Background: The purpose of this study is to evaluate the relationship between brain tumour location and core areas of cognitive and behavioural functioning for paediatric brain tumour survivors. The extant literature both supports and refutes an association between paediatric brain tumour location and neurocognitive outcomes. We examined neuropsychological test data to identify any differences in neurocognitive and behavioural profile associated with supratentorial versus infratentorial tumour location. Methods: Following Institutional Review Board approval, the medical records and neuropsychological test data collected between 1997 and 2002 for 70 children treated for brain tumour at Children's Hospital Los Angeles were reviewed. Fifty-one per cent of the participants had tumours located in the supratentorial regions of the brain, whereas 49% had infratentorial tumours. Primary medical treatments involved tumour resection (90%), cranial radiation therapy (76%), chemotherapy (71%), and 59% all three medical procedures. The two tumour location groups did not differ significantly in the cumulative treatment dose of irradiation to the tumour bed or in the dose delivered to the whole brain. Neuropsychological test data included measures of verbal and non-verbal intellectual functioning, attention/working memory, processing speed, verbal and visual memory, fine motor skills, visual-motor integration, academic achievement, and social-emotional functioning. Differences between the two groups were evaluated using ANOVA, t-tests and chi-squared statistical tests. Results: The supratentorial and infratentorial tumour location groups did not differ on measures of intellectual functioning. However, survivors of infratentorial tumours performed more poorly on selected measures of more specific cognitive functions and on parent-report of social-emotional functioning relative to survivors of supratentorial tumours, even when age at diagnosis was held as a covariate. Higher frequency of auditory deficits was noted in the infratentorial tumour group and was associated with lowered academic achievement scores. Conclusions: The differences by location found in more specific neurocognitive and social-emotional variables, after controlling for age at diagnosis, may possibly reflect tumour location-specific effects. However, this interpretation remains tentative given the limitations in our study and inability to control for the range of medical and treatment-related factors that may have contributed towards the outcomes observed in our sample. At the same time, most of our findings appear consistent with reports from recent studies in this area. (Contains 3 tables.)