Purpose: Ocular and visual symptoms are felt by many subjects during computer use. Previous studies have reported a reduced blink rate during computer operation and suggested that this may account for some of the symptoms, particularly dry eye. However, these earlier investigations did not include a control condition. To determine whether it is computer screen viewing that produces the change in blink rate, the present study compared blink patterns when reading from either a desktop computer monitor or a hard copy printed text under equivalent viewing conditions.
Methods: Subjects (N=118) were asked to undergo a continuous 20-minute reading task from either a desktop computer screen or a printed page at the same viewing distance. Identical text was used in the two sessions, with similar size and contrast. Target viewing angle and luminance were similar for the two conditions. Subjects were videotaped during the task to determine their blink rate. Immediately after the task, subjects completed a questionnaire regarding ocular symptoms experienced during the trial.
Results: Mean blink rates for the computer and hard copy conditions were 14.64 & 13.12 blinks per minute, respectively (p < 0.001). A significantly higher score of ocular symptoms was observed for the computer condition (p < 0.001).
Discussion: Blink rates were not reduced during computer operation when compared with an equivalent hard copy control condition. Suggestions state that the previously observed differences in blink rate are more likely to be produced by changes in cognitive demand rather than the method of presentation. However, during the use of computers, a higher score of ocular symptoms was noted which may have been associated with visual tiredness.
Conclusion: The study aims to compare the rate of blinking with and without a screen, to know if screens cause an influence on the blinking rate. Results show the increase in the blinking rate is more remarkable with screen than without screen. Future studies must focus on increasing the blink amplitude to determine whether this provides a viable method of reducing symptoms during computer use.
Our excessive and increasing use of computers has led to the development of a number of health problems. In contrast, our need for paper has decreased. Eye discomfort, fatigue, and stress are some of many work-related complaints and symptoms reported by many people who work on a computer. As computer use increases, so does the level of discomfort. The visual discomfort and related symptoms that occur in computer users should be treated as a growing health issue. Thus, arises the “Computer Vision Syndrome” which is the complex of eye and vision problems associated with experienced work while using the computer. A blink poses an interesting phenomenon that should be monitored and studied. In addition to being an important tool for scientific research, blinking is a necessity for our well-being, because of its
role as a protective barrier for the eye and as a distributor of the tear film on the ocular surface. The blink rate can be defined as the frequency during which spontaneous blinks occur and can be even used as a measure of visual fatigue. Several studies have shown that computer screens directly influence the blinking rate.
Between 64 & 90% of computer users experience visual symptoms, which may include eyestrain, headaches, ocular discomfort, dry eye, diplopic, and blurred vision either at near or when looking into the distance after prolonged computer use according to Thomson . A recent investigation of computer users in New York City noted that 40% of subjects reported tired eyes “at least half the time”, whereas 32 & 31% reported dry eye and eye discomfort, respectively, with this same frequency . Other studies have also found a strong association between
dry eye and computer-related symptoms, with longer periods of
computer work being associated with a higher prevalence of dry
eye . Investigations state that computer tasks reduce the blink
rate [4-6]. Tsubota and Nakamori  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 per minute and 7 per minute
when viewing the book or screen. These three testing conditions
varied in the method of presentation and in task demand. It
has been noted that as font size and contrast are reduced, blink
rates are reduced as well  or the cognitive demand of the task
increases [8,9]. Therefore, the differences observed by Tsubota
and Nakamori  may be related to changes in task difficulty and
are not the consequence of changing from printed material to
An important question arises: “Is the change in blink patterns
during computer use related specifically to viewing the electronic
screen or is it simply the result of performing a demanding nearvision
Indeed,  states that both the visual nature of the task
and the psychological status of the user should be considered
in studies examining eye blink activity during computer use.
Previous studies observing a reduced blink rate during computer
use have not included an equivalent hard copy control trial,
under the same viewing conditions. It has been suggested
that the poorer image quality of the electronic screen may be
responsible for the fluctuation in blink rate, when compared
with printed materials. However,  observed that degrading the
image quality by either inducing 1.00 diopter (D) of uncorrected
astigmatism or presenting a target at only 7% contrast did not
produce a significant change in blink rate for a given level of
cognitive demand. Furthermore,  reported an increased blink
rate when induced refractive error, glare, reduced contrast, and
accommodative stress (varying the accommodative stimulus by
±1.50 D during the course of the task). In addition,  found
that a significant reduction in blink rate is observed during the
introduction of an antireflection film over a computer monitor to
reduce glare produced. Since evidence lacks regarding the role of
the electronic screen in the change in blinking rates, the aim of the
present study was to determine if a difference in blink patterns
exists when subjects perform similar reading tasks from either a
computer monitor or printed materials.
The study involved 118 patients aged between 18 & 35 years.
They were examined in an optometry clinic. Subjects included
were those having a monocular visual acuity at least 10/10 for
far and near, with or without optical correction. Subjects excluded
were those having an uncorrected ametropia, having a squint or
an orthoptic problem, having an eye pathology or dry eye, having
undergone a previous eye surgery, patients suffering from a
general pathology influencing the visual system such as diabetes,
high blood pressure, etc.
The examination room was equipped with a slit lamp, an
autorefractometer, a digital phoropter, Schirmer’s test strips,
fluorescein strips, a projector of optotypes, a laptop, printed
papers and a video camera. The recruited patient was received in
a quiet room, with moderate lighting, of moderate temperature,
without ventilation or any other factor that could hinder the
examination of tear secretion and refraction. The contrast on the
computer screen and the printed paper was the same. Figure 1
the text read had the following characteristics identical on both
computer and paper: Size 14, Font Times New Roman, Black color,
a single space between words. Subjects wearing contact lenses
got rid of their lenses before 48 hours of our experience. Subjects
were not informed that their blinks were being monitored or
recorded as this could cause voluntary blinks. All of the subjects
were interviewed to fill out their examination sheets to ensure
first of all their age and the absence of any ophthalmological and
We started with the study of lacrimal secretion with the
Schirmer test (normal if more than 15mm per 5 minutes) and
the B.U.T. (normal if break after 10 seconds). Then, we performed
orthoptic tests including the motility test, PPC (Punctum
Proximum Convergence), cover/uncover test. These tests must
be within the normal limits. Visual acuity from far and near, raw
and with correction, was measured monocularly and binocularly.
It must be at least 10/10. To ensure that the patient is emmetropic
or perfectly corrected, we performed refraction. It was preceded
by a measurement on the autorefractometer. After completing the
information and carrying out the tests mentioned above, we made
the experience concrete, divided into two phases of 20 minutes
each, and separated by an interval of at least 24 hours.
The patient read identical text on a laptop screen during
the first phase and on printed paper during the second phase.
During these two phases, the subjects’ eyes were filmed using a
video camera from a cell phone (smartphone), held by a support.
Sufficient material was provided for 20 minutes of reading without
repetition. The recorded videos allowed us to count the number
of blinks during each phase. The results of the number of blinks
enabled us to determine the blink rate per minute. Immediately
after completion of each reading task, subjects completed a
written questionnaire asking about their level of ocular symptoms
observed during the two sessions. The results were recorded on a
scale ranging from 0-10 (0: zero, 10: severe).
The statistical test adapted in the study is the “analysis of
variance” as known as ANOVA. It is a test that is applied when one
or more discrete explanatory variables (the factors) that influence
the distribution of a variable continue to be explained. We can
compare Sig (The degree of significance) with α = 5% (The risk
of error). If Sig < α The qualitative variable has an effect on the
This part includes a descriptive presentation of the variables
studied throughout this study.
We considered 118 patients of an age group [18-35], taking
into account the theoretical increase in dry eyes, the decrease in
the blinking rate and the appearance of presbyopia with age. The
average age is 26.72 years with a standard deviation of 4.563.
The distribution of the population by gender was 56.78% male
and 43.22% female. 66.90% were emmetropes and 33.10% were
corrected ametropes [Tables 1-8]
The full population study can be found in Appendix A.
The ANOVA test carried out between the symptoms and the
experience “without screen” or “with screen” shows us that there
is a relationship between these two variables P-value ≤0.05. The
majority of subjects without a screen have a rate of 13.20 and
subjects with a screen have a rate of 15.30 (see Appendix D).
The average blink rate without and with screen are respectively
13.1161 and 14.6381. This means, as shown in the diagram, that
the average blinking rate with screen is slightly higher than that
without screen. The ANOVA test shows us a significant variation
P-value 0.00 < 0.05.
The aim of this research was to study the influence of
computer screens on the rate of blinking and the symptoms
observed. This was based on a comparison of the results obtained
by reading an identical text on a screen and on a printed paper,
since the influence of the screens causes a variation in the rate
of blinking and the appearance of symptoms. This work was
carried out on 118 subjects aged between 18 & 35, average is
26.72. 66.90% were emmetropes against 33.10% ametropes
perfectly corrected? Thus, this margin of difference allowed us
to establish a comparative study between the emmetropes and
the ametropes perfectly corrected. All patients underwent the
test with and without a screen. The average blink rate among
emmetropics was 13.12 without screen versus 14.50 with screen.
Whereas in perfectly corrected ametropics, the average was 13.10
without screen against 14.91 with screen. Fifty-six-point seventyeight
percent of the subjects were males and 43.22% were
females. Thus, this margin of difference allowed us to establish
a comparative study between males and females. All patients
underwent the test with and without a screen.
The average blink rate for males was 13.17 without screen
versus 14.82 with screen. Whereas in females, the average was
13.04 without screen against 14.39 with screen. By studying the
significance (p-value) without screen according to sex (p-value
= 0.563 > 0.05), the refractive state (p-value = 0.943 > 0.05) and
the profession (p-value = 0.534 > 0.05), we conclude that the
variations are not significant. On the other hand, by studying the
significance (p-value) with screen according to sex (p-value =
0.07> 0.05), the refractive state (p-value = 0.104 > 0.05) and the
profession (p-value = 0.135 > 0.05), we deduce that the variations
are also non-significant.
Compared to the symptoms found without a screen, we note
blurred vision scores while looking at text 0.41, blurred vision
looking from afar at the end of the task up close 0.36, difficulty
/ slowness in refocusing eyes from one distance to another 0.35,
irritated or burning eyes 0.75, dry eyes 1.00, eye fatigue 1.03, eye
discomfort 1.04, photophobia 0.22 and headache 0.40. On the
other hand, we find with screen an increase in the scores of all the
symptoms (Figure 2): blurred vision while looking at text 1.21,
blurred vision by looking from afar at the end of the task near
0.78, difficulty / slowness in refocusing eyes from one distance to
another 0.73, irritated or burning eyes 1.95, dry eyes 2.98, tired
eyes 3.29, eye discomfort 3.18, photophobia 0.77 and headache
0.85. The ANOVA statistical test shows a significant variation
(p-value < 0.05) for the 9 symptoms. The variation of the average
blink rate with screen is 14.63, greater than that without the value
13.11 screen. There is a significant variation according to ANOVA,
p-value = 0.00 < 0.05.
The total symptom score was higher regarding the computer
condition. However, between the 2 presentation methods, the
overall blink rate was not notably different. The absence of a
reduction in blink rate with computer use is contrary to previous
reports [4-6]. As noted earlier, these previous investigations did
not include an equivalent hard copy control trial. In the present
study, the hard copy condition was performed at the same
viewing distance and gaze angle and comprised the same reading
material while both target contrast and letter size were matched
as closely as possible. Since no difference was noted between the
two conditions, this would imply that the differences reported
previously were caused by variations in task demand rather than
being produced directly by the computer monitor. For example,
 found a significantly lower rate of blinking in office workers
viewing text on an electronic screen compared with when they
were relaxed. . observed a reduced blink rate when the
subjects were performing a computer task compared with “general
conversation”. In all cases, the increased cognitive demand of the
computer task could account for the change in blink rate reported
previously. Accordingly, evidence lacks to support the proposal
that performing a task using an electronic monitor will produce a
change in blink rate compared to carrying out the same task using
A review by Doughty  noted environmental conditions
affect the blink rate (e.g., ambient temperature and humidity),
gaze angle, and verbalization. Although the environmental
conditions were not strictly under control, all of the experimental
sessions were conducted in the same clinic, with variations in
temperature or humidity levels. Both gaze angle and verbalization
(i.e., the requirement to read the text aloud) were the same for
both conditions. Doughty  suggested that reading aloud (as
opposed to sitting in silence) might represent a “subtle stimulus”
to blinks, so one might conclude that slightly different results
could be reached if the subject read silently which is the popular
case in work environments. Although verbalization has been
shown to increase the blink rate, [12,13], moving the jaw on its
own does not significantly affect the blink rate . Accordingly,
task difficulty may form some sort of relationship between blink
rate and speech [14-18]. It seems reasonable to assume that, no
significant difference in blink rates between the computer and
hard copy conditions would have been noted even if the subjects
had read silently rather than aloud.
The aim of this study is to compare the rate of blinking
with and without a screen, in order to know if screens cause
an influence on the blinking rate and on the symptoms related.
According to the results, the increase in the blinking rate is more
remarkable with screen than without screen. By focusing on the
symptoms observed, we notice an increase of the scores with the
computer screen. The difference observed can be attributed to
the demand for the reading task, which is more complex on the
computer screen than on paper. This allows us to conclude that the symptoms present a more
severe score on the computer screen, as well as a slight increase
in the rate of blinking.
It shows that under tested conditions, no significant change in
blink rate is observed when using computers exclusively. The level
of cognitive demand is likely responsible for blink rate decline.
As mentioned before, increased symptoms during computer
operation were associated with a higher percentage of incomplete
blinks, future studies must focus on increasing the blink amplitude
to determine whether this provides a viable method of reducing
symptoms during computer use.