Use of electromyography (EMG) to monitor mastication is a relatively new concept in assessing food physical and sensory properties. Although expert assessment of cheese characteristics is widely used, the effect of training in sensory analysis on mastication patterns, as assessed using EMG is not well known. Nine samples of the same Mahon cheese (60 days ripening) were given to 24 subjects (8 experts, 16 untrained) and EMG recorded for each chewing sequence. Three samples were tested in a single session by each subject, and three sessions carried out on different days. EMG was recorded from four masticatory muscles for each subject. From EMG records the following was extracted: number of chews, chewing time, mean and maximum voltage of EMG bursts (i.e. chews) across chewing sequence, chewing work and chewing rate. No gender bias was found for the EMG parameters considered, therefore, as regards gender, each group was considered to be homogeneous. Variability within-subjects across samples was greater for experts than untrained subjects. Significant differences in chewing time, chewing work and chewing rate were found between the expert and untrained groups. Data analysis of the three sessions showed an influence of cognitive constructs, mediating states, on the chewing process. The experts were found to be goal driven as to their mastication process. Experts showed no significant differences between sessions, untrained subjects were found to vary their EMG output in successive sessions for number of chews, chewing time, mean voltage, and chewing work.