The review aimed at quantifying the evidence of performance effects resulting from occupational exposure to manganese. Epidemiological studies published between 1987 and 2008 were analyzed. The approach was based on the meta-analytical method of effect size estimates and sought to contribute to the following issues: (1) identification of the affected functions; (2) identification of sensitive neuropsychological tests; (3) analyses of exposure-effect relationships. Thirteen studies examining 958 exposed and 815 unexposed workers were included in the meta-analysis. Mean concentrations of inhalable manganese ranged from 0.05 to 1.59 mg/m(3), mean concentrations of manganese in whole blood ranged from 8.1 to 48.4 microg/L. Nineteen neuropsychological performance variables were analyzed as they were included in at least three of the identified studies. Apart from two outcomes, the overall effects displayed a negative impact of manganese on performance. Significant overall effects were obtained for six test variables; their size ranged from d=-0.23 to -0.36. Four of the variables measured motor speed and two of them speed of information processing. The analysis of exposure-effect relationships showed that larger effect sizes were more consistently associated with higher concentrations of inhalable manganese than with manganese in blood. The evidence of cognitive and motor performance effects is in accordance with the knowledge about accumulation of manganese in the basal ganglia and the effect of manganese on the neurotransmitter dopamine. Inconsistencies in the relationship between effect sizes and the biomarker manganese in blood were discussed in the context of results indicating that the biomarker might not be sufficiently meaningful for the neurobehavioral alterations. Simple motor performance tests with a distinct speed component seem to be highly recommendable for further studies, because they seem to be appropriate for measuring manganese-related changes, seem to provide homogenous results and their outcomes show consistent relations to exposure. The rigorous quantitative approach was especially appropriate for revealing exposure-effect relationships, but information about individual cumulative exposure would enhance the potential for risk assessment of manganese.