According to the neural efficiency hypothesis (NEH), individuals with higher intelligence display a more efficient brain activity when performing tasks within their area of competence. The NEH has already been supported using a broad variety of tasks. The aim of this study was to test the NEH in the context of mental arithmetic. Therefore we compared the extent of event-related (de)synchronization (ERD/ERS) in the upper alpha (10-12Hz) and theta frequency bands (4-7Hz) of individuals with higher and lower numerical intelligence (n = 19; n = 20, respectively) while they were solving additions.Individuals with higher numerical intelligence solved medium and difficult additions faster. Moreover, it was observed that these individuals had a tendency to correctly solve more medium and difficult tasks. However, these behavioral differences could not be associated with significant differences in brain activity. Nevertheless, we could observe trends that correlated with the NEH. In difficult additions, a negative correlation between numerical intelligence and the magnitude of the ERD in the upper alpha frequency band was observed. Since the magnitude of the ERD in the upper alpha frequency band serves as an indicator of general brain activation, these findings suggest a lower brain activity in individuals with higher numerical intelligence. The explorative analysis of the theta frequency band revealed a descriptively stronger ERS in individuals with higher numerical intelligence. These differences increased with the complexity of the arithmetic tasks. Thus indicating that individuals with higher numerical intelligence use a higher amount of fact retrieval. Being a pioneer study, this thesis gives initial indications for the validity of the NEH in mental arithmetic.