Verbal estimation of peak exertion intensity

The objectives of this research were to investigate the accuracy and precision with which trained and untrained participants estimate the magnitude of forceful exertion and to evaluate the mathematical relationship between actual and estimated exertion. Three groups of participants estimated, as a percentage of maximum voluntary contraction (%MVC), the magnitude of submaximal exertion for 12 simulated tasks. In addition to the control group, one group was exposed to one physical benchmark (100% MVC) and another to three benchmarks (25%, 75%, and 100% MVC) prior to force estimation. Error (estimated minus actual) significantly decreased (p < .0001) from 14% MVC to 4% MVC with one benchmark and to -3% MVC with three benchmarks, as compared with the control group. Furthermore, the standard deviation decreased significantly (p < .0001 ) from the control group (16.6% MVC) to the one-benchmark group (13.8% MVC) to the three-benchmark group (11.6% MVC), indicating improved precision. Significant interaction effects were observed, but their impact on main effects was negligible. Also, linear, power, and logarithmic regression models described the relationship between perceived and actual exertion equally well (R2 = .64-.81). Applications of this research include improving the accuracy and precision of field-based psychophysical estimates of forceful exertion for epidemiological research and other field-based analyses.