Search

Most experiments on monitoring behavior have employed visual signals. Because the visual modaliry is frequently pre-empted by task requirements, the characteristics of vigilance warning signals for non-visual modalities should be ascertained. Broadbent (1958) has suggested that there may be less deterioration of performance in monitoring auditory signals than in monitoring visual ones, as it is harder to not listen to a noise than not to look at a visual cue. The cutaneous channel also might have advanrages in that it is less likely to be occupied by other task requirements than is either the visual or auditory channel. While there have been very few vigilance experiments involving cutaneous cues, there have been a number employing auditory ones. In general, it appears that when 0's task was to report the presence or absence of an auditory signal, no decrement as a function of time on task was observed (Elliott, 1957; Martz 8: Harris, 1961 ). When changes in auditory signals were to be detected, a progressive deterioration of performance was reported (Solandt & Partridge, 1946; Mackworch, 1950; McGrath, 1960; Buckner, Harabedian, 8: McGrath, 1960). A similar generalizarion can probably be made regarding monitoring of visual signals (see Broadbent, 1958). Some recent experiments have reported conflicting results, however. The experiments by Elliott and by Martz and Harris, cited above, involved an ascending method of limits rather than a simple derection of the stimulus. In experiments by Loeb and Schmidt ( 1960a, b), which required Ss to respond rapidly to the presence of faint [lo db, sensation level (SL)] low frequency pure tones, there was a small but significant progressive increase in response time to the signals. Similarly, McCormack (1959) has reported a progressive increase in response time for detection of the presence of faint visual signals. In a more recent experiment (Hawkes & Loeb, 1961) involving response to slightly louder bursts of white noise (of rapid onset and decay) no progressive increase was exhibired. Whether the discrepancy in the findings in these experiments was due to differences in intensity, rise time, or spectrum of the signals was not established. In the latter experiment (Hawkes & Loeb, 1961), 0s also were required to respond to electrical cutaneous stimuli. Interestingly, though the electrical 'This investigation was aided by a contract between the Office of the Army Surgeon General and the University of Louisville. Appreciation for help in data colleaion is expressed to D. Kaufrnan, C. Roulston, C. Gettys, and E. A. Schmidt, and for administrarive support to R. H. Bixler.

Ten Ss made absolute judgments of electrocutaneous stimuli consisting of from one to three simultaneous stimulations of six loci. There were three loci in symmetrical positions on either side of S's body—one on each shoulder blade, one on each arm between the elbow and the shoulder, and one on each side of the body just above the belt line. Each S responded with one of two response alphabets during 11 sessions of 41 trials each, over a period of three days. Differences between the two response alphabets were not significant, and neither were differences attributable to the interaction of the response alphabets with the number of loci stimulated. Increases in the percentages of erroneous responses with increases in the number of loci stimulated were both large and statistically significant; these increases were interpreted as evidence of a central mechanism of cutaneous masking.

(96) Ray E. Trussell, Jack Elinson, and Morton L. Levin, "Comparison of Various Methods of Estimating the Prevalence of Chronic Disease in a Community," American Journal of Public Health, 46 (February, 1956), 173-182. (97) Urban Life Research Institute, Nursing Services in a Large Urban Hospital, New Orleans: Tulane University, 1954. (98) L. Wes Wager, "The Measurement of the Job Satisfaction of Professional Nurses," Research Studies of the State College of Washington, 20 (June, 1952), 45-55.