OBJECTIVES: The aim of this in vivo study was to investigate how reproducible individuals find their habitual intercuspation. METHODS: A new type of highly accurate measuring procedure was involved using an intraoral optical scanning device (Cerec Bluecam). First, a pilot test investigated the accuracy of the procedure within a standardized setting. With regard to the in vivo study, fifteen participants with full dentition were selected. For each of these individuals, eight scans were taken both in the morning and in the afternoon. Furthermore, during each session, scans were taken both in horizontal and upright position. In order to compare this to ubiquitous used processes, plaster replicas of five individuals were investigated as well. The scans were analysed, and the differences in the position of the lower jaw were calculated by a specialized superimposition program (n=570 comparisons/OraCheck). RESULTS: The results showed that there was no significant difference between the time of day and the position of the patient. The overall mean±SD value for locating the habitual intercuspation was 42±34m, however ranging from 22±9m to 77±58m for single individuals. On the other hand, the differences in positioning plaster replicas reached a mean of 135±77m. CONCLUSIONS: The reproducibility of the habitual intercuspation can be obtained under in vivo conditions by a newly developed and highly accurate measuring procedure. Individuals with full dentition show values in average of 42m. CLINICAL SIGNIFICANCE: Determining the occlusal jaw relation is an important precondition in restorative dentistry and many methods are proposed for a proper occlusal registration. Although much is known about in vitro accuracy of these techniques, little is known how reproducible the habitual occlusal position itself is found between individuals.