Recent interest concerning methylmercury that accumulates in the aquatic food chain appears to be directed not to the adverse effects of high-dose exposure in humans, but to the critical concentration at which methylmercury may affect the progeny of the exposed population. In epidemiological studies, however, uncertainties and limitations in estimating exposure make it difficult to quantify dose-response associations and can thereby lead to inaccuracies when deriving such concentrations. In this respect, benchmark dose calculation for quantitative outcomes may shed new light on the epidemiological procedure for estimating the critical concentration. After the epidemic outbreaks of methylmercury poisoning in Japan and Iraq, large-scale follow-up studies were carried out in the Faroe Islands, Seychelles and New Zealand, to clarify the effects of prenatal methylmercury exposure on child neurodevelopment in the latter half of the 1980s. This article presents an overview of the outcomes obtained from the Faroe Islands Prospective Study, as well as a brief interpretation of the benchmark dose calculation. Although the Faroe and Seychelles Islands studies did not seem to differ greatly in the study setting, such as the exposure level and sample size, the former study observed some significant dose-effect relationships between methylmercury concentrations at birth and neurobehavioral end-points, but the latter failed to find any significant associations except in one test. The discrepancy between the two conclusions is also discussed.