The purpose of this study was to examine possible differences between the field and laboratory ratings of perceived exertion (RPE) when the performance and physiological measures for the two conditions were equated. Furthermore, the interactive effects of local, central and overall RPE were examined. Finally, the question of the potential effects of attitudes on RPE was addressed. Eleven well-conditioned adult male marathon runners voluntarily participated in the study. After a period of treadmill habituation, biographical and anthropometric data were collected. This was followed by a VO₂ max test and a speed-matching session at 70% of VO₂ max to determine overground running speed at 3.8% and 7. 5% grade. Subjects then completed an attitudinal questionnaire and ran 4km overground. Finally, the above test was repeated on the treadmill, with the gradient and running speed of the overground condition being replicated. Physiological measures and differentiated RPE were obtained during the final two sessions. There were no physical environmental, task characteristic or performance differences between the overground and treadmill conditions. No heart rate or VO₂ differences were observed between the two conditions, but VE was significantly elevated in the laboratory. Local and overall RPE were significantly higher in the laboratory than in the field, but there was no difference for central RPE. Attitudes were more favourable towards the field than towards the laboratory work task. The results suggest that neither heart rate nor VO₂ are major factors directly influencing the perception of exertion. VE however appears to be a potent central signal mediating RPE. The results also indicate that local factors play a more important role in the perception of exertion than was previously thought. Attitudes towards a work task could possibly mediate the self-reports of exertion. The findings of this study also suggest that RPE are influenced by cognition to a large degree. Finally, environmental cues, or the ambience of a particular working environment, can exert a substantial influence on RPE. Direct perceptual translations from laboratory to field situations may therefore be invalid.