National Academies Press: OpenBook

Video Displays, Work, and Vision (1983)

Chapter: 7. Visual Tasks, Functions, and Symptoms

« Previous: 6. Anthropometry and Biomechanics in VDT Applications
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 143
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 144
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 145
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 146
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 147
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 148
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 149
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 150
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 151
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 152
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 153
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 154
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 155
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 156
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 157
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 158
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 159
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 160
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 161
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 162
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 163
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 164
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 165
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 166
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 167
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 168
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 169
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 170
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 171
Suggested Citation:"7. Visual Tasks, Functions, and Symptoms." National Research Council. 1983. Video Displays, Work, and Vision. Washington, DC: The National Academies Press. doi: 10.17226/169.
×
Page 172

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

7 Visual Tasks, Funchons, and Symptoms This chapter addresses several interrelated questions: Are the visual complaints of VDT workers qualitatively or quantitatively different from those of non-VDT workers performing comparable near-visual tasks? What aspects of VDT work and non-VDT work might influence visual symptoms? Are there unique features of VDT equipment or job tasks that affect the comfort and visual performance of workers? What does the often-used term 'visual fatigue" mean, and does it represent a useful concept? VISUAL ISSUES IN VDT STUDIES Field Surveys Visual Complaints Field surveys of the vision of VDT workers have received widespread attention in both popular literature (e.g., New York Committee for Occupational Safety and Health, 1980; Working Women, National Association of Office Workers, 1980; Dehiatteo et al., 1981; Working Women Education Fund, 1981) and technical literature (e.g., Grandjean and Vigliani, 1980) and have been the basis for concern about the well-being of VDT workers. A majority of surveys have reported that more than 50 percent of VDT workers indicated that they at least occasionally experienced some type of ocular discomfort (irritation, pain, or fatigue involving the eyes) or blurring or flickering of vision (Cakir et al., 1978; Gunnarsson and Soderberg, 1979; Coe et al., 1980; Elias et al., 1980; Gunnarsson and Soderberg, 1980; Rey and Meyer, 1980; Smith et al., 1980; Arndt et al., 1981; National Institute for Occupational Safety and Health, 1981; Sauter et al., 1981; Smith et al., 1981~. In studies that included comparison groups of non-VDT workers, a majority have reported that complaints 143

144 related to vision were more prevalent in the VDT subjects (Laubli et al., 1980; Rey and Meyer, 1980; Smith et al., 1980, 1981; Arndt et al., 1981; National Institute for Occupational Safety and Health, 1981; Sauter et al., 1981~. Coe and coworkers (1980) reported that while the prevalence of fatiguelike effects of visual discomfort was higher in VDT workers than in comparison groups, the preva- lence of irritantlike effects and other subjective eye complaints was similar in the two groups. Some surveys have reported that the prevalence of complaints of ocular discomfort increased with increasing percentage of work time spent working with VDTs (Rey and Meyer, 1980; Dainoff et al., 1981) and was higher in full-time than in part-time workers (Coe et al., 1980~. Some studies have reported that the preva- lence of complaints varied with the type of VDT work performed (Coe et al., 1980; Elias et al., 1980; Smith et al., 1981; National Institute for Occupational Safety and Health, 1981~. Several aspects of the designs of most of these surveys are problematic and limit the inferences that can be drawn from them (see Chapter 2~. The comfort and visual performance of VDT workers are important concerns. Unfortunately, however, studies have not established which factors in VDT work--e.g., image qual- ity, lighting conditions, workstation design, visual task require- ments, visual status of workers, job design, and psychosocial variables--are correlated with visual complaints, and no conclu- sions can be drawn as to whether reported symptoms can be attributed to use of VDTs per se, to other variables not unique to VDT work, or to some combination of these factors. The visual complaints of VDT workers described in the field studies appear to us to be qualitatively similar to complaints reported to clinicians by people of all ages and a wide variety of occupations. However, to detect subtle differences, if any, in the character of visual symptoms among groups of workers, clinical examinations rather than field surveys would be required. Measurements of Visual Status In addition to questionnaires on physical symptoms, some field surveys have also included routine measurements of the visual status of VDT workers and comparison groups of workers. Dainoff (1980) and Dainoff and coworkers (1981) found no significant differences in visual acuity or in lateral and vertical phoria measured before and after work on two groups of subjects who worked with VDTs for an average of 47 percent and 75 percent of their working time. Dainoff and coworkers reported that no obvious relationship was found between measures of visual status in individual subjects and subjective complaints of ocular dis-

145 comfort; their analyses, however, are not described. Coe and coworkers (1980) measured visual acuity, astigmatism, near phorias, amblyopia, and color blindness on four groups of VDT workers performing different tasks (input, editing, question~nd- answer, creative) and comparison groups of non-VDT workers: no differences between VDT and non-VDT workers were found on any of the measures except for acuity. The non-VDT workers had a higher prevalence of acuity defects, but this difference cannot be interpreted because none of the groups was measured for acuity before workers began their current job. Thus, in contrast to the findings of field surveys regarding visual discomfort and diffi- culties with seeing, defects in visual status have not been reported to occur more frequently among VDT workers than among non- VDT workers (see also the following section and the discussion later in this chapter). Experimental Field and Laboratory Studies of Visual Functions in VDT Works Several investigators have attempted to determine whether various visual functions change during performance of VDT tasks; the investigations have covered periods of up to several hours. In some cases, attempts have also been made to compare measures of visual function during VDT and non-VDT tasks. These studies have concluded that transient changes in accommodation (Haider et al., 1980; Ostberg, 1980; Ostberg et al., 1980; Mourant et al., 1981), convergence (Gunnarsson and Soderberg, 1980), and appar- ent visual acuity (Haider et al., 1980) occur during the use of VDTs. These changes have been asserted to be evidence of 'visual fatigue" and have been attributed by the investigators to unusual demands placed on the visual system during video viewing. Two of these studies (Gunnarsson and Soderberg, 1980; Haider et al., 1980) obtained questionnaire data on subjective visual complaints in addition to measuring visual function; however, the relationship between visual complaints and measured changes in visual function apparently was not assessed. Gunnarsson and Soderberg (1980) measured changes in the near points of accommodation and convergence before, during, and after work in groups of VDT workers under normal and intensified work conditions. They report that some of the subjects showed significant changes in the near point of convergence from the beginning to the end of the workday under the intensified con- dition. The authors concluded that the study shows changes in the - iThese studies are discussed in more detail in Appendix A.

146 near point of convergence to be a useful objective measure of visual strain associated with VDT work. Haider and coworkers (1980) measured visual acuity at 4 m in VDT workers performing a test protocol in a work setting and in a comparison group of non-VDT workers performing normal office work (primarily typing). They reported that a temporary reduction in visual acuity occurred following work in the VDT group while the comparison group showed almost no change in acuity. The authors referred to the reduction in acuity as "temporary myopization't resulting f ram "accommodation strain." {;)stberg (1980) and Ostberg and coworkers (1980) measured accommodative response and dark focus before and after work in three groups of VDT workers - group of air traffic controllers and two groups of telephone office VDT workers--each of which performed different types of tasks. Ostberg and coworkers reported a shift in dark focus and a reduction in accommodative response in the air traffic controllers and similar but smaller changes in the other two groups. They attributed these results to 'visual fatigue" associated with VDT work; the larger changes in the air traffic controllers were attributed to the greater visual demands of their work. Nlourant and coworkers (1981), in a laboratory study,, measured the time taken to focus from a near to a far target (referred to as outfocus time) and back to the near target (referred to as infocus t_) as a function of the type of visual display used, CRT or hard copy, and time on the task. The authors reported that both out- focus and infocus times were higher for subjects when viewing the CRT and that, for both types of display, times increased as a func- tion of time on task. The authors interpreted these results as evidence of fatigue in the accommodative or eye movement sys- tems, or both. They concluded that use of a CRT has a "measur- able fatigue impact on the visual mechanisms" and that this effect is greater for viewing VDTs than for viewing hard copy. Our analysis of these experimental field and laboratory studies indicates that, because of various aspects in the design, analysis, and interpretation of data, the results cannot be interpreted with any certainty. All of the studies used samples of opportunity; little information was presented on demographic variables and on the working environments of the different groups examined in the studies, despite their possible effects on outcomes. Appropriate control groups were not used; thus it is not possible to attribute the results of the studies specifically to VDT work. While Haider and coworkers used a comparison group, that group did not per- form a task comparable to that of the VDT group, and they presented no information on how the two groups compared on other variables. No information was presented in any of the

147 studies on the visual status of the subjects prior to the experi- ments or on the range of normal variability the subjects might be expected to exhibit; such information is, however, critical to evaluating changes in measures of visual function such as those reported in the studies. The field studies compared groups whose VDT tasks were not comparable; thus it is possible that the effects of task variables were confounded with the effects of the VDT. The techniques used in some studies were subject to such influences as general fatigue and motivation of the subjects. Although all of the studies reported results as statistically significant, little or no information was provided on the statistical analyses performed and the level of significance reached. Furthermore, the assertion in these studies that the reported temporary changes in measures of visual function are evidence of "visual fatigues' is based on the assumption that changes in these measures of visual function reflect changes in the fatigue state of the oculomotor system; this connection has not been scientifically established, despite an enormous amount of research on the issue over the past 50 years. Workers have been concerned that using VDTs might damage their eyes; however, our analysis (see below) of the temporary changes in measured visual function reported to follow VDT work, as well as non-VDT visual tasks, finds no sugges- tion of damage (in the sense of long-term irreversible anatomical or physiological changes) to the visual system. However, there does not seem to have been any research specifically to this point. The Need for Job and Task Analysis VDTs are used in many occupations, and jobs involving VDT work are highly diverse (see Chapter 1~. VDT jobs differ on many dimensions: for example, in the visual tasks involved, the time spent viewing the VDT, the familiarity of the information processed, and the work schedule. These factors shape the visual requirements placed on a worker and presumably affect the likelihood that visual and other complaints will occur. Unto r- tunately, this diversity has seldom been considered in VDT studies. There has not yet been any well~esigned research showing that there are characteristic sources of discomfort unique to VDT work, and the visual tasks characteristic of VDT jobs and non-VDT jobs have not been systematically compared. Any useful attempt to determine which factors may contribute to visual discomfort should include an analysis of job and task features. Such an analysis would consider how the eyes must be used to perform the task required. For example, in some data entry jobs the worker may only occasionally be required to view

148 the video screen. In such jobs the primary visual task may be reading hard copy from a document holder, a task also performed in many non-VDT jobs. Any visual discomfort experienced in this case would probably not be caused by the VDT itself. In contrast, some VDT jobs require prolonged viewing of the video screen, and many jobs require repetitive eye and head movements between the video screen and hardcopy materials. A systematic research program would be required to determine the relative effects of display characteristics, hardcopy characteristics, and other workstation features on the comfort and performance of workers in particular jobs. Psychosocial factors also may affect the experiencing and reporting of symptoms of visual discomfort. Workers in some jobs are allowed sufficient flexibility to temporarily modify their work pattern if they experience visual discomfort. Other jobs, how- ever, are much more rigid in time and activity requirements. It also seems likely that job satisfaction will affect workers' toler- ance for discomfort and their reporting of symptoms. The quality of VDT workstation design varies greatly even within given job categories. Worker comfort and performance can be affected substantially by display quality (Chapter 4), lighting and reflections (Chapter 5), and anthropometric features (Chapter 6~. Thus, job analyses should include consideration of the quality of these factors in individual workstations. Some problems with workstation features and visual task requirements are clearly not unique to VDT jobs: for example, workers in various non-VDT jobs—secretaries, editors, teachers, et=--must read hard copy or handwriting of poor legibility. In some VDT jobs, workers must deal with both poor-quality video displays and poor-quality hard copy, and they may experience cumulative effects in task dif- ficulty and the effort required. Are There Unique Features of VDT Tasks? Textual characters generated by VDTs differ in several respects from hard-copy characters. The effects of some of these display parameters on reading performance are discussed in Chapter 4. The self-luminous nature of a VDT and the transient presentation of material are obvious differences. Noticeable differences may also occur in contrast, sharpness of characters, image stability, and other features. Murch (1982a) argues, on the basis of a com- parison of accommodative response to CRTs and hard copter, that CRT displays generally do not provide a visual stimulus capable of

149 evoking optimal accommodation; the data he reports, however, do not seem to support this conclusion.2 VDTs typically use simplified fonts, chosen for engineering convenience, that contrast with the hard-copy fonts that have evolved through applied research and many years of practical experience. Image quality of a given display will in part deter- mine how well it compares to hard copy in terms of worker performance and comfort, but the quality (especially legibility) of hard copy is not always good either, especially in cases of photo- copies, carbons, and computer printouts. Problems with lighting and reflections are, of course, encountered in reading hard copy as well as in viewing VDTs, but some aspects of VDTs can be particularly problematic. For example, it is possible that users' accommodation and conver- gence may fluctuate between the characters displayed on the screen and images reflected on the screen (which are at a differ- ent optical distance), and this may adversely affect comfort and performance (see Chapter 5~. VDT workstation design can also present difficulties similar to but more pronounced than those encountered in traditional clerical workstations. For example, VDT workstations may impose additional postural constraints and requirements for head and eye movements (see Chapter 6~. Both visual and nonvisual aspects of VDT work are shaped by the character of the person-computer interaction. Business forms and text have evolved through centuries of use, while VDT for- mats have been designed primarily in accordance with program- mers' notions as to what will be efficient or convenient to the user. Research on VDT format design and its effects has barely begun (Granda, 1980; Kolers et al., 1981~. It is important to note that the software of a computer program sets the stage and determines the rules for interaction. Employees who function as lower-level users of a computer system play an especially sub- 2Murch used a laser optometer to measure the immediate accommodative responses of subjects to a pattern of X's displayed at a distance of 50 cm on six types of VDTs and on negative contrast hard copy. The accommodative response to each display varied only slightly from the ideal value of 2.0 diopters, ranging from 1.84 to 2.14 (the value for hard copy was 2.01), and all fell within the expected 0.25 - iopter depth of field expected under the measurement conditions of the study. No statistical analyses are reported, but the differences in measured refractive status do not appear to be significant. It is arguable whether the underlying premise of this study, that the accuracy of focus in response to a display provides an index of visibility for that display, is reasonable.

150 servient role to its programming--a role that requires them to place strong dependence on the entire computer system to carry out their jobs. There is, however, growing interest in designing "user-friendly" computer systems (see, e.g., Association for Computing Machinery et al., 1982~. In contrast to hard copy, VDT text is often active and trans- ient in nature. For example, text may be scrolled? moved up the screen to expose new material at the bottom and cover up pre- ceding material at the top. The few studies of scrolling under- taken to date (e.g., Kolers et al., 1981; Sekey and Teitz, 1982) do not suggest a problem. Kolers and coworkers found little differ- ence in observed eye movements between reading a static display and scrolled text, especially when the scrolled display was rapidly paced. The effect of scrolling on worker comfort over prolonged periods, however, has not been examined. Adequate research that would establish whether there is any- thing inherent in VDT tasks that can unavoidably cause visual difficulties not encountered in comparable non-VDT tasks has not been conducted. We suggest, however, that attention to quality of display design, control of lighting and reflections, and appro- priate job design (including display format and structure of operator-VDT interaction) would go a long way toward preventing difficulties (see Chapter 9~. Research on the effects of VDT work on the visual comfort and performance of workers should be designed to distinguish which factors (e.g., observable jitter, flick- er, screen-reflected images, poor design and format) contribute to which effects and whether problems can be reduced by changing display parameters. Without such research, development of effective displays and comfortable interaction conventions will depend on the slow and uncertain processes of evolution by trial and error. The Special Task of Reading Reading is a central task in most VDT and comparable non-VDT jobs. Display characteristics that affect short-term reading performance are discussed in Chapter 4. There has been little research systematically comparing sustained reading (for several hours) from VDTs and from printed materials. Bagnara (1980) found a decrease in performance in an error-detecting task over two hours of reading short news articles from VDTs, but the study failed to counterbalance the materials, it used only six operators, and- most important for the present point--it did not separate effects due to the reading task per se from those due to the VDT per se. Several studies have reported temporary changes in oculomotor function that were argued to result specifically from

151 VDT work; the findings of these studies are critically examined later in this chapter and in Appendix A. If significant oculomotor changes, decrements in reading performance, or visual discomfort were specific to VDT work (which has not been established), one would want to know what aspects of reading were affected by VDT characteristics. Appro- priately designed research is needed to compare the effects of different kinds of reading tasks on the visual-motor system and to examine the interaction of those effects with specific character- istics of VDTs. (The aspects of the reading task that would need to be considered are discussed below, and the relevant video display characteristics are discussed in Chapter 9.) The reading task requires a viewer to discriminate some or all of the symbols or strings of symbols from others of the set of symbols that might have been presented (for example, to distin- guish the letter A from other letters of the alphabet). This dim crimination task requires that a reader's eyes maintain some minimum degree of focus or accommodation and be directed so that the part of the array about which information is needed projects to the fovea of the eye. Visual acuity falls off very rapidly outside the fovea: only about four individual letters can be read as such without moving the eye, although characteristic word forms and lengths can, when they constitute meaningful text, convey some information for considerably longer distances from the fovea (up to 12 letter spaces) (see Woodworth, 1938; McConkie and Rayner, 1975, 1976~. Muscular activity is therefore needed t maintain accommodation and to move the eye to new fixations (through eye movements called saccades, discussed below). The fineness of detail that must be resolved varies greatly with the nature of the display itself and with the task. With dense and nonredundant text, a reader may need to resolve every distinctive feature of each symbol, while with highly redundant or familiar material, much less detail is needed to perform the same task. One reading task (e.g., proofreading) may require much more detailed vision than another (e.g., skimming for gist). The pre- cision of accommodation and the number of saccades that are needed to perform any task with any given body of text depend on the legibility of the display, the meaningful structure of the text, the nature of the reading task, and the reader's skills. The effects on reading performance of such variables as font, letter and word spacing, letter size, organization of materials on a page, contrast between letter and background, and reflectance or texture of the paper have been studied to some degree. The largest effects under normal circumstances are associated with the content of the material and its relation to the cognitive task, i.e., the variables of cognitive organization (Breland and Breland,

152 1944; Paterson and Tinker, 1956; Tinker, 1965; Bou ma, 1980; Engel, 1980; Timmers et al., 1980; Treurniet, 1980~. According to this conception of reading, tasks that involve low levels of cognitive demand allow a great deal of visual imperfec- tion: a lot of redundant information in a text allows a reader to miss a great deal and still "see" everything needed for the task. For more difficult reading tasks, a viewer is forced to process at the word-by-word or letter-by-letter level, and there is a greater cognitive load. Such a lack of redundancy places great demands on attention, which in turn requires both sustained concentration and detailed vision. The tasks that might create such cognitive load when performed under pressures for speed and accuracy include data entry or acquisition of nonredundant and unfamiliar materials. Such tasks would place more stringent demands on the discrimination of detail throughout the text. Although the rate of saccades does not appear to change much with task difficulty (Judd and Buswell, 1922), the saccades must be smaller and more precisely placed so that the fixations are closer together, and the emphasis on discriminating individual letters may place more stringent demands on the accommodative mechanisms. The Problematic Concepts of "Visual Fatigue" and "Eyestrain" The literature on visual complaints and symptoms of VDT workers contains many references to "visual fatigue" and "eyestrain," but these terms are often ambiguous. Carmichael and Dearborn, in their classic treatise Reading and Visual Fatigue (1947), discuss the topic in the context of problems . encountered in research on fatigue in general. They consider fatigue to have three aspects: the subjective experience of an individual performing prolonged work; changes in task perfor- mance over time; and physiological changes. In studies of these three aspects, there has been little success in relating changes in physiological parameters observed in fatigue experiments to either task performance or subjective aspects (Cameron, 1973; Smith, 1979~. The Carmichael and Dearborn study illustrates many of the issues encountered in research on effects of prolonged perfor- mance of visual tasks. They found no decrement in reading comprehension or eye movement performance (number of fixations and regressions) over a six-hour period among high school and college subjects reading hard-copy or microfilm text. Many of the subjects reported increasing tiredness or general discomfort over the six-hour period, but it appeared to be related more to the contraints on posture and activity than to the visual task. Apparently there were only a few ocular or visual complaints

153 (the subjects were not questioned specifically about visual or ocular symptoms but were asked to describe their subjective responses in general). The results of the Carmichael and Dearborn study contrast with the finding by Hoffman (1946) of decrements in reading performance by students over a four-hour period. Carmichael and Dearborn attributed the difference to motivation. Their study, unlike Hoffman's, employed incentives to maintain a high level of reading performance. Some studies have reported that highly motivated subjects can maintain effective task performance (e.g., driving or flying) for many hours even though they experience symptoms of subjective fatigue (e.g., tiredness, discomfort, sleepiness, reluctance to continue work), while other studies have reported decreased performance of tasks in periods as short as two hours (see Cameron, 1973~. The discrepancies among these studies may be a function of the increasing effort required to maintain performance over time, the level of arousal evoked by the partic- ular tasks, and the motivation of the subjects. According to Duke-Elder and Abrams (1970:559), "Eyestrain may be defined as the symptoms experienced in the conscious striving of the visual apparatus to clarify vision by ineffectual adjustments" (see also Borish, 1970~. Other authors (e.g., Ferguson et al., 1974) have objected to the term "eyestrain," which might suggest damage to the eyes. Asthenopia (used in general to refer to any subjective visual symptoms or distress resulting from use of one's eyes) may be a more useful term. Duke-Elder and Abrams classify the symptoms, which are quite varied, as visual (especially blurring), ocular (the eyes feel tired, hot, uncomfortable, or painful), referral (e.g., headaches), and functional (behavioral). The ocular discomfort is said to be due to muscular fatigue, but the mechanisms are not known. Duke-Elder and Abrams empha- size the role of higher perceptual processes in tiring from con- tinuous effort to interpret blurred and indistinct images. The causes of asthenopia are described as environmental (involving illumination, the nature of visual tasks, and the characteristics of the objects viewed), ocular (the patient's visual status), and constitutional (involving both physical health and emotional state). Refractive Errors and Visual Difficulties There are a number of clinical conditions involving small uncor- rected refractive errors and oculomotor imbalances that can cause visual difficulties with prolonged near work or critical detail work. When a person attempts to rectify those refractive errors or imbalances by continuous muscular effort, symptoms of asthenopia can develop. Such symptoms can be caused by uncor-

154 rected ametropia, accommodative difficulties (see also the next section), heterophoria, convergence difficulties, fusional inade- quacy, or aniseikonia. It is possible that some of the visual complaints of some VDT workers (as well as those of non-VDT workers) naight involve such clincial conditions. In such cases, a careful clinical examination would be useful, and refraction may relieve the symptoms. Myopia Myopia (nearsightedness) is an extremely common abnormality of ocular refraction occurring in varying degrees in almost one- quarter of the adult population of the United States (Roberts, 1978~. Myopia is less common in populations that are less technologically advanced (Taylor, 1981), and it is more common and of greater degree in adults than in children (Sorsby, 1972~. For these reasons it has been suggested that excessive use of the eyes for near work is the cause of myopia. There is, however, considerable controversy about the interaction of near work and myopia (Bear and Richler, 1982; Taylor, 1982~. As we discussed above, it is not clear whether the characteristics of VDTs or of the visual tasks performed in working with them differ from characteristics of other forms of near work with regard to a possible interaction with ocular refraction. OCULOMOTOR FACTORS AFFECTING VISUAL PERFORMANCE As discussed above, the concept of visual fatigue is one that has escaped canonical scientific definition, and there has been little progress in relating subjective measures of fatigue to perfor- mance of visual tasks or physiological functions. Several inves- tigators have measured oculomotor functions during prolonged performance of visual tasks (including VDT work) with the hope of finding objective correlates of visual complaints or even a phsyio- logical basis for those complaints. Recently, some investigators have argued that temporary changes in oculomotor functions reported to follow VDT work are evidence of visual fatigue. This argument is problematic for several reasons (some of which were discussed above>. the decrement in motor performance response to a stimulus may be due to adaptation of the sensory organs, habituation of central nervous system activity, or decline in motivation, as well as to true muscular fatigue. The effects of subjective (general) fatigue may in some cases be reversed by increased motivation. Efficiency, the ratio of response to effort,

155 has been suggested to be a more appropriate measure of fatigue than response decrement, but of course, efficiency—like subjec- tive fatigue—is difficult to measure (Cameron, 1973~. Further- more, observed oculomotor changes might in some cases be nor- mal adaptive responses to sustained work rather than indications of declining performance. In this section we discuss oculomotor factors that have been studied during sustained performance of visual tasks. The general functions of the oculomotor system are described to provide a context for considering the temporary changes that have been reported to follow sustained visual performance. We discuss the difficulties in relating these reported oculomotor changes to com- plaints of visual discomfort, and we review studies that have used visual tasks such as reading printed materials and microfilm, inspection tasks, tracking tasks, and, more recently, reading from VDTs. Eye Movements Eye-movement changes during sustained performance of visual tasks may be considered at several different levels of extraoculo- motor function: saccadic trajectory, dual-mode tracking move- ments, higher-level eye movements, and gaze and the vestibulo- ocular reflex. Saccadic Trajectory The eye (specifically, the fovea, the high-resolution, color- sensitive portion of the retina) is directed rapidly from one object of regard to another by saccadic movements. The performance of saccades has been examined in considerable detail, anatomic (Carpenter, 1977; Graybiel, 1977; Keller, 1977), physiologic {Bach-y-Rita, 1971; Evinger et al., 1977~9 and control system theoretic (Stark, 1968, 1971; Robinson, 1981~. A change of fixation of the eye involving a saccade is caused by a burst of neural activity to the extraocular agonist muscles. The activity rises to a peak value and then settles back to a lower plateau (Robinson, 1964; Fuchs and Luschei, 1970; Collins, 1975), a pattern sometimes called a pulse-step. Hering's law of equal innervation requires that the same pulse-step be applied to the corresponding muscles of both eyes. In normal circumstances the trajectories of saccades are stereotypical, and if certain parameters of the saccades (e.g., peak acceleration and peak velocity) are plotted against each other, a smooth curve called the main sequence is obtained.

156 Surprisingly, at least in adults, there is remarkably little variation in the main sequence (Bogen et al., 1974; Able et al., 1979), and it can therefore be used as a yardstick to measure changes in per- formance. The fine neurological control structure of the saccadic trajectory has been shown to decompose rapidly (within minutes) in repetitive trackings situations in at least two very different ways. First, central nervous system habituation allows the appear- ance of double saccades, each component of which falls naturally on the main sequence (Bahill and Stark, 1975~. The fact that these fit the main sequence provides evidence that even lower-level pulse-shaping neurons in the brainstem (as well as, of course, neuromuscular and muscular factors) are functioning quite nor- mally. Thus, it is clear in this instance at least that a well- documented appearance of oculomotor fatigue has been definitely localized to the central nervous system. This is especially important because increased motivational effort is known to compensate for fatigue processes; thus, it is often logically difficult to prove that the fatigue process does not lie in the peripheral neuromuscular apparatus (as one might expect). Also, pulse-step mismatches in the controller signal envelopes of neuronal bursting and tonic firing of nerve impulses occur and result in "glissades," which often violate Hering's law of equal innervation to corresponding muscles (Bahill and Stark, 1975; Bahill et al., 1976~. Second, the more peripheral neuromuscular fatigue type of decomposition of the saccadic trajectory produces generally slowed movements below the main sequence velocity~mplitude relationship and also irregular fragments showing erratic profiles of velocity and acceleration. These fragments have been found in patients with peripheral neural and neuromuscular diseases (Feldon et al., 1982) and lie above the main sequence as a result of "truncation" (similar to the truncation that occurs in overlapping voluntary nystagmus saccades) (Stark et al., 1980~. 3Reading VDT displays is largely performed by repetitive saccades at a rate of approximately four saccades per second. Thus there are similarities and differences between reading eye movements and repetitive tracking eye movements. Students of reading eye movements use tracking movements as a related but simpler form of sequential reading eye movements. Indeed, Pavlides (1981) has suggested that a deficiency in tracking accompanies abnormalities in reading eye movements (see discussion of reading eye move- ments above).

157 Dual Mode Tracking Eye Movements A second level of eye movement performance has to do with both smooth pursuit and saccadic eye movements. In the continual activity of moving one's eyes, an interactive dual-mode control system acts to turn the eyes conjugately Binocularly) throughout the visual field. In order to study these spontaneous movements, physiologists and bioengineers have developed a tracking task.--very similar to a test of skill in a penny arcade, where one tries to aim at a moving target- in which the task is to aim the eyes quickly and accurately. In this test a single entire saccadic response may drop out, and the sequence of responses then continues in phase with an alternating jumping target (Lion and Brockhurst,1951~. In addition, the tracking quality itself may deteriorate with gradually increased reaction times (minutes) and abnormal sequences of saccades (Stark et al., 1962~. Saccadic smooth pursuit may also occur as a compensatory adaptation when the gain of smooth pursuit decreases.5 Visual feedback is a crucial element in the control of eye movements at this level. Higher-Level Eye Movement Patterns Reading A third level of eye movement performance comprises reading and looking, which are composed of sequential patterns of saccadic eye movements. The reading task is complex, with cog- nitive, linguistic, visual, and eye movement factors all interacting (Huey, 1908 t19683; Woodworth, 1938; Hochberg, 1970; Smith, 1971; Fisher, 1975; McConkie and Rayner, 1975; Just and Carpen- ter, 1978; Rayner, 1978; Scinto, 1978; Stern, 1978; O'Regan, 1979; The tracking task is an excellent paradigm for human factors experimentation to define operator activities, for example, in controlling vehicles. In regard to the eye movement control system, the tracking task has been studied with the conceptual and mathematical tools of the control engineer (Fender and Nye, 1961; Young and Stark, 1963~. So far this methodology has not been applied to VDT use, but it is a very powerful tool for the quantitative study of task performance and its deterioration with sustained activity. Future studies of fatigue with VDT use would benefit from application of this methodology. It is important to note again Cameron's notion of efficiency: effort may compen- sate for decreased response and may maintain performance in highly motivated subjects (Cameron, 1973~. s Restoration may occur after a brief rest period (tens of seconds) or a startle dishabituation stimulus.

158 Spiro et al., 1980). Many studies have demonstrated poorer reading performance with les~than~ptimal typography or illumination (Tinker, 1939; Tinker and Paterson, 1939; Demilia, 1968~. However, several experienced and careful researchers have found no significant decrement in reading performance after six hours of reading concentrated material (Carmichael and Dearborn, 1947; Carmichael, 195~1952~.6 The eye-movement component of the reading task involves mainly a sequential pattern of saccades, about one for each word, although small common words (a? of the) are often skipped.7 About 10 to 20 percent of the saccades are regressions or backward saccades, generally to a linguistically critical word in the syntax of the sentence. Indeed, the French word verification for a regression assumes this linguistic role. At the end of a line, a larger saccade sweeps the eye to the left margin of the next line, and numerous corrective saccades occur, especially vertical saccades to place the eyes correctly on the line. Large saccades, such as occur when a user looks from a dark VDT to illuminated hard copy, may provide a light stimulus to the pupil and may also quickly change focal distance and stimulate an accommodative response. Vergence and accommodative response have not been thoroughly studied during reading. Although Carmichael and Dearborn (1947) reported that subjects who had spent several hours reading such concentrated material as Adam Smith's Wealth of Nations reported feeling fatigued, 60 of __ . . the 80 subjects fe t they could continue to read if provided a rest. 7Reading rates of 300 words per minute are near maximum. The maximum rate of saccadic generation is 5 per second, which is equivalent to a reading rate of 300 words per minute (Adler- Grinberg and Stark, 1978~. (The extra number of regression saccades is roughly equated to the number of skipped small words.) This high frequency of saccades must be considered as a possible source of interaction with the flicker produced by VDT regeneration rates (David Bridgeman, Department of Psychology, University of California, Santa Cruz, personal communication, 1 981~. Many studies have shown that poor readers with or without specific dyslexia often have irregular reading eye movement patterns and prolonged fixations. These and backward staircases have been found in peripheral motor dyslexia secondary to the diplopia of a peripheral oculomotor paresis. One may speculate that minor forms of dyslexia in VDT operators may surface and produce complaints.

159 Picture Scanning Some VDT users examine graphs or charts, and engineers engaged in computer-assisted design (CAD) typically work with orthographic projections or perspective displays of machine components. It is possible that some of these workers examine diagrams with sequences of eye movements that have been dubbed scanpaths. The scanpath, which is observed under certain laboratory conditions, is a nonrandom sequence of sac- cadic eye movements that scans a picture in an idiosyncratic pattern for each subject. Scanpaths appear to be generated by internal cognitive models in a checking phase of pattern recog- nition. Recent evidence for this comes from the work of Stark, Noton, and Ellis who used ambiguous and fragmented figures and asked subjects to signal their alternating perceptions (Noton and Stark, 1971a, 1971b, 1971c; Stark and Ellis, 1981~. With the same subject and the same physical pictures, the eye-movement scan- path developed different patterns, each associated with a par- ticular perception. Scanpaths are modified with continued examination of the same pattern (Yarbus, 1967), but the implica- tions of this finding are not known. Minor changes in scanpaths may not be noticed.8 Gaze and the Vestibulo~cular Reflex The vestibulo-ocular reflex is a primitive phylogenetic mechanism that stabilizes the direction of gaze in space. In the brainstem there are connections between the oculomotor system and the vestibular apparatus that monitor balance and posture with respect to gravity. If a head movement occurs, a signal that generates a compensatory eye movement (in the absence of volition) is sent from the vestibular apparatus to the oculomotor system. Thus, when looking at a visual target, a person may turn his or her body or head without losing visual contact (Baloh and Honrubia, 1979; Wilson and Jones, 1979; Henn et al., 1980~. Coordinated gaze movement normally has an initial eye-in- orbit saccade onto the target followed by a synkinetic and much slower head movement. At the level of the electromyographic (EMG) signal, latencies are synchronous, but because the visco- ~Drivers engaged in long, continuing driving tasks show central concentrations of eye fixations rather than normal, widespread, active looking and searching patterns (McDowell and Rockwell, 1978~. The question of preview control, how far ahead the driver looks, may be important to the concentration of fixations by fatigue (Tomizuka and Whitney, 1975~.

160 inertial dynamics of head and neck muscles are different from the viscoelastic dynamics of eye and extraocular muscles, the saccade is over before the head position has changed. The vestibulo-ocular reflex generated by head acceleration drives the compensatory eye movement, eye-in-orbit, in the opposite direction so that gaze, eye-imspace, remains on target. Thus, gaze movement has the advantage of the rapidity of the eye saccade, and at the end of the movement the eye is in primary position in the orbit and the target is in the straightforward position, since directional sense is determined by head position. Diseases of the oculomotor system, the vestibular apparatus, or the central nervous system may cause incomplete visual com- pensation for body movement. In such circumstances a disequi- librium may result because visual space does not correspond with the subject's egocentric image of space. This sensation results in nausea, vomiting, and vertigo. The same symptoms may occur in the absence of disease if a person uses psychotropic drugs such as tranquilizers, sedatives, or alcohol (Rashbass and Russet, 1961; Wilkinson et al., 1974; Barmack and Pettorosi, 1980; Favilla et al., 1981). Blinking There are two forms of blinking motion, a reflex action occurring about once in every 2 to 10 seconds, and voluntary-force eye closure (Hung et al., 1977~. Blinking stimulates tear production and provides a pumping-out action for contaminants in the eye.9 The normal blink rate decreases with age and with the use of drugs such as tranquilizers. An abnormally high or low rate of blinking may be related (although no clinical evidence exists) to the ocular irritation syndrome of redness and injection (i.e., hyperemia) of sclera and cornea, the sensation of heat and burning of the anterior portion of the eye, and headache. Some visual tasks requiring concentrated attention reduce the rate of blinking (which may lead to dry eyes and irritation), but excessive blinking has been associated with reading difficult text. Transient quenching of the light input to the rods and cones of the retina may have a resetting effect on the excitability of these receptors and all of the subsequent visual neuron - -that is, the bipolar, horizontal, amacrine, and ganglion cells of the retina and neurons in the lateral geniculate body; the visual cortex; the parietal cortex; and the frontal eye fields. Thus blinking could be considered both as a corneal and as a visual "windshield wiper."

161 Blinking has been reported to be associated with fatigue since at least 1895 (Katz, as reported by Kravklov, 1974; Bartley, 1942; Luckiesh, 1947; Krivohlavy et al., 1969), but it has also been denied (Bitterman, 1945,1946,1947; Brozek et al., 1950; Wood et al., 1950; Holland and Tarlow, 1972~. Luckiesh and Floss (1937,1942) conducted a number of exper meets and concluded that the rate of involuntary blinking was a dependable criterion for ease of seeing. Subsequent work by McFarland and coworkers (;1942), Bitterman (1945,1946), and Tinker (1946), which failed to support Luckiesh's findings, touched off a debate in the literature (Bitterman, 1946,1947; Luckiesh, 1946, 1947~. Luckiesh defended his results, claiming that the different experimental conditions used by other investigators naturally led to different results. Bitterman and others argued that any mea- sure that required strict and rather narrow conditions could not be used as an index of 'visual effort" or 'visual fatigue." A later study by Tinker attempted to repeat in as much detail as possible the conditions of the Luckiesh and Moss experiments. Tinker was unable to confirm their results and concludes (Tinker, 1949:560~: "Under the conditions of this experiment, rate of involuntary blinking does not reflect differences in ease of see- ing." Brozek and coworkers (1950) measured several ocular functions at different time intervals for subjects performing "hard visual work" and found no significant increase in the blink rate. Finally, Wood and Bitterman (1950) had subjects perform a reading task in which visual effort and performance were varied sepa- rately. They found that blink rate was independent of visual effort but was inversely correlated with performance. Triadic Near Reflex: Combined Focusing, Convergence, and Aperture Mechanisms The mechanisms of the near reflex, that is, the triad of vergence, accommodation, and pupil diameter, are closely interlinked, each exerting some control on the other. This section first considers each mechanism separately and then discusses interactions between the three. Accommodation Focusing is achieved by both a lentincular and an extralenticular mechanism. The active ciliary muscle applies tension and stretches the peripheral portion of the zonule of Zinn, which causes the lenticular portion of the system to assume a particular

162 shape (i.e., a particular state of accommodation) determined by the elasticity of the lens and surrounding capsule.~° Blur is one of the stimuli for accommodation (Fincham, 1937, 1951; Allen, 1955; Phillips and Stark, 1977) and is detected via the foveal cones under photopic light conditions. While blur can tell the accommodative control system that the image is out of focus, a static value of blur is an even error signal: that is, it cannot indicate which way to change the accommodation in order to correct the error. The system overcomes this deficit by acting as a hill-climbing controller. In this way the control system can germane one Erection ot change that reduces blur and can then initiate a steady change in that direction until blur is minimized. The question of blur introduces a possible factor in visual distress during use of VDTs. The inexpensive terminals in use in many VDT installations have some inherent design features that prevent them from producing sharply defined (i.e., resolvable or legible) characters (Sakrison, 1977; Kajiya and Ullner, 1981~. The accommodative mechanism of a user's eye might therefore be robbed of the error signal it needs in order to achieve good visual acuity. This may also affect the comfort of VDT operators. Many other clues help the accommodative mechanism. Through synkinesis, for example, the vergence mechanism can produce a change in accommodation without requiring a blur signal to drive accommodation directly (see above). The accommodative system has a control-bias level with a set point of about 1 diopter (although there is considerable variation among people in this value). Accommodation drifts to this bias 1 ~ ._ ~ ~ i°Thus the dual, indirect, active theory of the accommodative mechanism was put forward by Helmholtz (1867) and has received support over the past century (Rohen and Rentsch, 1969; Saladin and Stark, 1975). It is a dual mechanism because both lenticular and extralenticular elements participate, indirect because the ciliary muscle does not act directly on the lens as a sphincter but acts indirectly by unloading the axial portion of the ciliary ligament, and active because activity in the ciliary muscle produces an increase in accommodation. See Stark (1968) for a discussion of the concept of a hill- climbing controller (which attempts to minimize blur but does not try to correct to zero error). The accommodative mechanism also has many other nonlinear characteristics (Stark, 1968~. It also has a leaky integrator (Krishnan and Stark, 1975) or leaky memory with about a 10-second time constant.

163 levelly by reducing illumination below photopic levels (night myopia), by removing a focusable image (space myopia) (Whiteside, 1957), or by making the system "open loop" by electronic methods (Phillips and Stark, 1977~. The accommodative control-bias level is sometimes called the dark focus. It is not peculiar to darkness, however; it applies equally to empty-field conditions and to opemloop conditions. Because the bias level is set by synkinetic control of the vergence system (see above), it is not a true focused state of accommodation. Static Factors in Accommodation The range of accommodation is - up to 16 diopters in children but only about 10 diopters in young adults. The gradual restriction of range of accommodation is called presbyopia, but since it begins in childhood, it is clearly not a degenerative condition.~3 By the age of 40, normal adults need reading glasses or bifocals as a static refractive prosthesis.~4 Certain other static disabilities are seen clinically. Accommo- dative spasm (fixed accommodation at the near point) usually also involves vergence, indicating that it originates in abnormal central nervous system control rather than in ocular muscles. Accom- modative asthenopia (fixed accommodation at the far point) occurs without evident accommodative effort, again suggesting that the ciliary muscles are not involved. Because these static conditions occur most frequently in persons aged 30-40 years, they may be related to the development of presbyopia. Accommo- dative spasm and accommodative asthenopia are examples of minor ocular deficits that may be related to the incidence of complaints of ocular discomfort in VDT users. Tithe drift toward bias level is called a lead by clinicians when the subject is focusing at infinity and a lazy lag when the subject is focusing at a near target. lithe ciliary muscle has normal strength and activity (Saladin and Stark, 1975), but since the growth of the lens continues, producing a larger lens with a decreased curvature and increased radius, there is a recession of the near point. Meredith Morgan, Elwin Marg, and Lawrence W. Stark believe that by the time one understands the mechanisms of accommoda- tion, pupillary contraction is one's only mechanism of accommo- dation; that is, increasing the depth of focus by pupillary constric- tion is the only way in which a presbyope can approach his or her near point.

164 Dynamic Factors in Accommodation The latency of accommoda- tion and the rate at which it can be changed vary among subjects and for the same subject under different conditions (Shirachi et al., 1978~.~5 Disfacility of accommodation, which is less severe than accommodative spasm or accommodative asthenopia, shows responses delayed beyond the usual latency and slowed rates of change (Liu et al., 1979~.~6 Responses reach the required accommodative levels but then drift back. Oscillation in level of accommodation beyond that normally found in healthy subjects can also occur. Fatigue and Accommodation The earliest work on fatigue and accommodation consisted of clinical impressions only, and there continues to be clinical literature (e.g., Weber, 1950~. Ergographic studies have been performed for the last 50 years (Lancaster and Williams, 1914; Howe, 1916; Blatt, 1934 Berens and Stark, 1932; Kurtz, 1937, 1938; Hofstetter, 1943; Ostberg, 1980; and Kintz and Bowker, 1982~.~7 Some of these studies have found deficits in accommodation that Methods of experimentally and clinically analyzing these dynamics include ergographic recordings and, more recently, the laser optometer. In these methods the subject positions the target so that it falls within his or her depth of focus. This positioning may be recorded as a measure of subjective appreciation of blur or laser speckle direction. A new laser ergograph method depends upon subjective appreciation of directional movement of a "speckle pattern" secondary to separation of the conjugate plane of the image from the retina. The image is formed by a rotating drum illuminated by a laser. Objective methods of analyzing accommodative dynamics include: the lensometer principle, which measures the blur of a target on the retina; the Scheiner method, in which blur is converted to prismatic shift (Malmstro~ al., 1981~; retinoscopic methods, which use directional image motion as an indication of accommodation error; and the third Purkinje image method (O'Neill and Stark, 1968), which directly measures lens anterior pole position. wit is interesting that both presbyopes and children can show accommodative disfacility. Both subjective (Hofstetter, 1943) and objective (Liu et al., 1979) observations have shown that disfacil- ity can be successfully treated by exercise. lithe ergograph was designed to measure and record objectively a sequence of continual movements so that if fatigue occurs and the amplitude of the movements decreases, it can be noted.

165 are usually recessions of the near point during sustained perfor- mance of a variety of near-work tasks. They all agree that the results are variable. For example, Gunnarsson and Soderberg (1980) report recessions of the near point for younger VDT workers and some transient approach of the near point during the early part of the day for older VDT workers. Saito and coworkers (1981), however, report approach of the near point with younger workers. Hofstetter feels the rapid restoration with attention indicates that the locus of fatigue is in the central nervous system and not in the muscles; that fatigue shows up in a masked, nonseeing eye almost to the same extent as in a seeing eye (Berens and Sells, 1954) rules out fatigue in sensory processes. Various investigators believe that what is lacking in their research is a distinction between fatigue and boredom and, especially, control of motivational factors that might combat performance decrements. Another basic issue is the question of whether the objective measurements define abnormally functioning accommodation or simply an adaptive response (especially if the pupil becomes smaller). That is, for an overlearned task involving muscular effort, an uncon- scious response is to reduce muscular effort to the lowest level compatible with task performance. Indeed, most often acuity and other measures of vision remain normal. Perhaps a skilled VDT operator shows an adaptive response that should not be construed as a sign of fatigue. A few authors have performed dynamic studies with fatigued subjects. Malmstrom and coworkers (1981) used a tracking task in which subjects followed near and far sinusoidally moving targets for a 6.5-minute period. Minor reductions in accommodative amplitude occurred. Logical defects in this study (and in many other studies) include the absence of necessity to maintain sharp focus on the target and the lack of control of possible papillary constriction. Krivohlavy and coworkers (1969) used subjective methods to show that the rate of change of accommodation was slowed in fatigued subjects, who took longer to cycle fixation between near and far targets; this effect was significantly corre- lated with reduced performance in a reading task. Several studies have reported changes in users' accommoda- tion in the course of work at VDTs. These studies have numerous methodological problems that make their interpretation difficult (see the discussion in Appendix A). For example, the study by Gunnarsson and Soderberg (1980) of changes in near points of accommodation and convergence did not include a control group of non-VDT workers performing similar visual tasks. Ostberg t6 Krueger (1980) has reviewed the physiology of and the clinical approach to accommodation.

166 (1980) and Ostberg and coworkers (1980) reported decreases in accommodative response and shifts in dark focus point among air traffic controllers after two hours of VDT work. No significant changes were found among two groups of telephone operators performing work involving VDTs. This study did not include any control groups of workers performing similar tasks without VDTs. The authors suggest that the differences in effects among the groups were due to the greater visual demands of air traffic con- trollers, and they argue that these findings are evidence of 'visual fatigue" caused by VDT work. However, the failure to include appropriate control groups, especially in light of the stressful nature of air traffic control work, makes it impossible to deter- mine whether these effects were associated with VDT work per se. Lurch (1982b) found no change in dark focus (i.e., bias level), accommodation, or convergence following a two-hour intensive visual search task at either a raster video display or a direct-view storage tube terminal. Haider and coworkers (1980) reported a very small transient decrease in visual acuity during the course of three hours of VDT work. No change in acuity was found in a non-VDT comparison group performing general office work. However, the work of the comparison group apparently differed from that of the VDT group in several respects, so it cannot be determined whether the reported change in acuity was related specifically to VDT task features. Small differences in acuity were reported between groups that worked with VDTs with green characters and those with yellow characters. Haider and coworkers suggested that the change in acuity signified "temporary myopization," probably due to "accommodation strain." Murch (1982b) also reported a reduc- tion in acuity following visual search at a raster video display, but not with a direct-view storage tube display. The finding suggests that the acuity change was related in some way to the raster display since the two tasks were well matched for other features. Follow-up of this preliminary study would be useful. Mourant and coworkers (1981) reported that the time required to move eye fixation and focus between near and far points slowed following a visual search task with a video display but not with a hard-copy display. The approach of this study is potentially valuable, but a number of methodological flaws make the results difficult to interpret. Eye movement records were evaluated subjectively, and reliability was not assessed (see Appendix A). Accommodation was probably not a factor in the reported effects since presbyopes and young workers showed similar changes, and both the distances used, 0.16 diopters and (apparently) about 2.2 diopters could have been within the depth of focus of the subjects ' eyes at the same time. Murch (1982b) reported that no significant change in focus speed from a near to a far target occurred

167 following VDT work, but he did not include details of his measurement procedures. In summary, none of the studies reporting oculomotor changes following VDT work has demonstrated that the changes are related to visual task features specific to VDT work. The significance of the reported oculomotor changes is not clear, but no evidence has been presented that they are problematic for workers. A serious defect in these VDT studies, and in other studies described in this section, is that papillary aperture was not measured. The pupil is known to constrict with sleepiness, boredom, or fatigue and, if constricted, would increase the depth of focus, reducing the requirement for accommodation. Note that unless pupil size is measured and depth of focus calculated, it is not possible to state that a certain measured accommodative state is or is not optimal. Vergence Vergence is the oculomotor control system that maintains binoc- ular fixation. As an object moves closer or farther away, the two visual axes rotate disjunctively and symmetrically to maintain bifixation on the point of interest. Vergence uses the same muscles as version, in which both eyes rotate in the same direc- tion toward an object. The vergence system operates much more slowly than the versional (or turning) system, probably because the synkinetic coordination of vergence with pupillary and accom- modative changes acts through smooth intraocular muscles. Dis- parity vergence is driven by the perception of disparate images of an intended object in the two eyes. The eyes rotate disjunctively so as to superimpose the two images. The brainstem mechanisms are not as well understood for vergence as for saccades. Stimulation of many areas in the brainstem induces convergence (Bender and Shanzer, 1964), and recordings have been made in the brainstem correlated with con- vergence motions (Keller and Robinson, 1972), but the exact path- ways have not been defined. Since these motions are disjunctive and since the main sequence relationships are different from those of saccades (Stark, 1975; Stark and Bahill, 1979), control of vergence probably involves different pathways from those of saccades. The usual method of examining vergence evaluates the stationary component, but when vergence must change, it does so by a smooth trajectory. Unlike saccadic motions, the smooth changes of vergence do not suppress visu al input or remap space into a stable perceptual space: that is, ~ he world may appear to move during changes of vergence. Of ale eye movements, ver-

168 gence demonstrates the most individual variation. The effects of aging, initial refractive state, or the mechanical tendencies of the eyes to cross or become wall-eyed (esophoria or exophoria) have all been described in the clinical literature. Some types of VDT use require continual vergence activity as the operator looks back and forth between written and video text at different distances. Also, as the image is refreshed on the VDT, there may be displacement of the letters, which may stim- ulate disparity vergence. Both kinds of vergence activities might contribute to ocular discomfort. Changes in the vergence system can often be demonstrated during prolonged performance of a visual task. Luckiesh and Moss (1935a, 1935b) showed that two aspects of the vergence system change with sustained activity. Using reading and inspection tasks lasting up to four hours, they showed that the distance from the observer at which bifixation is no longer possible as the target moves closer is increased with sustained activity. If this near point is expressed in diopters, the parameter is called the ampl~- tude of convergence; consequently, the amplitude of convergence decreases with sustained activity. Luckiesh and Moss also showed that muscle imbalance is increased with prolonged visual activity. Changes in balance (phorias) were found by Cobb and Floss (1925) to vary in different directions among subjects when one eye was covered and the amount of esophoria or exophoria of the covered nonfixating eye was estimated. The decreased amplitude of convergence caused by visual inspection work was also shown by Brozek and coworkers (1950~. Mahto (1972) suggested, on the basis of a small clinical study, that convergence insufficiency is the usual cause of complaints of eyestrain in persons aged 15 to 40 years. The association of close work with these complaints was also noted. The vergence system has been shown to be much more easily disrupted by centrally acting drugs and fatigue than the versional system (Rashbass, 1959; Westheimer and Rashbass, 196 1; Westheimer, 1963~.~9 Pupil Pupil as a Regulator of Light Level The pupil acts as an effective control mechanism for retinal illumination only for small changes of illumination. Larger changes, varying over many log units, are handled by multiple retinal adaptive mechanisms. A more impor- tant function of the pupil in regard to VDT studies is that pupil- , . ~9 Double vision is a sensory consequence of alcohol effects on the central nervous system controller of the vergence system.

169 lary constriction increases depth of focus and hence reduces the need for accommodation. For a presbyopic subject, pupillary constriction may be the major accommodative factor. A recent study by Cakir and coworkers (1978) has suggested that transient adaptation glare may produce significant transient changes in pupil size in subjects alternately viewing positive- and negative-contrast text projected by a slide projector. They noted that viewing positive~ontrast (light characters on a dark back- ground) VDT displays is often alternated with viewing illuminated negative-contrast (dark characters on a light background) hard- copy source documents, and they raised concerns that VDT work- ers might experience overexposure of their retinas because of alternately looking between the two. Rupp (1981) criticized this study and suggested that the pupillary response may have been caused by the brief period of darkness on the display appearing between the alternating fields. Dainoff (1982) reported that Rupp and coworkers have performed preliminary experiments (using a positive-contrast VDT and an illuminated negative~ontrast page of printed material) that confirm Rupp's conjecture. It should also be remembered that transient adaptation does not depend on pupillary changes alone, and changes in threshold have been reported when pupil size is controlled (Boynton, 1961; Rinalducci, 1967~. No studies have examined pupillary changes occurring when both VDT display and hard copy have the same (positive or negative) contrast. Pupillary Constriction Associated with Prolonged Visual Work Pupillary constriction following prolonged visual work within an institutional setting has been studied by Geacintov and Peavler (1974), who showed a significant pupillary~iameter decrease over the day's work in telephone operators using microfiche or tele- phone books. This work on pupillary constriction as an indicator of workload should be considered in the context of work on pupil- lary dilation as an indicator of emotional interest and arousal. Pupil as a Factor in Visuomotor Discomfort Bartley (1938) showed that slow flicker, 1 to 6 flashes per second, was of too high a frequency to be followed by pupillary movement; instead, a maintained constriction occurred (the Troelstra effect)20 and bathe Troelstra effect (Troelstra, 1968) is a maintained constriction caused by a sequence of light flashes that occur too frequently for the pupillary muscles to follow singly. More recent control studies of the pupil have shown that it can follow at least up to 3 Hz, so that it is not clear whether the static constriction or a small oscillation not indicated by older studies underlies Bartley's results.

170 discomfort was reported. Halstead (1941) used a mydriatic, scopolamine (which blocks papillary constriction), when he repeated the Bartley experiment, and under those conditions no ocular strain was reported by the subjects. Fugate and Fry (1956), King (1972), and Fry and King (1975) performed a related experi- ment. They used very bright light flashes well above sensory threshold and defined a ''border between comfort and discomfort"; and because they also believed that papillary constriction was the source of discomfort, they used atropine to block both the sphinc- ter and ciliary muscles. This approach substantially reduced the reported discomfort in their subjects. These studies implicate an oculomotor mechanism, repetitive papillary constriction, involved in visual discomfort and further demonstrate relief after elimi- nating that mechanism by means of drug paralysis. In apparent partial contradiction to these studies, Heaton (1966) found that hyoscine, a mydriatic and blocker of ciliary muscle, did not relieve various complaints of pain with eyestrain. The finding most related to the results of previous investigators was that pain after administration of eserine, a miotic drug, was different but neither greater nor less than that due to eyestrain. Pupillary Hippus in Relation to Habituation and Sleepiness Random contractions of the pupil (hippos) are a wel~known clin- ical phenomenon and are present in all normal persons. The amplitude of hippus is largest at moderate pupil sizes and decreases for both large and small values. Large-amplitude hippus has long been associated with a list of vague ailments, and Lowenstein and coworkers (1963) have related this to fatigue in a variety of clinical, normal, and abnormal settings. They point to the possibility of immediate recovery to normal hippos either by instructions to the subject or by psychosensory stimuli.22 This immediate dishabituation suggests that the fatigue is habituation of central nervous system origin. Synkinesis of Accommodation, Vergence, and Pupillary Constric- tion The three mechanisms controlling accommodation, vergence, and papillary constriction are intimately linked; each is capable of driving the other to some extent. Depth of focus is controlled by ZiUsui and Stark (1978) and others have considered hippus as noise with low band-pass characteristics. 22Lowenstein and Loewenfeld (1952) showed a continually diminished responsiveness with fatigue, sleepiness, and boredom. Immediately after a sudden alerting stimulus, they obtained psychosensory dilation of the pupil and increased constriction response to a light flash from subjects.

171 the pupil moving in synkinesis with accommodation and conver- gence. In the section on accommodation (see above), it was pointed out how this asymmetrical interaction between the control of accommodation and pupillary constriction could reduce the need for accommodative amplitude. This triadic synkinesis (O'Neill and Stark, 1968) is an essential part of the control of accommodation. Vergence drives accom- modation (Krishnan et al., 1977), and this has been found to be the main mechanism for change of focus. Disparity of an intended stimulus, the binocular parallax that is a signal to the vergence system, can be detected by the visual system even when the disparate images are blurred. However, disparity throws the retinal image off the foveae and thus removes the blur error signal to accommodation, since blur is only detected via foveal cones. Thus, in ordinary vision, as in switching attention from a VDT screen to nearer work or to distance vision, it is not the defocused images that drive accommodation but the disparate images that drive vergence. The accommodative state is corrected by interaction via the vergence~ccommodation mechanism. Accommodation then produces an accommodative pupillary constriction that increases the depth of focus, which lessens the requirement for accommodation on a near target. If at this stage the material on the VDT screen is not sharply focused, the accommodative system will be stimulated to reduce blur, and the accommodatio~vergence system and the pupillary constriction system will consequently adjust. SUMMARY AND CONCLUSIONS Most of the questions posed at the beginning of this chapter can- not be answered adequately on the basis of existing literature. A number of studies have reported higher incidences of ocular com- plaints and temporary alterations in oculomotor functions among VDT workers than among non-VDT workers. Most of these studies are flawed, and they do not establish whether these differences are related to parameters unique to VDT work. These studies have done little to increase understanding of problems associated with VDT use and the differences between VDT tasks and other visual tasks that require prolonged near work. "Visual fatigue" remains a nebulous concept after many years of research and discussion (National Research Council, 1939; Carmichael and Dearborn, 1947~. We suggest that it is more useful to refer to specific phenomena, such as performance decrement, oculomotor changes, and complaints of visual or ocular symptoms than loosely to invoke the term '"visual fatigue."

172 Complaints of ocular discomfort or visual symptoms following prolonged near-visual work are by no means unique to VDT work- ers; some studies suggest, however, that they may be more prevalent among workers who use VDTs than among those who do not. The oculomotor changes reported to follow VDT work are qualitatively similar to those found after other near-visual work; the significance of these changes is not clear. Most of the reviewed studies on oculomotor factors had severe methodological deficits (such as absence of pupillary measurement in testing accommodation)? and their results do not support the notion of a specific oculomotor physiological change with prolonged visual work. Even if the numerous suggestive findings are taken at face value as indicating a change in oculomotor state, an important question of interpretation remains. Are these changes a result of "fatigue" or are they a skillful adaptation to the task whereby practice allows effort to be minimized while still permitting adequate performance? The pupillary constriction experiments are the only studies that point to a specific phenomenon associated with discomfort. Two independent, established investigative groups suggested that drugs that stop papillary constriction also minimize subjective feelings of discomfort. The relationships between visual task performance, subjective fatigue, oculomotor changes, and motivational and attentional factors remain obscure; careful research that considers all these factors would be required to improve our understanding (see Chapter 10~.

Next: 8. Job Design and Organizational Variables »
Video Displays, Work, and Vision Get This Book
×
Buy Paperback | $60.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Along with the widespread use of computers have come growing fears that working in front of video display terminals (VDTs) can irritate and even damage the eyes. Separating scientific fact from popular opinion, this report takes a critical look at the link between VDT use and eye discomfort and disease as well as at changes in visual performance and oculomotor function. Drawing on information from ergonomics, illuminating engineering, and industrial and organizational psychology, the report gives practical advice on optimal workstation design to improve the comfort, performance, and job satisfaction of VDT users.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!