Objective: To test the hypothesis that stunted Nepalese children have an altered stress response system when compared with matched nonstunted children in response to a battery of psychological tests. Design: Case-control study. Setting: Poor urban areas of Kathmandu, Nepal. Subjects: A total of 64 stunted (less than -2 s.d. height-for-age) children compared with 64 nonstunted (4 À1s.d. height-forage) schoolchildren between 8 and 10 y old matched for school and sex. Methods: A psychological test session was administered, which included mental arithmetic and two tests of working memory. Salivary cortisol samples were obtained at five points during testing, and heart rate was measured during testing and also at baseline. Salivary cortisol samples were also obtained once early in the morning. Hemoglobin was assessed at the testing session, and extensive data were obtained on the social background of the children's families. Results: Stunted Nepalese children showed a blunted physiologic response (salivary cortisol and heart rate) to psychological stressors (Po0.05) when compared with nonstunted children, but were not different from the nonstunted children in baseline measures, when controlling for social background. The two groups were not different in terms of social background. Conclusions: These findings suggest that childhood growth retardation may be associated with changes in physiological arousal, and that stunting could be associated with hyporesponsivity in response to psychological stress.