Concerns about the use of laboratory animals in biomedical research and testing and the relevance of animal data to the human condition are not new issues. Toxicologists have long been aware of the difficulties in interpreting and extrapolating data derived from animals to humans because of species-specific susceptibilities to toxic effects and differences in exposure magnitudes. Such difficulties are compounded by the possibility that data generated in young, healthy, single strain test animals may not apply to a human population that is old, diseased, malnourished, or diverse in its genetic makeup. Since the human population is more heterogeneous than experimental animal strains, it may be sensitive to a wider range of doses than animals. The chapter describes some of the above issues, discusses the potential of current in vitro neurotoxicity tests to address the needs of risk assessment and hazard identification, the advantages and limitations of such models, the scientific validation process, and finally, future directions that might be pursued to encourage their deployment in risk assessment.