|
Article Excerpt
INTRODUCTION
Researchers and practitioners of occupational ergonomics are faced with a limited number of valid and practical tools for assessing the magnitude of physical exertion. To assess the physical demands of using a screwdriver, for example, one might compare the torque required to fasten the screw with the corresponding strength of the target worker population. Although this approach is conceptually straightforward, a lack of valid tools to efficiently evaluate the physical demands of work poses a challenge. Techniques such as direct force measurement and electromyography (EMG) are often not practical for widespread application in occupational settings, so workers themselves are often relied upon to estimate intensity of physical exertion. This paper describes the results of an experiment that investigated the use of psychophysical magnitude estimation as a tool for evaluating forceful exertion.
The benefits of using psychophysical methods to assess exposure to forceful exertions are numerous. From a practical standpoint, psychophysical methods are less costly, in terms of time and money, as compared with the high cost and setup time of instrument-based techniques (Sinclair, 1995). Also, psychophysical methods provide an integrative measure in that they synthesize the many signals elicited from the peripheral working muscles, joints, and central nervous system (Borg, 1990). A criticism of psychophysical methods, however, is that they are inherently subjective and, consequently, are suspect in terms of validity. Direct force measurement and EMG provide "objective" data and may generally be more credible. This view, however, may be limited. EMG, though objective in that it provides a biophysical measurement of the muscle load, is subject to many parameters controlled by the researcher, including muscle selection, calibration technique, signal processing, data analysis, and presentation issues. These variables can have a profound impact on the results that are obtained (Solomonow, 2000).
Like any measurement technique, a psychophysical estimate has some degree of error associated with it, and to make meaningful use of the data it is important to quantify the accuracy (agreement between measurement and actual value) and precision (variability of measurement) that should be expected, as well as the factors that affect the error, so that researchers and practitioners can apply this information to collected data. Previous research (e.g., Chin, Bishu, & Hallbeck, 1995; Cooper, Grimby, Jones, & Edwards, 1979: Krombholz, 1985) has investigated the magnitude of estimation error, but those studies focused on a narrow range of exertions that do not represent those performed in occupational settings.
There is some limited evidence that the precision and accuracy with which participants verbally estimate the magnitude of exertion may improve with training. Deeb (1999) investigated the effects of training on participants' ability to estimate the weight of objects. A significant improvement in accuracy was found when they first held weights of known magnitude prior to estimating the unknown weight. Based on these findings, it is hypothesized that the precision and accuracy of estimation will improve when a participant is systematically exposed to "physical benchmarks" to compare with the exertion being estimated. For example, if a participant is exposed to a maximum exertion prior to estimating a submaximal exertion, the estimate will be more accurate than if he or she had no prior exposure to the benchmark. The primary objective of this research was to examine this hypothesis by investigating the effect of training on the accuracy and precision with which participants verbally estimate the magnitude of exertion intensity for simulated tasks that vary with respect to the degrees of freedom needed to perform them.
A considerable amount of attention has been given to modeling the relationship between perceived and actual force production. Early research by Fechner (1860/1966) investigated the relationship between perceived and physical quantities, and from this came the concept of a "just noticeable difference" (JND), which is the incremental change in perception of the stimulus as its intensity is increased. According to the Weber-Fechner law, the JND is a geometric function of stimulus intensity, meaning the perceived sensation is a log function of the stimulus. However, this theory proved inadequate for very low and very high intensities. As a result, S. S. Stevens (1957) proposed a generic power law, in which the perceived quantity grows as a power function of the actual quantity,.
Researchers have investigated the relationship between perceived and actual exertion, and their results have varied. Whereas several researchers (e.g., Gamberale, Ljungberg, Annwall, & Kilbom, 1987; Ljungberg, Gamberale, & Kilbom, 1982; J. C. Stevens & Mack, 1959) have supported the model proposed by S. S. Stevens (1957), finding exponents ranging from 1.4 to 2.4, others (e.g., Deeb, 1999; Krombholz, 1985) have found the relationship to be linear. Furthermore, other researchers (Chin et al., 1995; Cooper et al., 1979) have found that the relationship could be described equally well by a power or linear function, noting that the latter is a special case of the former. Still others (Weiss, 1981) have argued that no universal psychophysical law exists and that the relationship between perceived and actual quantities is determined by the specific context of the experiment. In light of the discrepancy in the literature, a secondary objective of this experiment was to evaluate the mathematical relationship between perceived and actual exertions.
METHODS
To achieve the stated objectives, we performed an experiment in which participants estimated the exertion intensity required to just budge a stationary handle on a work simulator. Verbal estimates were compared with the known resistance levels, which ranged between 0% and 100%, of the participant's static, maximum voluntary contraction (%MVC). The experiment consisted of testing three participant groups: one control (no-benchmark) group and two other groups, which varied in the number of reference benchmarks to which they were exposed prior to estimating the intensity of submaximal exertions. Each group participated in two data collection phases, although the sequence of these phases varied by group. In the first phase, standardized maximum static strength measurements were collected. In the second phase, participants verbally estimated submaximal exertion intensities. Prior to exertion estimation, one group performed an additional phase in which they were exposed to two additional resistance benchmarks.
Equipment
Equipment consisted of a work simulator and a data acquisition system to record measurements. The simulator (Baltimore Therapeutic Equipment Company, 1992), shown in Figure 1, measures and controls force produced at the...
|
