducing human AD pathology have been developed, which are also useful tools for therapeutic screening. Although none of these models fully reproduce the key features of human AD, the experimental models do provide important insight into the pathological changes which occur in AD. This review summarizes the commonly used experimental models of AD and also discusses how the models may be used to decipher the pathogenesis underlying AD and to screen novel therapies for this disease.