“Computer vision syndrome” (CVS) is defined by the American Optometric Association as the combination of eye and vision problems associated with the use of computers. These symptoms result from the individual having insufficient visual capabilities to perform the computer task comfortably.1 In 2003, it was estimated that 55% of jobs involved computer use.2 It seems likely that this number has now increased, and when combined with nonvocational computer use for activities such as e-mail, Internet access, and entertainment, one might suggest that computer usage is now almost universal.3 Indeed, a recent report suggested that adults may spend, on average, approximately 8.5 hours per day viewing electronic screens.4
It is difficult to estimate accurately the prevalence of ocular and visual symptoms associated with viewing electronic screens because both working conditions and the methods used to quantify symptoms vary widely. Nevertheless, Thomson5 suggested that between 64 and 90% of computer users experience visual symptoms that may include eyestrain, headaches, ocular discomfort, dry eye, diplopia, and blurred vision either at near or when looking into the distance after prolonged computer use. A recent investigation of office workers in New York City noted that 40% of subjects reported tired eyes “at least half the time,” whereas 32 and 31% reported dry eye and eye discomfort, respectively, with this same frequency.6 Other reports have also noted a strong association between dry eye and CVS, with longer periods of computer work being associated with a higher prevalence of dry eye.7
The prevalence of dry eye varies between 15 and 30% of the general population (depending on the precise criterion adopted) but increases with age and is greater in females.7–11 Furthermore, contact lens wearers may experience poor vision and dry eye symptoms because of poor lens surface wetting and reduced tear production. A review by Blehm et al.12 suggested that dry eye during computer operation could be caused by either a reduced blink rate or increased corneal exposure produced by the primary gaze position of the monitor (particularly in the case of desktop computers). Other factors that could contribute to symptoms of dry eye during computer use include poor environmental conditions caused by excessive heating or air conditioning and/or poor air quality caused by physical, biological, or chemical contaminants.13,14
Several investigations have shown that blink rate is reduced during computer operation. For example, Tsubota and Nakamori15 compared the rate of blinking in 104 office workers when they were relaxed, reading a book, or viewing text on an electronic screen. Mean blink rates were 22 per minute while relaxed but only 10 or 7 per minute when viewing the book or screen, respectively. In addition, Patel et al.16 observed mean blink rates before and during computer operation of 18.4 and 3.6 per minute, respectively. In addition, they noted a significant relationship between the stability of the precorneal tear film and the interval between blinks. Schlote et al.17 found that the reduced blink rate associated with computer use was also accompanied by distinct patterns of blinking. For example, some patients (all of whom had symptoms of dry eye) exhibited alternating interblink periods of longer and shorter duration. These authors hypothesized that the change in interblink duration during computer operation represented cognitive adaptation to the task. Furthermore, no significant correlation was observed between clinical measurements of the ocular tear film (tear breakup time, Schirmer I, or Jones tests) and the observed blink rate during computer operation.
It has also been reported that blink rate decreases as font size and contrast are reduced18 or the cognitive demand of the task increases.19,20 In addition, Sheedy et al.21 noted that voluntary eyelid squinting reduced the blink rate significantly. Therefore, the poorer image quality of the electronic text (as evidenced by increased reports of blurred vision during the course of a computer task, when compared with printed materials22) may adversely affect the blink rate. Interestingly, the application of topical elastoviscous solutions to the cornea does not modify the reduced blink rate associated with computer use.23 Reduced blinking may also exaggerate symptoms of preexisting dry eye, which could be exacerbated by other aspects of the work environment as previously noted, as well as factors such as contact lens wear and increasing age (particularly in females).
Whereas blink rate has been shown to decrease significantly with computer use,16,17,23 an additional factor to consider is the completeness of the blink, that is, does the upper lid cover the exposed cornea completely during the blink process. Himebaugh et al.19 analyzed the blink amplitude during a number of tasks, including computer operation, and observed that incomplete blinking was common, task dependent, and present in all subjects. These included both individuals with symptoms of dry eye and aged-matched normals. It is unclear whether incomplete blinking is undesirable. Harrison et al.24 noted that partial blinking is associated with staining of (and presumably damage to) the inferior cornea.25,26 Yet incomplete blinking is commonly found in asymptomatic patients,26 and provided that the portion of the cornea over the pupil is covered by the upper eyelid, one would expect to find uninterrupted clear vision. Accordingly, the aim of the present study was to determine whether incomplete blinking during computer operation is indeed associated with increased ocular and visual symptoms. A second investigation examined whether increasing the blink rate by means of an audible tone (subjects were instructed to blink on hearing the tone) would reduce the prevalence of post task symptoms.
The experiments were carried out on 21 visually normal subjects (9 men, 12 women) having a mean age of 24.4 years (range, 21 to 29 years). All had habitual visual acuity of at least 6/6 in each eye. None had strabismus or manifest ocular disease. The study followed the tenets of the Declaration of Helsinki, and informed consent was obtained from all subjects after an explanation of the nature and possible consequences of the study. The protocol was approved by the institutional review board at the SUNY State College of Optometry. Subjects read text aloud from a desktop computer screen (Compaq Evo 5500 with a 15-in monitor) at a viewing distance of 50 cm for a continuous 15-minute period. A forehead rest was used throughout the task to maintain a constant viewing angle and working distance. The screen was positioned so that about 75% of the visible screen area was positioned below the subject’s eye level. The upper and lower edges of the visible screen area lay approximately 7.5 degrees above and 22 degrees below eye level, respectively, at the 50-cm viewing distance. The text was composed of cognitively demanding stories taken from the Internet. The text was single spaced, black, 10-point Times New Roman font, with a contrast of approximately 80%. Target luminance was approximately 15 cd m2. The subject scrolled through the text as required using the computer mouse. To ensure concentration, subjects were told that they would be asked questions about the passage at the end of the session.
During the task, subjects were videotaped using a Logitech QuickCam web camera (Logitech, Fremont, CA) positioned immediately to the side of the computer monitor. The video recording was downloaded and stored using Logitech QuickCam software version 10.51.2029. After the trial was completed, the recording was reviewed to determine the mean blink rate per minute during the 15-minute session. Although the Web camera was in clear view, subjects were not told that their blinks were being monitored or recorded because this may produce conscious changes in this function.
In addition, the completeness of each blink was quantified using the following criteria:
- Grade 1. The upper eyelid failed to reach the top of the subject’s pupil while he or she viewed the computer screen.
- Grade 2. The upper eyelid reached the top but not the bottom of the subject’s pupil while he or she viewed the computer screen.
- Grade 3. No cornea was visible as the subject completed his or her blink.
A blink score was computed by multiplying the number of blinks in each category by the grade level. Immediately after each session, subjects completed a written questionnaire developed by Hayes et al.27 composed of 10 questions concerning the level of ocular discomfort experienced during the computer task.
In a second trial, the effect of controlling the blink rate was monitored. All testing conditions were similar to those previously described, except that a Web-based metronome program (www.webMetronome.com) was used to provide an audible beep every 4 seconds (i.e., 15 times per minute) and subjects were instructed to blink their eyes each time they heard a beep. This was designed to produce a blink rate of at least 15 per minute. Again, the written questionnaire regarding the level of ocular discomfort experienced during the computer task was completed after the session.
The mean symptom scores after the computer task for the two trials, that is, where blink rate was not and was controlled are shown in Table 1. The total symptom score, plotted as a function of the blink rate per minute for each subject for the uncontrolled blink condition, is shown in Fig. 1. A significant negative correlation was observed (r = 0.43; p = 0.05), with subjects having the lowest blink rate showing the highest symptom score. The total symptom score as a function of the blink score (i.e., the product of the number of blinks and the grading of each blink) for the uncontrolled blink trial is shown in Fig. 2. A significant negative correlation was also observed (r = 0.46; p = 0.035). In addition, the total symptom score was plotted as a function of the percentage of blinks that were deemed incomplete in the uncontrolled condition (i.e., grade 1 or 2). This function is shown in Fig. 3. A significant positive correlation was noted (r = 0.63; p = 0.002) between these parameters.
The mean blink rates for the uncontrolled and metronome (controlled blink) sessions were 11.29 blinks per minute (SEM, 1.67) and 23.45 blinks per minute (SEM, 1.65), respectively (t = 11.63; p < 0.0001). Although no significant difference was observed between the two conditions for any of the symptoms questioned (Table 1), in all cases, the mean symptom scores were extremely low. Therefore, data from the six subjects having the highest mean symptom scores in the uncontrolled blink condition were reexamined, and these results are shown in Table 2. Again, no significant difference in symptoms was noted between these two subgroups.
The findings of the present study confirm that increased symptoms during computer operation are associated with both a reduced blink rate and an increased percentage of incomplete blinks. The observation of a reduced blink rate is consistent with that of previous studies15–17 and may at least partially explain the higher prevalence of symptoms when viewing electronic screens. However, increasing the blink rate by means of an audible tone did not reduce CVS symptoms significantly in either the entire population tested or in the six subjects reporting the highest symptom scores in the control condition. It should be noted that, for this subgroup, increased blinking produced a 48% reduction in the mean total symptom score (from 37.17 to 19.33; Table 2). This did not reach statistical significance, probably because of the small number of subjects (n = 6). In addition, a noticeable reduction in the symptom of eyestrain was also observed in this subgroup (from 4.67 to 3.00; p = 0.14). Although these differences are worthy of further testing in a larger sample, several subjects reported that increased voluntary blinking in response to the audible tone actually interfered with their ability to perform the task satisfactorily. Although a lower metronome rate might have made task performance easier, it would also have decreased the difference between the control and metronome conditions. A limitation of the present study is that the magnitude of post task symptoms in the subjects tested was relatively low (Table 1). Different results may have been found if either a more demanding task (such as greater cognitive demand or longer duration) had been used or recruitment was limited to symptomatic subjects.
With regard to incomplete versus full blinks, Himebaugh et al.19 observed incomplete blinking in 100% of their subjects who were composed of 16 dry eye and 16 control individuals. They also reported that 20% of blinks covered less than 50% of the exposed cornea. As noted previously, chronic incomplete blinking has been associated with damage to the inferior corneal epithelium24–26 and inferior preblink tear breakup. Furthermore, Harrison et al.24 examined tear stability after both incomplete and full blinks and found that the newly deposited tear film was actually more stable after an incomplete blink in dry eye subjects. The authors hypothesized that, in dry eye patients having reduced tear volume, the tear film could be thinned or stretched further (thereby making it less stable) after a complete blink when compared with a partial blink.
A key question is whether the change in blink patterns during computer use is related specifically to viewing the electronic screen or is simply the result of performing a demanding near-vision task. Previous investigations noting a reduced blink rate during computer operation have not included a hard copy printed control trial, that is, where subjects perform the same task in printed (nonelectronic) form under equivalent conditions. Although it has been observed that blink rate decreases with task difficulty,28–30 this is not specific to computer use. It has been suggested that the poorer image quality of the electronic screen, when compared with printed materials, may be responsible for the change in blink rate.22 However, Gowrisankaran et al.31 observed that degrading the image quality by either inducing 1.00 diopter of uncorrected astigmatism or presenting the target at only 7% contrast did not produce a significant change in blink rate for a given level of cognitive load, whereas Gowrisankaran et al.18 reported that induced refractive error, glare, reduced contrast, and accommodative stress (varying the accommodative stimulus by ±1.50 diopters during the course of the task) produced an increased blink rate. In addition, Miyake-Kashima et al.32 found that introduction of an antireflection film over a computer monitor to reduce glare produced a significant reduction in blink rate. Therefore, one cannot account for the reduced blink rate observed in the present investigation on the basis of a degraded visual stimulus.
Gaze angle has also been cited as a critical parameter in accounting for symptoms during computer operation. It has been proposed that the primary position of most desktop computer monitors, when compared with the more common downgaze adopted when viewing printed text at near, could account for the difference in blink patterns. The rationale for this difference is that increased corneal exposure in primary gaze may lead to increased drying of the ocular surface. On this basis, one might expect an increased blink rate in the primary position, whereas most investigations of CVS have noted a reduced blink rate during computer use. Furthermore, the results of previous studies into the effect of gaze angle on blink patterns are contradictory. For example, both Cho et al.33 and Nielsen et al.34 reported that blink rates were significantly lower when a reading task was performed in downgaze, whereas Murube and Murube35 found a higher blink rate with downgaze viewing. Gowrisankaran et al.18 noted that upgaze viewing (25 degrees above the primary position) produced no significant change in blink rate.
In view of these contradictory findings, it is difficult to explain the reduced blink rate during computer viewing and even to state whether this is specific to the electronic task. One might question whether similar results would have been observed had a similar protocol been used for viewing cognitively demanding, hard copy, printed materials. Current work in our laboratory is comparing the effects of gaze angle and stimulus format (print vs. electronic) on blink rate. Nevertheless, if ocular dryness is indeed a significant factor in CVS, then the results of the present investigation suggest that incomplete blinking may be to blame. Furthermore, the results of the metronome trial indicate that simply telling the subject to blink more may not reduce the symptoms significantly, and may actually impair his or her ability to perform the task. Future work should be directed toward trying to achieve complete corneal coverage during blinking in an attempt to alleviate the symptoms associated with CVS. Alternative therapies such as instillation of lubricating drops or increasing ambient humidity may also represent useful treatment regimens.
It has been reported that CVS produces significant symptoms in approximately 40% of office workers at least half the time.6 In addition, between 15 and 30% of the general population (depending on the precise criterion adopted) have symptoms of dry eye.7–11 It seems likely that there will be considerable overlap in individuals experiencing both of these conditions. This study has demonstrated that CVS symptoms are greatest in subjects with the lowest blink rates and the highest percentage of incomplete blinks. Therapies need to be developed to improve the quality of the anterior ocular surface and thereby alleviate these relatively common conditions.
SUNY College of Optometry
33 West 42nd St
New York, NY 10036
Received January 7, 2013; accepted February 21, 2013.
3. Rosenfield M, Howarth PA, Sheedy JE, Crossland MD. Vision and IT displays: a whole new visual world. Ophthalmic Physiol Opt 2012; 32: 363–6.
5. Thomson WD. Eye problems and visual display terminals—the facts and the fallacies. Ophthalmic Physiol Opt 1998; 18: 111–9.
6. Portello JK, Rosenfield M, Bababekova Y, Estrada JM, Leon A. Computer-related visual symptoms in office workers. Ophthalmic Physiol Opt 2012; 32: 375–82.
7. Uchino M, Schaumberg DA, Dogru M, Uchino Y, Fukagawa K, Shimmura S, Satoh T, Takebayashi T, Tsubota K. Prevalence of dry eye
disease among Japanese visual display terminal users. Ophthalmology 2008; 115: 1982–8.
8. Moss SE, Klein R, Klein BE. Prevalence of and risk factors for dry eye
syndrome. Arch Ophthalmol 2000; 118: 1264–8.
9. Hikichi T, Yoshida A, Fukui Y, Hamano T, Ri M, Araki K, Horimoto K, Takamura E, Kitagawa K, Oyama M, Danyo Y, Kondo S, Fujishima H, Toda I, Tsubota K. Prevalence of dry eye
in Japanese eye centers. Graefes Arch Clin Exp Ophthalmol 1995; 233: 555–8.
10. Schaumberg DA, Sullivan DA, Buring JE, Dana MR. Prevalence of dry eye
syndrome among US women. Am J Ophthalmol 2003; 136: 318–26.
11. Schaumberg DA, Dana R, Buring JE, Sullivan DA. Prevalence of dry eye
disease among US men: estimates from the Physicians’ Health Studies. Arch Ophthalmol 2009; 127: 763–8.
12. Blehm C, Vishnu S, Khattak A, Mitra S, Yee RW. Computer vision syndrome
: a review. Surv Ophthalmol 2005; 50: 253–62.
13. Piccoli B. A critical appraisal of current knowledge and future directions of ergophthalmology: consensus document of the ICOH Committee on “Work and Vision.” Ergonomics 2003; 46: 384–406.
14. Salibello C, Nilsen E. Is there a typical VDT patient? A demographic analysis. J Am Optom Assoc 1995; 66: 479–83.
15. Tsubota K, Nakamori K.Dry eyes and video display terminals. N Engl J Med 1993; 328: 584.
16. Patel S, Henderson R, Bradley L, Galloway B, Hunter L. Effect of visual display unit use on blink rate
and tear stability. Optom Vis Sci 1991; 68: 888–92.
17. Schlote T, Kadner G, Freudenthaler N. Marked reduction and distinct patterns of eye blinking in patients with moderately dry eyes during video display terminal use. Graefes Arch Clin Exp Ophthalmol 2004; 242: 306–12.
18. Gowrisankaran S, Sheedy JE, Hayes JR. Eyelid squint response to asthenopia-inducing conditions. Optom Vis Sci 2007; 84: 611–9.
19. Himebaugh NL, Begley CG, Bradley A, Wilkinson JA. Blinking and tear break-up during four visual tasks. Optom Vis Sci 2009; 86: E106–14.
20. Jansen ME, Begley CG, Himebaugh NH, Port NL. Effect of contact lens wear and a near task on tear film break-up. Optom Vis Sci 2010; 87: 350–7.
21. Sheedy JE, Gowrisankaran S, Hayes JR. Blink rate
decreases with eyelid squint. Optom Vis Sci 2005; 82: 905–11.
22. Chu C, Rosenfield M, Portello JK, Benzoni JA, Collier JD. A comparison of symptoms after viewing text on a computer screen and hardcopy. Ophthalmic Physiol Opt 2011; 31: 29–32.
23. Acosta MC, Gallar J, Belmonte C. The influence of eye solutions on blinking and ocular comfort at rest and during work at video display terminals. Exp Eye Res 1999; 68: 663–9.
24. Harrison WW, Begley CG, Liu H, Chen M, Garcia M, Smith JA. Menisci and fullness of the blink in dry eye
. Optom Vis Sci 2008; 85: 706–14.
25. Abelson MB, Holly FJ. A tentative mechanism for inferior punctate keratopathy. Am J Ophthalmol 1977; 83: 866–9.
26. Collins MJ, Iskander DR, Saunders A, Hook S, Anthony E, Gillon R. Blinking patterns and corneal staining. Eye Contact Lens 2006; 32: 287–93.
27. Hayes JR, Sheedy JE, Stelmack JA, Heaney CA. Computer use, symptoms, and quality of life. Optom Vis Sci 2007; 84: 738–44.
28. Tanaka Y, Yamaoka K. Blink activity and task difficulty. Percept Mot Skills 1993; 77: 55–66.
29. York M, Ong J, Robbins JC. Variation in blink rate
associated with contact lens wear and task difficulty. Am J Optom Arch Am Acad Optom 1971; 48: 461–7.
30. Wong KK, Wan WY, Kaye SB. Blinking and operating: cognition versus vision. Br J Ophthalmol 2002; 86: 479.
31. Gowrisankaran S, Nahar NK, Hayes JR, Sheedy JE. Asthenopia and blink rate
under visual and cognitive loads. Optom Vis Sci 2012; 89: 97–104.
32. Miyake-Kashima M, Dogru M, Nojima T, Murase M, Matsumoto Y, Tsubota K. The effect of antireflection film use on blink rate
and asthenopic symptoms during visual display terminal work. Cornea
2005; 24: 567–70.
33. Cho P, Sheng C, Chan C, Lee R, Tam J. Baseline blink rates and the effect of visual task difficulty and position of gaze. Curr Eye Res 2000; 20: 64–70.
34. Nielsen PK, Sogaard K, Skotte J, Wolkoff P. Ocular surface area and human eye blink frequency during VDU work: the effect of monitor position and task. Eur J Appl Physiol 2008; 103: 1–7.
35. Murube J, Murube E. Near vision accommodation in horizontality with VDT: why low blinking and dry eye
? Adv Exp Med Biol 2002; 506: 1205–11.
Keywords:© 2013 American Academy of Optometry
blink rate; computer vision syndrome; cornea; dry eye; reading