Objective: The aim of this study was to investigate different assessment methods of a two-colour chewing gum test for masticatory efficiency to determine its validity for research and clinical purposes. Materials and methods: Twenty adult volunteers, eleven women and nine men (mean age of 27.5 years), participated in this study. All participants perceived their masticatory efficiency as normal. The task was to chew five samples of a two-colour chewing gum for 5, 10, 20, 30 and 50 cycles respectively. Maximum bite force was measured. All samples were assessed twice by two independent operators both, as 'bolus' and after flattening to 1 mm thick 'wafers'. The latter were scanned and the unmixed pixels counted using Adobe Photoshop Elements((R)) to calculate the ratio of unmixed colour to the total surface. Results: Digital image processing confirmed a significant correlation between colour mixing and chewing duration (P < 0.001). Subjective assessment proved less accurate with fair to substantial intra-examiner agreement for 'bolus' (0.20 < kappa < 0.63) and substantial to almost perfect agreement for 'wafer' (0.60 < kappa < 0.88). Inter-examiner agreement was consistently moderate or substantial only for specimen chewed 20 cycles or longer. No significant correlation was found between the colour mixture and the maximum bite force. Conclusion: Digital image processing of the two-colour chewing gum test specimen provides reliable quantitative data for chewing efficiency. Visual assessments were less reliable but might still be useful in screening for chewing deficiencies in a clinical setting. In this context, the test should be performed with a flattened specimen chewed, probably for 20 cycles.