The purpose of this laboratory study was to investigate the effects of surface electromyography (EMG)- and mechanomyography (MMG)-based audio and visual biofeedback during computer work. Standardized computer work was performed for 3 min with/without time constraint and biofeedback in a randomized order. Biofeedback was given on the basis of an individual preset threshold value for the right trapezius EMG and MMG signal and a time factor (repetition of events above the threshold). The duration of muscle activity above the preset threshold, the right trapezius EMG and MMG root mean square (RMS) values as well as the work performance in terms of number of completed graph/mouse clicks/errors, the rating of perceived exertion (RPE) and the usefulness of the biofeedback were assessed. The duration of muscle activity above the threshold was significantly lower with MMG compared with EMG as source of biofeedback (p < 0.05). Biofeedback led to a significant decrease in the right trapezius EMG RMS, lower RPE and decreased number of errors and mouse clicks, but also decreased number of completed graphs (p < 0.05). Audio and visual biofeedbacks were as effective. MMG-based biofeedback is a potential reliable alternative to EMG in ergonomics. A lowering of the trapezius muscle activity may contribute to diminish the risk of work related musculoskeletal disorders development.