The present study examined the effects of instruction type and boredom proneness (BP) on vigilance performance, workload, and boredom. Subjects completed the Boredom Proneness Scale and were assigned to high and low groups based on their scores. They then monitored a VDT for critical signals. Half the subjects were instructed to detect "critical" flickers (detection emphasis), and the remaining subjects were instructed to relax but to respond to any flickers observed (relaxation emphasis). Subjects also provided pre- and postvigil ratings of workload, stress, and boredom. A performance decrement was observed for all conditions. Low-BP subjects outperformed high-BP subjects and reported less boredom. Thus the results from the present study provide evidence for the long-sought, elusive link between trait boredom and performance in vigilance. In addition, subjects who received relaxation-emphasis instructions reported lower workload, frustration, and stress for the vigil than did those receiving detection-emphasis instructions. These results are discussed in terms of a recent dynamic model of stress as it relates to sustained attention.