This study was derived from a model of the development and maintenance of occupational pain (Wright, 1987), particularly occupational overuse syndrome. The model suggests that occupational pain is maintained by a pain cycle that has several entry points, including that of psychosocial job stress. Part of the model, starting at this point, led to the proposition that job stress is positively associated with anxiety, the bracing of the muscles of the arms, neck and shoulders, and musculoskeletal discomfort, of keyboard operators. Each of the variables was measured in a sample of 47 male and female visual display unit operators employed by several public and private organisations. Three questionnaires were modified for the purposes of the study: the Stress Diagnostic Survey (Ivancevich, Matteson, & Dorin, 1988), the State-Trait Anxiety Inventory (Speilberger, 1983) and the Nordic Questionnaires (Kuorinka et al., 1987). Surface muscle tension was recorded using electromyography, as subjects worked and stopped working. Results showed some statistically significant correlations between variables, supporting in part the hypotheses tested. Because the study is of cross-sectional design, cause-and-effect relationships are not conclusive. However, the significance of the results in the prevention of occupational overuse syndrome, by reducing muscle tension, strengthening muscles, and taking rest pauses, is discussed.