Great progress has been made over the past 40 years in understanding the hazards of exposure to a small number of developmental neurotoxicants. Lead, polychlorinated biphenyls, and methylmercury are all good examples of science-based approaches to characterizing the hazard to the developing nervous systems from environmental contaminants. However, very little effort has been spent to address the challenge of assessing the potential developmental neurotoxic hazard of the thousands of other chemicals in common commercial use. The extensive time, financial and animal resource requirements for current regulatory testing guideline methods make this an untenable solution to this challenge. A new testing paradigm is needed that uses time and cost-efficient methods to screen large numbers of chemicals for developmental neurotoxicity (DNT). In silico models are needed to provide rapid chemical structure-based screening. In vitro techniques are being developed to provide rapid and efficient testing in cell-free and cell-based systems. In addition, the use of alternative species, such as zebrafish, will provide efficient models for testing the effects of chemicals in organisms with intact developing nervous systems. Finally, these methods and models need to be used in an integrated fashion to provide the data needs for hazard assessment in a manner that is problem-driven and cost-efficient. This paper summarizes discussions on these issues from the symposium 'Developmental neurotoxicity testing: Scientific approaches towards the next generation to protecting the developing nervous system of children' held at the 2011 annual meeting of the Japanese Teratology Society.