ArticlePDF Available

Visual discomfort while watching stereoscopic three-dimensional movies at the cinema

Authors:

Abstract and Figures

PurposeThis study investigates discomfort symptoms while watching Stereoscopic three-dimensional (S3D) movies in the ‘real’ condition of a cinema. In particular, it had two main objectives: to evaluate the presence and nature of visual discomfort while watching S3D movies, and to compare visual symptoms during S3D and 2D viewing.Method Cinema spectators of S3D or 2D films were interviewed by questionnaire at the theatre exit of different multiplex cinemas immediately after viewing a movie.ResultsA total of 854 subjects were interviewed (mean age 23.7 ± 10.9 years; range 8–81 years; 392 females and 462 males). Five hundred and ninety-nine of them viewed different S3D movies, and 255 subjects viewed a 2D version of a film seen in S3D by 251 subjects from the S3D group for a between-subjects design for that comparison. Exploratory factor analysis revealed two factors underlying symptoms: External Symptoms Factors (ESF) with a mean ± S.D. symptom score of 1.51 ± 0.58 comprised of eye burning, eye ache, eye strain, eye irritation and tearing; and Internal Symptoms Factors (ISF) with a mean ± S.D. symptom score of 1.38 ± 0.51 comprised of blur, double vision, headache, dizziness and nausea. ISF and ESF were significantly correlated (Spearman r = 0.55; p = 0.001) but with external symptoms significantly higher than internal ones (Wilcoxon Signed-ranks test; p = 0.001). The age of participants did not significantly affect symptoms. However, females had higher scores than males for both ESF and ISF, and myopes had higher ISF scores than hyperopes. Newly released movies provided lower ESF scores than older movies, while the seat position of spectators had minimal effect. Symptoms while viewing S3D movies were significantly and negatively correlated to the duration of wearing S3D glasses. Kruskal–Wallis results showed that symptoms were significantly greater for S3D compared to those of 2D movies, both for ISF (p = 0.001) and for ESF (p = 0.001).Conclusions In short, the analysis of the symptoms experienced by S3D movie spectators based on retrospective visual comfort assessments, showed a higher level of external symptoms (eye burning, eye ache, tearing, etc.) when compared to the internal ones that are typically more perceptual (blurred vision, double vision, headache, etc.). Furthermore, spectators of S3D movies reported statistically higher symptoms when compared to 2D spectators.
Content may be subject to copyright.
Visual discomfort while watching stereoscopic
three-dimensional movies at the cinema
Fabrizio Zeri
1
and Stefano Livi
2
1
Optics and Optometry, Department of Sciences, Roma Tre University, Rome, and
2
Department of Social and Developmental Psychology, University of
Rome “Sapienza”, Rome, Italy
Citation information: Zeri F, Livi S. Visual discomfort while watching stereoscopic three-dimensional movies at the cinema. Ophthalmic Physiol Opt
2015; 35: 271282. doi: 10.1111/opo.12194
Keywords: 2D movie, stereoscopic three-
dimensional movie, visual discomfort, visual
symptoms
Correspondence: Fabrizio Zeri
E-mail address: fabrizio.zeri@uniroma1.it
Received: 22 September 2014; Accepted: 13
December 2014; Published Online: 13 February
2015
Abstract
Purpose: This study investigates discomfort symptoms while watching Stereo-
scopic three-dimensional (S3D) movies in the ‘real’ condition of a cinema. In par-
ticular, it had two main objectives: to evaluate the presence and nature of visual
discomfort while watching S3D movies, and to compare visual symptoms during
S3D and 2D viewing.
Method: Cinema spectators of S3D or 2D films were interviewed by questionnaire
at the theatre exit of different multiplex cinemas immediately after viewing a movie.
Results: A total of 854 subjects were interviewed (mean age 23.7 10.9 years;
range 881 years; 392 females and 462 males). Five hundred and ninety-nine of
them viewed different S3D movies, and 255 subjects viewed a 2D version of a film
seen in S3D by 251 subjects from the S3D group for a between-subjects design for
that comparison. Exploratory factor analysis revealed two factors underlying
symptoms: External Symptoms Factors (ESF) with a mean S.D. symptom score
of 1.51 0.58 comprised of eye burning, eye ache, eye strain, eye irritation and
tearing; and Internal Symptoms Factors (ISF) with a mean S.D. symptom score
of 1.38 0.51 comprised of blur, double vision, headache, dizziness and nausea.
ISF and ESF were significantly correlated (Spearman r=0.55; p=0.001) but with
external symptoms significantly higher than internal ones (Wilcoxon Signed-
ranks test; p=0.001). The age of participants did not significantly affect symp-
toms. However, females had higher scores than males for both ESF and ISF, and
myopes had higher ISF scores than hyperopes. Newly released movies provided
lower ESF scores than older movies, while the seat position of spectators had min-
imal effect. Symptoms while viewing S3D movies were significantly and negatively
correlated to the duration of wearing S3D glasses. KruskalWallis results showed
that symptoms were significantly greater for S3D compared to those of 2D mov-
ies, both for ISF (p=0.001) and for ESF (p=0.001).
Conclusions: In short, the analysis of the symptoms experienced by S3D movie
spectators based on retrospective visual comfort assessments, showed a higher
level of external symptoms (eye burning, eye ache, tearing, etc.) when compared
to the internal ones that are typically more perceptual (blurred vision, double
vision, headache, etc.). Furthermore, spectators of S3D movies reported statisti-
cally higher symptoms when compared to 2D spectators.
Introduction
Since the beginning of the first decade of the twenty-first
century, the world of cinematography has witnessed a
sudden renaissance of stereoscopic three-dimensional (S3D)
cinema. In this period, the number of S3D movies produced
and distributed, as well as the number of theatres equipped
with the available digital S3D screens, has steadily increased.
©2015 The Authors Ophthalmic & Physiological Optics ©2015 The College of Optometrists
Ophthalmic & Physiological Optics 35 (2015) 271–282
271
Ophthalmic & Physiological Optics ISSN 0275-5408
The strong success of The Polar Express,releasedintheatres
in 2003, ignited this new era in the history of S3D cinema.
1
S3D movies have been reported to provide more emotion
and a higher sense of immersion,
2,3
and this may explain the
reason for the recent S3D releases generating three times
more revenue per screen, than 2D versions.
1
However, evi-
dence of improvement in emotional experiences with stereo-
scopic stimulus has not always been found.
4
The use of S3D has been sporadic in the history of cin-
ema, so there is no certainty that the current rebirth will
last. Notwithstanding this, the question about the health
effects of viewing S3D movies has recently sparked a heated
debate among laypeople and visual scientists (e.g. Refs.
5,6). While there are several studies that have investigated
symptoms of discomfort while viewing S3D displays and
S3D home theatre systems,
3,710
few studies have been con-
ducted to directly address the possible side effects of view-
ing S3D movies in cinemas. Yang et al.
3
briefly mention a
pilot study of 96 adults that discovered few, low-intensity
symptoms that quickly dissipated, which were more appar-
ent in the 3D compared to the 2D viewers.
11
Surprisingly,
dry eye symptoms were more severe for the 2D viewers.
Another survey showed that S3D movies can increase ocu-
lomotor symptoms, nausea and disorientation compared
with 2D movies.
12
Concerning home theatre systems, Yang
et al.
3
reported greater and more frequent perception of
ocular and visual symptoms, such as blurred vision, double
vision, dizziness, disorientation and nausea during S3D
viewing rather than 2D viewing. More recently, Read &
Bohr
10
reported that only a minority of viewers, around
14%, experienced symptoms such as headache and eye-
strain, due to S3D viewing.
It is possible to highlight five main sources that can
potentially be a cause of visual discomfort for people view-
ing S3D movies: perceptual conflicts from depth informa-
tion, physiological conflict between accommodation and
convergence, retinal rivalry, an excessive amount of retinal
disparity, and lastly perceptual distortion induced by S3D
production. Between these sources, a certain degree of
interconnection can be found.
One example of perceptual conflicts between depth infor-
mation (that can be divided into pictorials, kinetics and
physiological; for a review see Ref. 1315) is the break of
the stereoscopic window or bending.
1
The physiological conflict between accommodation and
convergence has been implicated as the most likely problem
during S3D viewing, and has been studied exten-
sively.
5,6,9,1621
Retinal rivalry can also occur in S3D cinema and varies
with the S3D cinema projection system (for example, time
multiplexing versus light polarisation vs wavelength multi-
plexing).
1
The crosstalk phenomenon,
22
the incomplete bar-
rier of the right-eye filter for left images and vice-versa, is
an example that produces double contours (ghosting)
induced retinal rivalry.
Also, excessive amount of retinal disparity induced by a
S3D film can be an issue. There is large consensus among
researchers that the amount of disparity in stereoscopic
images should be limited within a range called the ‘com-
fortable viewing range’.
1,5,7,9,23,24
However, the comfortable
viewing range is not strictly defined,
25
and even though
image disparity doesn’t change once a film is produced, the
viewing distance of the observer can change, therefore sig-
nifying that the angular disparity of the viewer cannot be
strictly controlled by the film producer. Moreover, the
speed of disparity variation in space and time is another
important factor for discomfort control.
2628
The last potential source of visual discomfort, the percep-
tual distortion induced by S3D production, is typically related
to technical problems during production of the film (the
process of image acquisition) that can cause several types of
image distortion. There are three main aspects that can
affect image quality during acquisition: the distance
between the cameras, the convergence distance (the distance
away from the cameras at which the optical axes of the cam-
eras intersect), and the field of view of the cameras (deter-
mined by the CCD format size and lens focal length).
29
Well-known distortion effects include Keystone distortion,
lens distortion, shear effect, and the cardboard effect.
1,29
The five main sources of discomfort just described can
affect people to different degrees of severity, due to a poten-
tially high number of individual differences. For example, dif-
ferent interpupillary distances (a wide range exists
30
)can
affect the position of a stereoscopic object, even though the
amount of disparity of that object on the screen is the same: a
smaller interpupillary distance produces a higher stereoscopic
depth.
5
Similarly, different accommodation-convergence
relationships (in terms of AC/A or CA/C ratio) can affect
how well viewers handle and tolerate conflict between accom-
modation and convergence stimuli.
6,31
For example, Yang
et al.
3
have found that presbyopes are less likely to report
accommodation and/or vergence-related symptoms during
S3D film viewing. The presumed rationale is because presby-
opic people have significantly reduced ability to vary accom-
modative responses when compared to younger spectators.
Different levels of visual discomfort may also be due to
differences in viewing position with S3D, such as the dis-
tance from and angle to the screen, as well as the screen
dimensions.
1
Yang et al.
3
have demonstrated that sitting in
an oblique position reduced perceived immersion but at
the same time reduced motion symptoms in S3D viewing.
The present research had two main objectives:
1. To evaluate the presence and kind of visual discomfort
when watching S3D movies at a cinema (according to the
Sheedy’s symptoms structure),
32
taking into account the
effect of viewer characteristics (age, gender, self-reported
©2015 The Authors Ophthalmic & Physiological Optics ©2015 The College of Optometrists
Ophthalmic & Physiological Optics 35 (2015) 271–282
272
S3D movies and visual discomfort F Zeri and S Livi
visual correction), cinema setting variables (seat posi-
tion, movie release date) and, finally, variables related to
the use of S3D glasses and S3D subjective perception.
2. The second aim was to compare visual symptoms dur-
ing S3D and 2D viewing (i.e. the effect of the dimension
of the movie presentation on symptoms).
Methods
Participants and procedure
Cinema spectators of S3D film were interviewed between
2010 and 2011 on a voluntary basis for anonymous data col-
lection using a uniform questionnaire (see next section)
which was completed immediately after the conclusion of
the movie, at the exits of the different multiplex cinema the-
atres. The resulting data therefore included several S3D films
(Table 1), all of them were dubbed from the original version.
Moreover, in order to compare visual symptoms during S3D
and 2D viewing, the same data collection strategy was
applied with 2D spectators of the last film studied in S3D:
Pirates of the Caribbean: On Stranger Tides’(Table 1). In all
the multiplex cinema theatres where the data were collected,
a passive glasses 3D system of projection (circularised polari-
sation) based on RealD digital technology, was utilised.
A total of 854 subjects were interviewed in Italy
(mean standard deviation age of 23.7 10.9 years,
range 881 years; 392 females and 462 males). The sample
included 283 (33%) spectators that declared having a
refractive error at distance: 208 interviewees (74% of those
with refractive error) reported having myopia or myopic
astigmatism; 43 interviewees (15% of those with refractive
error) reported having hyperopia or hyperopic astigma-
tism; and the remaining subjects (11%) declared not know-
ing their refractive status. Among those with refractive
errors, 84% used vision correction while viewing the movie
(58% glasses, 39% contact lenses and 3% did not specify).
Questionnaire
The questionnaire used in the interviews was developed to
study S3D film viewing symptoms, was made up of four
sections pertinent to this study. A second version of the
questionnaire was developed specifically to study the symp-
toms of spectators viewing 2D film. This second question-
naire was comprised of the same S3D questionnaire except
for section 3, concerning aspects of S3D perception, which
was removed.
Section 1: spectators’ information
This section included four variables linked to personal
characteristics of the interviewee: gender, age, type of
refractive error for distance and if so, whether this
correction was used during the movie (with a specific
question regarding whether spectacles or contact lenses
were used).
Section 2: cinema setting information
This section was related to viewing conditions. The first
variable investigated was the interviewee’s approximate seat
position in the theatre. In order to establish the position, a
sketch of a generic rectangular floor plan of the cinema
divided into nine sectors (3 93 grid) was provided in the
questionnaire:
1. Three blocks of rows in longitudinal direction from the
screen to the end of the theatre;
2. Three blocks of seats in horizontal direction (left, cen-
tral and right).
In this way, the sectors 1, 2 and 3 were the first block of
rows in front of the screen on the left, centre and right
respectively. Sectors 4, 5 and 6 were the second block of
rows (left, central and right, respectively). Finally, sectors 7,
8 and 9 were the farthest block of rows from the screen
(left, central and right respectively).
For the subsample of 251 subjects that viewed the S3D
version of Pirates of the Caribbean, and the 255 subjects that
viewed a 2D version of the same film in the same multiplex
cinema, the relative position for the nine sectors was con-
verted. This was done via comparison of each theatre’s
floor plan where the movies were projected (see below),
using two absolute values for each single sector (referring
to the centre position of the sector):
1. An absolute value of the angular position in respect to
the centre of the screen; and
2. An absolute value of the visual angle subtended by the
screen position.
The second variable investigated in this section was
the time spent using the S3D glasses. To assist the sub-
ject’s judgment on time, an analogue scale was provided
whereby the subject had to indicate the amount of time
spent wearing S3D glasses during the movie just viewed.
This was a 10 cm line with descriptors at both ends
and in the middle (never,always and almost half time,
respectively) and ticks to effectively create a 10-point
scale.
Table 1. Number of spectators for each viewed film
Film
Number of spectators
interviewed
Pirates of the Caribbean S3D version 251
Avatar S3D 101
Gulliver S3D 85
Sanctum S3D 73
Alice in Wonderland S3D 46
Others S3D 43
Pirates of the Caribbean 2D version 255
©2015 The Authors Ophthalmic & Physiological Optics ©2015 The College of Optometrists
Ophthalmic & Physiological Optics 35 (2015) 271–282
273
F Zeri and S Livi S3D movies and visual discomfort
Section 3: subjective S3D experience
This section surveyed the subject’s stereoscopic perception
in terms of the length of perceived stereoscopic sensation
and intensity of the effect. The first variable that considered
temporal judgment used an analogue scale as described in
the previous section. The subject had to indicate the
amount of time in which he/she was able to see stereo-
scopic effects by placing a mark along a continuous line,
with the two ends described as ‘never’ and ‘always’. To
assess perceived strength of stereoscopic effects, the ques-
tionnaire used a Likert scale that ranged from 1 (nothing)
to 5 (very much).
Section 4: visual discomfort
Subjectively perceived visual discomfort during movie
viewing was measured by a Likert scale that ranged from 1
(nothing) to 5 (very much). Eleven specific symptoms were
investigated for degree of discomfort. The first nine symp-
toms were taken directly from the list of asthenopic symp-
toms used by Sheedy et al.
32
in 2003. In our study, the
original scale used by these authors (ranged from 1 to 100)
was reduced to a five-point Likert scale. The items were eye
burning, eye ache, eyestrain, eye irritation, tearing, blurred
vision, double vision, dryness and headache. Two more items
(dizziness and nausea) were added because some S3D spec-
tators interviewed in a pilot study reported these kinds of
symptoms in an open-ended question about the sensation
felt during viewing. These symptoms have been also studied
by other authors as symptoms linked to S3D-induced
motion sickness.
3,12
Data analyses
First, an exploratory factor analysis (Principal Component
Analysis, with Promax rotation) was first performed on the
11 symptom items to evaluate if clustering overlapped with
the one obtained in Sheedy’s paper.
32
In order to test the
first aim of the study, Spearman Rho was performed to
examine correlations among 11 symptoms. KruskalWallis
and Wilcoxon Signed-ranks tests were used to compare
ESF and ISF scores. More specifically, this study investi-
gated the effect of different variables: spectator characteris-
tics (gender, age, type of refractive error for distance and
actual use of correction during movie watching), cinema
setting variables (release date of film, spectator’s seat posi-
tion in the theatre, time spent using S3D glasses while
movie watching), and subjective S3D experience (length of
stereoscopic sensation perceived and intensity of S3D per-
ception) on the clustered asthenopic symptoms obtained in
the exploratory factor analysis.
To compare visual symptoms during S3D and 2D view-
ing, KruskalWallis Tests and logistic regression models
were performed using screen dimension (S3D vs 2D) as the
first factor, and several variables relative to spectators and
the viewing condition (age, gender, ametropia, kind of
ametropia, type of vision correction, spectator’s seat posi-
tion in the theatre) as the second factor. The clustered as-
thenopic symptoms obtained in the exploratory factor
analysis served as the dependent variable. All statistical
analysis was performed using SPSS 20 (http://www.ibm.-
com/software/analytics/spss/).
Results
Study 1: S3D symptoms
Sample
For the first study, 599 subjects (mean standard devia-
tion age of 24.4 11.8 years; range 881 years; 259
females, 43% and 340 males, 57%) that viewed a S3D
movie, completed the questionnaire. One hundred and
ninety-three (32%) of the interviewees declared having a
refractive error at distance: 141 (73% of those with refrac-
tive error) reported having myopia or myopic astigmatism,
37 (19% of those with refractive error) reported having
hyperopia or hyperopic astigmatism, and the remaining
subjects declared not knowing their refractive status.
Among them 155 subjects (80%) used vision correction
while viewing the movie (57% glasses, 40% contact lenses
and 3% did not specify the type of correction).
The seat position of the sample of S3D spectators is
reported in Figure 1. Thirteen of the respondents did not
provide this information. The distribution in the theatre is
not uniform. This is evident for every sample analysed: the
overall sample (all S3D and 2D spectators), the sample of
S3D spectators and finally the Pirates of the Caribbean 2D
version sample. The central block of seats parallel to the
screen (sectors 4, 5 and 6) and perpendicular to the screen
(sectors 2, 5 and 8) were chosen more frequently by the
spectators. Frequency of actual spectators seat position
(sectors) is statistically different compared to random dis-
tribution (v
2
=698.4; p<0.001). Nevertheless, the specta-
tor’s seat position distribution may reflect not only the
choice of the spectator but also the fact that in a multiplex
theatre it is possible to book the ticket online and the sys-
tem assigns the seat in a way that strictly reflects the
observed distribution (first the seats in the centre of the
theatre and then the others).
The mean standard deviation time of wearing S3D
glasses while watching the movie (on the questionnaire’s 1
10 scale) was 9.2 1.5 (range 110), implying an average
of about 92% of the movie’s duration. Fifty-eight per cent
reported wearing the S3D glasses for the whole film.
Mean standard deviation time spent with stereoscopic
perception and the mean standard deviation strength of
those stereoscopic effects were 7.7 2.6 (range 010 on a
©2015 The Authors Ophthalmic & Physiological Optics ©2015 The College of Optometrists
Ophthalmic & Physiological Optics 35 (2015) 271–282
274
S3D movies and visual discomfort F Zeri and S Livi
10 point scale) and 3.4 0.9 (range 15 on a 5 point
scale), respectively. Thirty-eight per cent reported they had
full 3D perception for the whole film and 9% reported they
had full strength 3D effects.
Visual discomfort in S3D
Table 2 reports statistical information (median, mean,
standard deviation, asymmetry, kurtosis, minimum and
maximum values, inter-quartile range) about the 11 symp-
toms. Similar to Sheedy et al.’s study
32
all subjective
reported symptoms (apart from strain) were not normally
distributed. Symptoms had median values of 1 (except for
strain with a median equal to 2), although all subjective
answers covered the complete range from 1 to 5.
Factor validity of the symptoms scale
The correlation matrix among symptoms (Table 3) reveals
that all the symptoms are significantly correlated. Explor-
atory factor analysis was used to reduce the data and iden-
tify a restricted number of factors underlying the 11
symptoms (see Sheedy et al.
32
). Principal component
analysis was used for factor extraction and Promax with
Kaiser normalisation was used for factor rotation.
33
Parallel
Analysis (PA)
34
was applied as extraction algorithm and
compares the observed eigenvalues with those produced by
the analysis of a correlation matrix based on the same num-
ber of random variables.
35
The mean eigenvalues for all
behaviours across 100 random data sets were calculated
and compared with the observed Scree plot. The factors
extracted were those with eigenvalues greater than the point
on the plot where the observed and random eigenvalues
cross. This technique has been demonstrated to be the most
accurate means of deciding on the optimal number of fac-
tors to extract.
35
The Parallel Analysis suggested a two-factor solution,
and so two factors were extracted using principal compo-
nent analysis. The two-factor model explained 53% of the
sample variance (Table 4), with the first factor accounting
for 41% of variance and the second factor 13% (while the
third and fourth factors explained 9% and 7% of the vari-
ance respectively), demonstrating that although the two
factors are statistically reasonable, a general and latent fac-
Figure 1. Interviewees’ viewing position in the theatre (%). Sectors 1, 2 and 3 are the block of rows closest to the screen on the left, centre and right
respectively. Sectors 4, 5 and 6 represent the middle block of rows and sectors 7, 8 and 9 are the rows farthest from the screen.
Table 2. Descriptive statistics of perceived symptoms (on a 15 scale) from 599 interviewed S3D viewers
Median % scoring >1 IQ range Skewness Kurtosis Mean S.D.
Burning 1 40 12 1.41 1.64 1.6 0.8
Ache 1 23 11 2.21 5.26 1.3 0.6
Strain 2 56 13 0.86 0.03 1.9 1.0
Irritation 1 29 12 1.89 3.42 1.4 0.7
Tearing 1 25 11 2.37 6.02 1.4 0.7
Blur 1 49 12 1.08 0.38 1.7 0.9
Double Vision 1 25 11 2.14 4.33 1.4 0.7
Dryness 1 24 11 2.47 6.53 1.4 0.8
Headache 1 28 12 2.12 4.48 1.4 0.8
Dizziness 1 19 11 2.94 9.86 1.3 0.6
Nausea 1 10 11 4.61 23.65 1.2 0.6
S.D., standard deviation; IQ range, inter-quartile range.
©2015 The Authors Ophthalmic & Physiological Optics ©2015 The College of Optometrists
Ophthalmic & Physiological Optics 35 (2015) 271–282
275
F Zeri and S Livi S3D movies and visual discomfort
tor of perceived symptoms is still present underlying the
two extracted factors. In fact, the two factors extracted are
well correlated (r=0.53). The pattern matrix of the Pro-
max rotation (Table 4 and Figure 2) reveals that these fac-
tors are very similar to the ones obtained in the previous
work of Sheedy et al.
32
These researchers called these two
major factors the ‘External Symptom Factor’ (ESF) that
includes burning, ache, strain, irritation and tearing and
the ‘Internal Symptom Factor’ (ISF) that includes blur,
double vision, headache, dizziness and nausea. In our data,
dryness symptoms were primarily in ISF (0.38) while in the
Sheedy et al.
32
study it was included in the ESF. However
in our data dryness also had a high secondary loading in
ESF at 0.32. From an optometric perspective, it could be
possible to define these two factors, ESF and ISF, as ocular
(localised at the eye level) and perceptual (they are more
linked to the vision process) respectively.
The mean of symptom severity for ISF and ESF overall
were computed from their respective items identified by
factor analysis, above. The mean (S.D.) symptom sever-
ity was 1.38 (0.51) and 1.51 (0.58) for ISF and ESF
respectively. A Wilcoxon Signed-ranks test indicated that
ESF were significantly higher than ISF symptoms
(Z=5.69, p=0.001). The two scales were significantly
correlated (Spearman r=0.55; p=0.001). Internal consis-
tency of the scales were satisfactory for ISF (Cronbach’s
alpha =0.79) and ESF (Cronbach’s alpha =0.83).
Symptoms relationship with spectator characteristics, cinema
setting variables and subjective S3D experience
Spectator characteristics. Using the KruskalWallis test,
gender had a significant effect for ISF and ESF, with females
reporting more problems than males: for ESF, the female
mean S.D. score was 1.57 0.63 and the male
mean S.D. was 1.45 0.53 (H
1
=6.35, p=0.012),
while for ISF the female mean S.D. was 1.43 0.51 and
the male mean S.D. was 1.35 0.52 (H
1
=6.35,
p=0.049). No significant correlation was found with age
(for ISF, Spearman Rho =0.04, p=0.34; for ESF Spear-
man Rho =0.06, p=0.14).
For ametropia we considered three levels: myopic
(myope or myopic astigmatism), hyperopic (hyperope or
hyperopic astigmatism) and presumed emmetropic (16
participants with a reported refractive error had not
Table 3. Spearman correlation matrix among symptoms seen in S3D (all coefficients are p<0.01)
Burning Ache Strain Irritation Tearing Blur Double Dryness Headache Dizziness Nausea
Burning
Ache 0.45
Strain 0.48 0.50
Irritation 0.55 0.40 0.41
Tearing 0.34 0.26 0.31 0.47
Blur 0.24 0.33 0.49 0.24 0.20
Double Vision 0.16 0.34 0.34 0.24 0.26 0.50
Dryness 0.29 0.32 0.34 0.32 0.28 0.28 0.27
Headache 0.28 0.38 0.39 0.25 0.26 0.36 0.32 0.33
Dizziness 0.20 0.30 0.28 0.17 0.20 0.28 0.27 0.22 0.50
Nausea 0.21 0.24 0.20 0.21 0.19 0.27 0.31 0.28 0.34 0.50
Table 4. Pattern matrix of the oblimin rotation (S3D). In bold primary
loadings.
Components
12
Burning 0.88 0.19
Irritation 0.87 0.10
Tearing 0.73 0.05
Strain 0.63 0.19
Ache 0.57 0.24
Dizziness 0.11 0.83
Nausea 0.11 0.75
Headache 0.10 0.70
Double Vision 0.06 0.68
Blur 0.13 0.61
Dryness 0.33 0.38
Burning
Irritation
Tearing
Strain
Ache
Dizziness Nausea Headache
Double Vision Blur
Dryness
–1.00
0.00
1.00
–1.00 –
0.50
0.00 0.50 1.00
Component 2
Component 1
Figure 2. Factor loading plots (S3D).
©2015 The Authors Ophthalmic & Physiological Optics ©2015 The College of Optometrists
Ophthalmic & Physiological Optics 35 (2015) 271–282
276
S3D movies and visual discomfort F Zeri and S Livi
declared which type, so were discarded for this analysis).
For ESF, no statistical differences were found (H
2
=1.25,
p=0.54), while for ISF, differences were significant
(H
2
=7.49, p=0.024): post hoc comparison revealed that
myopes experienced higher levels of ISF symptoms com-
pared to people with emmetropia (p=0.01), while no sta-
tistical differences were found between other groups
(Figure 3). Moreover, no differences were found either for
ISF or ESF regardless of the type of ametropia (myopia or
hyperopia) for the modality of vision correction (glasses,
contact lenses or absence of correction).
Cinema setting variables. The seat position in the theatre
(relative position in the nine sectors) did not significantly
affect ESF symptoms (H
8
=8.46; p=0.39) but reached a
significant level for ISF (H
8
=17.63; p=0.024). Post hoc
analysis revealed that the only significant difference among
the nine seat positions was between sector 5 (mean
ISF =1.45) and sector 9 (mean ISF =1.18) (H
8
=12.50;
p=0.001); those values represent the maximum and mini-
mum ISF symptoms registered in the nine seat sectors. In
the subsample of 251 subjects viewing the S3D version of
Pirates of the Caribbean, the relationship between symp-
toms and the absolute value of viewing position with
respect to the centre of the screen and the absolute value of
visual angle subtended by the screen position were calcu-
lated. Results show no significant correlation (all Spearman
rlower than |0.08|, p>0.10). Symptoms were significantly
and negatively correlated with the proportion of time the
S3D glasses were worn (ISF, Spearman r=0.16;
p=0.001; ESF, Spearman r=0.22; p=0.001). Finally,
concerning the six S3D movies represented in this study
(see Table 1), no differences were found for ISF
(H
5
=10.97; p=0.052), while statistical differences were
found for ESF (H
5
=18.54; p=0.002). Post hoc compari-
sons show that ESF symptoms reported for Sanctum were
significantly lower than for Avatar (H
1
=15.81; p=0.001)
and Alice in Wonderland (H
1
=10.56; p=0.001), and ESF
symptoms for Pirates of the Caribbean were lower than for
Avatar (H
1
=7.74; p=0.005) (see Figure 4).
Subjective S3D experience. Both ISF and ESF were corre-
lated with S3D movie experience: symptoms were nega-
tively correlated with both length of perceived stereoscopic
sensation (ISF, Spearman r=0.27; p=0.001; ESF,
Spearman r=0.23; p=0.001) and strength of stereo-
scopic effects (ISF, Spearman r=0.25; p=0.001; ESF,
Spearman r=0.22).
Study 2: comparison between symptoms in S3D and 2D
Sample
For the second study we interviewed another sample of 255
participants viewing a 2D version of Pirates of the Caribbean
in the same period. The symptoms reported in the 2D group
of people were compared with those of a subsample of 251
people from the S3D group described above that viewed the
S3D version of the same film (Pirates of the Caribbean),
allowing a between-subjects comparison of the 2D and S3D
versions of the same film (506 participants in total).
The 255 spectators that viewed the 2D version had a
mean standard deviation age of 22.0 8.4 years, (range
849 years). One hundred and thirty-three were female
and 122 male. Eighty-nine subjects (35%) declared being
ametropic: 67 subjects (75% of those with refractive error)
reported having myopia or myopic astigmatism, six sub-
jects (7% of those with refractive error) reported having
hyperopia or hyperopic astigmatism and the remaining
18% declared not knowing their refractive error. Among
ametropes, 82 subjects (92%) used vision correction while
viewing the movie: 48 (59%) glasses, 31 (38%) contact
lenses and three subjects (4%) did not specify the type of
correction used. The frequency of the spectator’s seat
Figure 3. Mean and standard error of ISF and ESF perceived symptoms from the S3D sample as a function of self-reported refractive correction.
©2015 The Authors Ophthalmic & Physiological Optics ©2015 The College of Optometrists
Ophthalmic & Physiological Optics 35 (2015) 271–282
277
F Zeri and S Livi S3D movies and visual discomfort
position (sectors) was statistically different compared to a
random distribution (v
2
=206.2; p<0.001) (Figure 1),
with a propensity toward central seating.
The 251 subjects viewing the S3D version of Pirates of the
Caribbean had a mean standard deviation age of
22.0 11.1 years (range 965 years). Ninety-six were
female and 155 male. Seventy-four subjects (30%) declared
being ametropes: 65 subjects (88% of those with refractive
error) reported having myopia or myopic astigmatism, five
subjects (7% of those with refractive error) reported having
hyperopia or hyperopic astigmatism and the remaining 6%
declared not knowing their refractive status. Sixty-three
subjects (85%) used vision correction while viewing the
movie: 35 (56%) glasses and 28 (44%) contact lenses. As
with the comparison group, distribution of the spectators’
seat position (sectors) was statistically different compared
to a random distribution (v
2
=325.5; p<0.001) (Fig-
ure 1).
The two samples do not differ in terms of age
(F
1,504
=1.77; p>0.10) ametropia (v
2
=0.34; p>0.10),
vision correction (v
2
=3.08; p>0.10) or vision correction
use (v
2
=1.88; p>0.10). A statistically significant differ-
ence was found between 2D and S3D spectators’ seat posi-
tion (v
2
=24.7; p<0.01) (Figure 1).
Relationship between S3D and 2D symptoms
KruskalWallis tests indicated that symptoms were signifi-
cantly higher for S3D compared to 2D movies both for ISF
(mean S.D.; S3D 1.38 0.55 vs 2D 1.15 0.33;
H
1
=51.84, p=0.001) and for ESF (S3D 1.48 0.59 vs
2D 1.22 0.43; H
1
=46.35, p=0.001) (Figure 5). Each
individual symptom was significantly different at
p=0.001, except for nausea (p=0.036).
A logistic regression was performed in order to test the
main effects of movie dimension and gender. Results
showed a significant effect for movie dimension (for ISF:
B=1.48; S.E. =0.26; Wald =22.43; p=0.001; for ESF:
B=1.20; S.E. =0.25; Wald =22.48; p=0.001) but not
for gender nor for the interaction.
In order to assess whether age influenced and interacted
with movie dimension (S3D vs 2D), as previous studies
demonstrated for some viewing symptoms,
3
we divided the
age sample in quartiles: (1) ranging from 8 to 15
(N=112); (2) ranging from 16 to 18 (N=134); (3) rang-
ing from 19 to 24 (N=129); (4) ranging from 25 to 65
(N=134). Results of ISF confirmed the effect of movie
dimension (B=1.62; S.E. =0.48; Wald =11.27;
p=0.001) but there was no significant effect concerning
age or the interaction. The same results were found for ESF
with a positive main effect of movie dimension (B=1.05;
S.E. =0.48; Wald =4.85; p=0.028), but no effect of age
nor interaction.
Two logistic regression models were performed with
movie dimension (S3D vs 2D) as the first factor and pres-
ence or absence of ametropia as the second factor predict-
ing dichotomised ISF (absence of symptoms (N=248) vs
presence of symptoms (N=258)) and ESF absence of
symptoms (N=248) vs presence of symptoms (N=258).
ISF main effects were significant both for movie dimension
(B=1.34; S.E. =0.22; Wald =35.78; p=0.001) and also
the presence of ametropia (B=0.64; S.E. =0.28;
Wald =5.45; p=0.02), while the interaction was not sig-
nificant. The same pattern was evident for ESF: the movie
dimension (B=1.32; S.E. =0.22; Wald =34.92;
p=0.001) and ametropia (B=0.94; S.E. =0.28;
Wald =11.62; p=0.001) were significant but not the
interaction.
Ametropia included three levels: myopia (simple myopia
or myopic astigmatism), hyperopia (simple hyperopia or
hyperopic astigmatism) and emmetropia. Logistic analysis
revealed that movie dimension (S3D vs 2D) was again
Figure 4. Mean and S.D. of ISF and ESF perceived symptoms from the S3D sample as a function of the different film viewed.
©2015 The Authors Ophthalmic & Physiological Optics ©2015 The College of Optometrists
Ophthalmic & Physiological Optics 35 (2015) 271–282
278
S3D movies and visual discomfort F Zeri and S Livi
significant for ISF and ESF at similar levels to the previous
analysis. Ametropia type was significant for ISF (B=0.32;
S.E. =0.15; Wald =4.69; p=0.030) and ESF (B=0.54;
S.E. =0.15; Wald =12.86; p=0.001), but the interactions
were not significant.
Vision correction included three levels: glasses, contact
lenses and no correction. Once again, a significant main
effect for movie dimension (S3D vs 2D) for ISF and ESF
was found at similar levels to the previous analyses. In
addition, vision correction was significant both for ISF
(B=0.49; S.E. =0.18; Wald =35.58; p=0.001) and ESF
(B=1.34; S.E. =0.22; Wald =35.78; p=0.001), but
there were no significant interactions. In order to better
understand the significant main effect of vision correction
and compare it with the results of study 1, we ran two
separate KruskalWallis tests, one for S3D and one for
2D. Results showed that for S3D no significant differ-
ences were found among the three kinds of visual correc-
tion, hence confirming the results of the first study but
also the Solimini study
12
that found no effect of the use
of glasses or contact lenses on S3D symptoms. Instead,
significant differences were found with 2D movie watch-
ing in perceived ISF (H
2
=8.67, p=0.013), where partic-
ipants with contact lenses reported higher level of
internal symptoms (although overall with lower values
than with S3D): the mean S.D. scores were
1.15 0.29 (glasses), 1.27 0.46 (contact lenses) and
1.13 0.31 (no correction). For ESF significant differ-
ences (H
2
=8.87, p=0.012) were due to higher scores
for both glasses (1.31 0.56) and contact lenses
(1.31 0.50) compared to no correction (1.18 0.50)
with 2D movie watching.
Finally, in order to analyse the interaction between movie
dimension and seat position, two logistic regression were
performed, using movie dimension (again 2D and S3D)
and seat position (nine levels), one for each symptom.
Results again showed a main effect of movie dimension
(for ISF: B=1.44; S.E. =0.50; Wald =8.16; p=0.004; for
ESF: B=1.44; S.E. =0.50; Wald =8.16; p=0.004), but
not seat position nor interaction.
Discussion
The present study is an attempt to investigate symptoms of
discomfort while watching S3D movies in the ‘real’ condi-
tions of a cinema. In particular, the study had two main
objectives: to evaluate the presence and type of visual dis-
comfort while watching S3D movies and to compare visual
symptoms between S3D and 2D viewing. In regard to the
first aim, it was found that symptoms experienced by S3D
spectators could be categorised into two main groups:
External Symptoms (eye burning, eye ache, eye strain, eye
irritation and tearing) and Internal Symptoms (blur, dou-
ble vision, headache, dizziness and nausea). The former
family of symptoms is typically ocular and each symptom
can be localised in the area of the eyes. In contrast, the sec-
ond family of symptoms is more perceptual. This division
into internal and external symptoms was proposed by Shee-
dy et al.
32
for asthenopia linked to stressful visual tasks.
Our clusters of symptoms almost perfectly overlap theirs.
Only ‘dryness’ presents with a subtle difference between the
two studies: in Sheedy’s study,
32
it was included among
external symptoms, while in the present study it was shared
between the two classes of symptoms.
Nausea
Figure 5. Mean and S.D. of perceived symptoms from the sample of interviewed S3D (N=251) and 2D (N=255) viewers. The first two bars refer
to the average of symptoms that comprise ESF and ISF.
©2015 The Authors Ophthalmic & Physiological Optics ©2015 The College of Optometrists
Ophthalmic & Physiological Optics 35 (2015) 271–282
279
F Zeri and S Livi S3D movies and visual discomfort
Sheedy et al.
32
found that external sensations were higher
in the condition of forced dry eyes or upward gaze, and
internal sensations were significantly higher in the case of
accommodative or convergence stress. Considering that the
physiological conflict between accommodation and conver-
gence has been implicated as the main problem during S3D
viewing among several potential sources of visual discom-
fort,
5,6,9,1621
one could expect a higher level of internal
symptoms in cinema S3D viewing. Contrary to this, in our
study, external symptoms were higher than internal even
though the two families of symptoms were well correlated.
The apparent difference in findings may be for many rea-
sons including the specific research question asked and
study designs. Alternatively, because this current study was
based in commercial cinemas and the majority of other
S3D symptom studies were not, a significant environmental
component may have been revealed by our study.
More recently, Yang et al.
3
have grouped the symptoms
differently: external ocular (such as eye burning, gritty eye,
etc.), internal ocular (such as pulled eye, pain in the eye),
physical (such as neck or shoulder ache), visual (such as
blurry or double vision) and finally motion sickness (dizzi-
ness, nausea and disorientation). This different way of clus-
tering was based on a different set of symptoms; moreover,
subjective viewing symptoms were measured in comparison
to a baseline before movie viewing, and S3D viewing was in
a home theatre setting. These research design differences
make it difficult to compare our results to theirs. Even so,
they found that stereoscopic S3D viewing can lead to both
some heightened ‘ocular’ symptoms (pulled sensation and
pain inside the eye that they classified as internal) and ‘per-
ceptual’ (double images that they classified as visual; disori-
entation and dizziness that they classified as motion
sickness). But the perceptual symptoms showed a larger
increase with S3D in that study.
An interesting outcome of the present study is that gen-
der had an effect both on ESF and ISF, with females experi-
encing more symptoms than males. This is in accordance
with the results of Solimini.
12
He suggested that women
with susceptible visual-vestibular systems may experience
more S3D symptoms. As these data were self-reported, it is
possible that females were more willing to report symp-
toms/problems than males.
In relation to ametropia, we ascertained that myopia was
linked to a higher level of subjective ISF symptoms. In our
second study, we found that ametropia induced more
symptoms in spectators (both in S3D or 2D). Further stud-
ies should be carried out to confirm this outcome, and to
also evaluate the degree of ametropia influence (that was
not investigated in our study). However, it should be taken
into consideration that the assessment of the refractive
error by self-report has some weaknesses in terms of speci-
ficity and sensitivity, although it may be more precise in
determining myopic refractive status.
36, 37
Moreover we
have no knowledge as to whether the self-reported classifi-
cation of refractive error could be affected by the age of
interviewee.
Only ISF symptoms experienced by S3D spectators seem
be affected by seat position of spectators with the only sig-
nificant comparison being between sector 5 (higher symp-
toms) and sector 9. This result is not easy to explain. Yang
et al.
3
have found a clearer interaction between film dimen-
sion (2D vs S3D), and seat positioning in relation to
motion symptoms (nausea, dizziness). A possible explana-
tion of the difference in the results, is that our spectators
spontaneously chose their respective positions. This choice
would have been made to limit their visual problems.
Instead, in Yang’s study,
3
the spectators were randomly
assigned to the different seating positions. Further, Yang’s
study
3
utilised a typical home theatre layout, as opposed to
this study’s commercial theatre, although they also men-
tioned an earlier pilot study which had also found that seat-
ing had an influence on symptoms in commercial theatres.
Another finding from this study was that spectators had
more symptoms if they wore S3D glasses for a shorter time
during movie viewing (negative correlation between symp-
toms and length of S3D glasses wearing; ISF, Spearman
r=0.16; p=0.001; ESF, Spearman r=0.22;
p=0.001), if they experienced shorter S3D perception dur-
ing viewing (negative correlation with S3D perception
duration, ISF, Spearman r=0.27; p=0.001; ESF, Spear-
man r=0.23; p=0.001) or if they had a lower strength
of stereoscopic effects (ISF, Spearman r=0.25;
p=0.001; ESF, Spearman r=0.22). A possible explana-
tion for this, could be that those with lower strength of ste-
reoscopic effects, as well as shorter perceived stereoscopic
sensation, may have visual problems, thus leaving them
prone to several potential symptoms or perceptual prob-
lems, such as binocular rivalry; in turn, this may lead to
taking the S3D glasses off. The same logic may apply to
those who have more comfortable S3D vision: with better
visual abilities, and more stable S3D perception, they can
use S3D glasses for a longer time without experiencing the
need to take them off.
Finally, a variable that affected only ESF symptoms was
the release date of the film. A more recent film like Sanctum
gave lower symptoms when compared to Avatar or Alice in
Wonderland (see Figure 5). This conclusion is interesting,
especially when considering the fact that Avatar and Sanc-
tum (both produced by James Cameron) have been pro-
duced using the same technique. A plausible explanation
could be that the stereoscopic box in the more recent mov-
ies has been narrowed compared to the older ones, as well
as the greater attention given to depth continuity.
1
This
would have reduced the amount of retinal disparity, and
limited the accommodative-convergence conflict. There is
©2015 The Authors Ophthalmic & Physiological Optics ©2015 The College of Optometrists
Ophthalmic & Physiological Optics 35 (2015) 271–282
280
S3D movies and visual discomfort F Zeri and S Livi
also another explanation linked to the self-selection of S3D
film audiences: with time, people who experienced symp-
toms during S3D movies viewing, maybe less likely to view
new S3D movies. This did not happen at the beginning of
the S3D renaissance with the blockbuster Avatar that
attracted many spectators with no recent S3D experiences.
In any case, comments and discussion of the last result
should be accompanied with caution in view of the fact that
the particular setting of the real cinema viewing did not
allow the control of numerous variables, many of them
being unknown (such as the stereoscopic box of the single
film).
In regard to the second aim of the study, results showed
that S3D symptoms (ESF, ISF or the individual ones) were
statistically higher than the 2D symptoms reported for the
same film. The seat position in the theatre had no effect on
this outcome, which was also evidenced in the first study
(concerning ESF only) as discussed above.
On a final note, both studies concluded that symptoms
were not age-related. The first consideration regarding this
outcome, seems to suggest that a smaller interpupillary dis-
tance in younger people, causing a larger stereoscopic
effect, does not affect symptoms. This result should be con-
firmed by further researches on an increased number of
young participants in which interpupillary distance is mea-
sured directly. Also for what concern older participants we
didn’t find evidence of any changes of symptoms. Yang
et al.
3
found that presbyopic people are less likely to report
accommodation and/or convergence-related symptoms.
We probably did not have a number of presbyopes in our
sample to verify this, or the differences in test conditions
that may influence detection of this variable (e.g. viewing
distances, movie choice, screen features, etc.).
To summarise, the current study analysed the symptoms
experienced by S3D movie spectators based on a retrospec-
tive visual comfort assessment. During S3D viewing, spec-
tators reported a higher level of external symptoms
(localised close to the eye such as eye burning, eye ache,
tearing, etc.) compared to internal ones that are typically
more perceptual (blurred vision, double vision, headache,
etc.). The specific film affected the level of symptoms for
S3D viewers. Spectators of S3D movies reported statistically
higher symptoms compared to 2D spectators for the same
movie, but this result was not affected by the age of inter-
viewees, the seat position (if chosen freely), being of little
importance, it seems.
As S3D content and viewing conditions continue to
develop, further studies of the many variables on symp-
tomatology, prevention and potential treatment interven-
tions are warranted. The question of S3D viewing
causing more symptoms, than 2D viewing, appears to be
confirmed by this study, as well as others cited through-
out this article. Although the average symptom increase
for S3D conditions is typically small compared to 2D,
this does not describe the more extreme symptoms
reported by individual viewers, nor the apparent lack of
symptoms in others. With S3D presentation permeating
education, many forms of entertainment, advertising,
etc., meaningful work to identify characteristics that
increase or decrease the likelihood of an individual hav-
ing notable symptoms with S3D viewing, including the
effective variations of S3D production, display or treat-
ment interventions, would be beneficial to those with
S3D perception or symptom challenges.
Acknowledgements
This research received no specific grant from any funding
agency in the public, commercial, or not-for-profit sectors.
The authors wish to acknowledge the invaluable contribu-
tion of Scott Cooper for having reviewed the manuscript
provided his comments and advice. The authors are
thankful to Stefano Donato for his work in the pilot study,
Gabriele Campetella for his work in the data collection for
the second study, Gianpaolo Giusti and Lauro Barbaresi of
Technicolor SpA for their useful suggestions in the devel-
opment of the research and Gianluca De Pasquale for his
support in addressing the various topic raised in the
review process. Moreover, a special thanks goes to the
UCI Cinemas Direction and Laura De Francesco, Maria
Chiara Zeri, Fabrizio Mondragon Parra and Francesca Es-
posito for their help in the data collection of the first
study.
Disclosure
The authors report no conflicts of interest and have no pro-
prietary interest in any of the materials mentioned in this
article.
References
1. Mendiburu B. 3D Movie Making. Stereoscopic Digital Cinema
from Script to Screen. Elsevier: Burlington, 2009.
2. Brooks FP. What’s real about virtual reality? IEEE Comput
Graph Appl 1999; 19: 1627.
3. Yang S, Schlieski T, Selmins B et al. Stereoscopic viewing
and reported perceived immersion and symptoms. Optom
Vis Sci 2012; 89: 10681080.
4. Ba~
nos RM, Botella C, Rubi
o I, Quero S, Garc
ıa-Palacios A &
Alca~
niz M. Presence and emotions in virtual environments:
the influence of stereoscopy. Cyberpsychol Behav 2008; 11:
18.
5. Lambooij M, Ijsselsteijn W, Fortuin M & Heynderickx I.
Visual discomfort and visual fatigue of stereoscopic displays:
a review. J Imaging Sci Technol 2009; 53: 030201-(14).
©2015 The Authors Ophthalmic & Physiological Optics ©2015 The College of Optometrists
Ophthalmic & Physiological Optics 35 (2015) 271–282
281
F Zeri and S Livi S3D movies and visual discomfort
6. Howarth PA. Potential hazards of viewing 3-D stereoscopic
television, cinema and computer games: a review. Ophthal-
mic Physiol Opt 2011; 31: 111122.
7. Wopking M. Viewing comfort with stereoscopic pictures: an
experimental study on the subjective effects of disparity
magnitude and depth of focus. J Soc Inf Disp 1995; 3: 101
103.
8. Kooi FL & Toet A. Visual comfort of binocular and 3D dis-
plays. Displays 2004; 25: 99108.
9. Shibata T, Kim J, Hoffman DM & Banks MS. The zone of
comfort: predicting visual discomfort with stereo displays. J
Vis 2011; 11: 129.
10. Read JCA & Bohr I. User experience while viewing stereo-
scopic 3D television. Ergonomics 2014; 57: 11401153.
11. Cooper SC, Gietzen JA, Hayes JR & Sheedy JE. Compari-
son of eye and vision-related symptoms when viewing a 2d
versus 3d movies. American Academy of Optometry Scien-
tific Poster Session November 1720, 2010. San Francisco,
CA.
12. Solimini AG. Are there side effects to watching 3D movies?
A prospective crossover observational study on visually
induced motion sickness. PLoS One 2013; 8: 18.
13. Maffei L & Fiorentini A. Arte e Cervello. Zanichelli Editore:
Bologna, 1995.
14. Cutting JE & Vishton PM. Perceiving layout and knowing
distance: the integration, relative potency and contextual use
of different information about depth. In: Perception of Space
and Motion (Epstein W & Rogers S, editors), Academic
Press: New York, 1995; pp. 69117.
15. Goldstein EB. Sensation and Perception, 6th edn. Wads-
worth-Thomson Learning: Pacific Grove, 2002.
16. Inoue T & Ohzu H. Accommodative responses to stereo-
scopic three-dimensional display. Appl Opt 1997; 36: 4509
4515.
17. Ukai K & Kato Y. The use of video refraction to measure the
dynamic properties of the near triad in observers of a 3-D
display. Ophthalmic Physiol Opt 2002; 22: 385388.
18. Emoto M, Niida T & Okano F. Repeated vergence adapta-
tion causes the decline of visual. J Dis Technol 2005; 1: 328
340.
19. Okada Y, Ukai K, Wolffsohn JS et al. Target spatial fre-
quency determines the response to conflicting defocus- and
convergence-driven accommodative stimuli. Vision Res
2006; 46: 475484.
20. Ukai K & Howarth PA. Visual fatigue caused by viewing ste-
reoscopic motion images: background, theories, and obser-
vations. Displays 2008; 29: 106116.
21. Hoffman DM, Girshick AR, Akeley K & Banks MS. Vergence
accommodation conflicts hinder visual performance and
cause visual fatigue. J Vis 2008; 8: 130.
22. Seunti
ens PJH, Meesters LMJ & IJsselsteijn WA. Perceptual
attributes of crosstalk in 3D images. Displays 2005; 26: 177
183.
23. Jones GR, Lee D, Holliman NS & Ezra D. Controlling per-
ceived depth in stereoscopic images. Proc SPIE 2001; 4297:
4253.
24. Sun G & Holliman N. Evaluating Methods for Controlling
Depth Perception in Stereoscopic Cinematography. IS&T/SPIE
Electronic Imaging. International Society for Optics and
Photonics, 2009: 72370I-72370I-12.
25. Chen MJ, Kwon DK, Cormack LK & Bovik AC. Optimizing
3D image display using the stereoacuity function. In Image
Processing (ICIP), 2012 19th IEEE International Conference
on Image Processing (pp. 617620). IEEE.
26. Yano S, Ide S, Mitsuhashi T & Thwaites H. A study of visual
fatigue and visual comfort for 3D HDTV/HDTV images.
Displays 2002; 23: 191201.
27. Speranza F, Tam WJ, Renaud R & Hur N. Effect of disparity
and motion on visual comfort of stereoscopic images. Proc
SPIE 2006; 6055: 94103.
28. Lee S, Jung Y, Sohn H, Ro YM & Wook Park H. Visual dis-
comfort induced by fast salient object motion in stereo-
scopic video. Proc SPIE 2011; 7863: 112.
29. Woods A, Docherty T & Koch R. Image distortions in ste-
reoscopic video systems. Proc SPIE 1993; 1915: 113.
30. Dodgson NA. Variation and extrema of human interpupil-
lary distance. In: Stereoscopic Displays and Virtual Reality
Systems XI (Woods AJ, Merritt JO, Benton SA & Bolas MT,
editors), Proc SPIE 2004; 529.
31. Fukushima T, Torii M, Ukai K, Wolffsohn JS & Gilmartin B.
The relationship between CA/C ratio and individual differ-
ences in dynamic accommodative responses while viewing
stereoscopic images. J Vis 2009; 21: 113.
32. Sheedy JE, Hayes JN & Engle J. Is all asthenopia the same?
Optom Vis Sci 2003; 80: 732739.
33. Fabrigar LR, Wegener DT, MacCallum RC & Strahan EJ.
Evaluating the use of exploratory factor analysis in psycho-
logical research. Psychol Methods 1999; 4: 272299.
34. Horn JL. A rationale and test for the number of factors in
factor analysis. Psychometrika 1965; 30: 179185.
35. Patil VH, Singh SN, Mishra S & Todd Donavan D. Efficient
theory development and factor retention criteria: abandon
the ‘eigenvalue greater than one’ criterion. J Bus Res 2008;
61: 162170.
36. Walline JJ, Zadnik K & Mutti DO. Validity of surveys
reporting myopia, astigmatism, and presbyopia. Optom Vis
Sci 1996; 73: 376381.
37. Breslin KM, O’Donoghue L & Saunders KJ. An investigation
into the validity of self-reported classification of refractive
error. Ophthalmic Physiol Opt 2014; 34: 346352.
©2015 The Authors Ophthalmic & Physiological Optics ©2015 The College of Optometrists
Ophthalmic & Physiological Optics 35 (2015) 271–282
282
S3D movies and visual discomfort F Zeri and S Livi
... An excessive amount of retinal disparity between the two images places a large demand on the vergence system to fuse the two images. Image distortion is another factor related to the technical problems of the device [7]. ...
... It is already known from previous studies that watching 3D stereoscopic movies or video games without using a VR headset can still lead to both visual and nonvisual symptoms. Importantly, these symptoms may reflect certain oculomotor function changes [7,21] . Stone found that cybersickness symptoms were quite limited when 3D interactive video games played using VR headsets were combined with physical motion for about 20 minutes [8] . ...
Article
Full-text available
Purpose: Three-dimensional (3D) video games played using virtual reality head-mounted displays (VR-HMDs) are becoming increasingly popular. However, the use of this technology may lead to visual symptoms and discomfort by disrupting the normal linkage between the accommodation and vergence systems. This study aimed to investigate the effect of playing 3D video games using VR-HMDs on different oculomotor functions (i.e., accommodation and/or vergence system parameters) and to quantify symptoms associated with playing games using VR-HMD technology. Methods: Twenty-six male and female young adults (age range 19-27 years) with normal binocular vision completed this study. Different clinical accommodative and vergence measurements were collected before and after playing 3D video games using a VR-HMD system for 45 minutes. In addition, visual-and non-visual-related symptoms were measured before and after the video game sessions using the standardized Simulator Sickness Questionnaire. Results: The majority of accommodative parameters-including negative relative accommodation, accommodative accuracy, and monocular and binocular accommodative facilities-were significantly affected after playing 3D video games. With respect to vergence system measurements, the horizontal negative fusional vergence range at near and vergence facility test outcomes were affected significantly after the 3D video game sessions. Significant increments in different types of symptoms (i.e., visual and nonvisual) were also observed after playing 3D video games using the VR-HMD system. Conclusions: Playing 3D video games using VR-HMD systems can lead to a deterioration of certain oculomotor functions (i.e., accommodative and vergence systems). Players can be expected to experience eyestrain and discomfort after just 45 minutes of playing.
... As the opposite of experiencing a symptom is not conceivable (but would rather refer to not experiencing the symptom) a bipolar rating scale is not suitable. While the number of categories of the scales we build upon varies between 4 and 100, they are all unipolar scales (e.g., [40,86,95]). Furthermore, verbally labelling categories increases test-retest reliability [75]. ...
... Typically, symptom severity or pain scales are labelled from "none" to "severe", but differ in labelling the single categories in between [3,24,45]. Other labels are "no discomfort"/"very bad discomfort" [40] or "nothing"/"very much" [95]. We chose to stay consistent with typical symptom severity scales but decided to verbalize all seven categories, as this was found to facilitate participants' mental representation of the measured construct. ...
... Cybersickness describes several oculomotor symptoms. Questionnaires, like the virtual reality sickness questionnaire (VRSQ) designed for cybersickness Sevinc and Berkman 2020;Cid et al. 2021), or the simulator sickness questionnaire (SSQ) by Kennedy et al. (1993), list similar symptoms to those for visual fatigue like the computer vision syndrome questionnaire (CVS-Q) (Seguí et al. 2015;Sheppard and Wolffsohn 2018) or questionnaires developed for discomfort during stereoscopic viewing (Lambooij et al. 2007(Lambooij et al. , 2009Zeri and Livi 2015). Namely, "headache" and "blurred vision" are common symptoms reported for both states. ...
Article
Full-text available
This narrative review synthesizes and introduces 386 previous works about virtual reality-induced symptoms and effects by focusing on cybersickness, visual fatigue, muscle fatigue, acute stress, and mental overload. Usually, these VRISE are treated independently in the literature, although virtual reality is increasingly considered an option to replace PCs at the workplace, which encourages us to consider them all at once. We emphasize the context of office-like tasks in VR, gathering 57 articles meeting our inclusion/exclusion criteria. Cybersickness symptoms, influenced by fifty factors, could prevent workers from using VR. It is studied but requires more research to reach a theoretical consensus. VR can lead to more visual fatigue than other screen uses, influenced by fifteen factors, mainly due to vergence-accommodation conflicts. This side effect requires more testing and clarification on how it differs from cybersickness. VR can provoke muscle fatigue and musculoskeletal discomfort, influenced by fifteen factors, depending on tasks and interactions. VR could lead to acute stress due to technostress, task difficulty, time pressure, and public speaking. VR also potentially leads to mental overload, mainly due to task load, time pressure, and intrinsically due interaction and interface of the virtual environment. We propose a research agenda to tackle VR ergonomics and risks issues at the workplace.
... The compression, blur and frame-freeze distortions degrade videos in the stages of format conversion, representation, coding and transmission [8]. Zeri and Livi [9] interviewed 854 people. They recognized frequent symptoms of eye strain, blurred vision and a burning sensation after watching 3D films in movie theaters. ...
... Eye tracking measures can be interestingly be combined with Quality of experience assessment via questionnaires (Alexander et al., 2005;Hupont et al., 2015;Egan et al., 2016;Kong & Liu, 2019). In the VR learning context, Presence (Slater, 2009;Schroeder et al., 2017;Selzer et al., 2019), Flow (Csikzentmihalyi, 1990Wong & Csikszentmihalyi, 1991;Hamari & Koivisto, 2014;Kiili et al., 2014;Majaranta & Bulling, 2014;Sharek & Wiebe, 2014;Tozman et al., 2015), visual discomfort or oculomotor symptoms (Zeri & Livi, 2015; H. K. Kim et al., 2018;Caldas et al., 2020;Porcino et al., 2020), and subjective cognitive load (such as the NASA-TLX) (Hart & Staveland, 1988;Matthews et al., 2020) can be considered. Ultimately, users should feel a high presence and flow when learning in VR (S. ...
Article
Virtual Reality Head-Mounted Displays (HMDs) reached the consumer market and are used for learning purposes. Risks regarding visual fatigue and high cognitive load arise while using HMDs. These risks could impact learning efficiency. Visual fatigue and cognitive load can be measured with eye tracking, a technique that is progressively implemented in HMDs. Thus, we investigate how to assess visual fatigue and cognitive load via eye tracking. We conducted this review based on five research questions. We first described visual fatigue and possible cognitive overload while learning with HMDs. The review indicates that visual fatigue can be measured with blinks and cognitive load with pupil diameter based on thirty-seven included papers. Yet, distinguishing visual fatigue from cognitive load with such measures is challenging due to possible links between them. Despite measure interpretation issues, eye tracking is promising for live assessment. More researches are needed to make data interpretation more robust and document human factor risks when learning with HMDs.
... • Visual discomfort: measured via the questionnaire by Zeri and Livi (2015) consisting of 11 items. ...
Article
Full-text available
Purpose Do apparatuses and eyestrain have effects on learning performances and quality of experience? Materials and Methods: 42 participants played a serious game simulating a job interview with a Samsung Gear VR Head-Mounted Display (HMD) or a computer screen. Participants were randomly assigned to 3 groups: PC, HMD biocular, and HMD stereoscopy (S3D). Participants played the game thrice. Eyestrain was assessed pre- and post-exposure with six optometric measures. Learning performances were obtained in-game. Quality of experience was measured with questionnaires. Results: eyestrain was higher with HMDs than PC based on Punctum Proximum of accommodation but similar between biocular and S3D. Knowledge gain and retention were similar with HMDs and PC based on scores and response time. All groups improved response time but without statistically significant differences between HMDs and PC. Visual discomfort difference was statistically significant between PC and HMDs (biocular and S3D). Flow difference was statistically significant between PC and HMDs (biocular and S3D), with the PC group reporting higher Flow than HMD-S3D. Conclusion: short-term learning is similar between PC and HMDs. Groups initially using HMDs continued improving during long-term learning but without statistically significant difference compared to PC. Eyestrain and visual discomfort were higher with HMDs than PC. Flow was higher with the PC group. Our results show that eyestrain does not seem to decrease learning.
... Cybersickness describes several oculomotor symptoms. Questionnaires, like the Virtual Reality Sickness Questionnaire (VRSQ) designed for cybersickness Sevinc and Berkman 2020;Cid et al. 2021), or the Simulator Sickness Questionnaire (SSQ) by Kennedy et al. (1993), list similar symptoms to those for visual fatigue like the computer vision syndrome questionnaire (CVS-Q) (Seguí et al. 2015;Sheppard and Wolffsohn 2018) or questionnaires developed for discomfort during stereoscopic viewing (Lambooij et al. 2007(Lambooij et al. , 2009Zeri and Livi 2015). Namely, Pre-Print version "headache" and "blurred vision" are common symptoms reported for both states. ...
Preprint
INFINITY project EU Report (https://h2020-infinity.eu/) This document provides an evaluation of the potential impacts of the use of immersive technologies on the users: potential positive impacts as well as potential negative impacts. We have conducted literature analysis in the different topics and present a report of what has been previously described in various fields, including investigation activities when available. We consider impacts on 3 dimensions: cognition, health, and well-being. I am the author of section 3 "ERGONOMIC RISKS OF VIRTUAL REALITY FOR LAW ENFORCEMENT AGENCIES"
... Это соответствует данным, полученным на основе субъективных впечатлений зрителей в кинотеатре, которые отмечали после просмотра 3D-фильмов ощущение усталости, жжение глаз, головную боль и др. [13]. Полагают, что визуальный дискомфорт возникает у относительно небольшой части зрителей, и он также может быть связан с эффектом «ноцебо» от самих очков, особенно у женщин [14]. ...
Article
Full-text available
Different properties of displays, features of visual perception of three-dimensional images and other conditions, probably affect the effectiveness of motor rehabilitation when using a visual feedback channel and virtual reality technology. A brief review presents the latest publications on a choice of 2D or 3D displays. It is concluded that the presence of many features not only creates difficulties in comparing the effects of using various equipment, but also provides the potential for targeted display selection for a particular rehabilitation task.
Article
Digital eye strain (DES), caused by prolonged exposure to digital screens, stresses the visual system and negatively affects users’ well-being and productivity. While DES is well-studied in computer displays, its impact on users of virtual reality (VR) head-mounted displays (HMDs) is largely unexplored—despite that some of their key properties (e.g., the vergence-accommodation conflict) make VR-HMDs particularly prone. This work provides the first comprehensive investigation into DES in VR HMDs. We present results from a survey with 68 experienced users to understand DES symptoms in VR-HMDs. To help address DES, we investigate eye exercises resulting from survey answers and blue light filtering in three user studies (N = 71). Results demonstrate that eye exercises, but not blue light filtering, can effectively reduce DES. We conclude with an extensive analysis of the user studies and condense our findings in 10 key challenges that guide future work in this emerging research area.
Chapter
Full-text available
Evaluation is an important activity in software development since it allows detecting issues like design flaws under different analysis perspectives. However, conducting a satisfactory assessment is challenging, especially considering new interactive technologies such as virtual reality (VR). This paper presents a systematic review revealing the most used evaluation methods in this context to deepen the understanding of the evaluation of educational applications employing VR. Considering a search in the ACM database with a filter for the last six years, we selected 1351 initial studies and 81 for analysis. The results show that the tests of usability, flow, and technology acceptance model are the most common, as well as the methods used in many articles present weaknesses due to the lack of an adequate theoretical foundation. Accordingly, we strengthen the relevance of using well-defined support theories in all stages of an evaluation.
Chapter
Full-text available
The layout in most natural environments can be perceived through the use of nine or more sources of information. This number is greater than that available for the perception of any other property in any modality of perception. Oddly enough, how perceivers select and/or combine them has been relatively unstudied. This chapter focuses briefly on the issues inhibiting its study, and on what is known about integration, then in detail on an assessment of nine sources of information—occlusion, relative size, relative density, height in the visual field, aerial perspective, motion perspective, binocular disparities, convergence, and accommodation— and their relative utility at different distances. From a comparison of their ordinal depth-threshold functions, we postulate three different classes of distance around an observer-personal space, action space, and vista space. Within each space, we suggest a smaller number of sources act in consort, with different relative strengths, in offering the perceiver information about layout. We then apply this system to the study of representations of layout in art, to the development of the perception of layout by infants, and to an assessment of the scientific study of layout.
Article
Full-text available
Despite the widespread use of exploratory factor analysis in psychological research, researchers often make questionable decisions when conducting these analyses. This article reviews the major design and analytical decisions that must be made when conducting a factor analysis and notes that each of these decisions has important consequences for the obtained results. Recommendations that have been made in the methodological literature are discussed. Analyses of 3 existing empirical data sets are used to illustrate how questionable decisions in conducting factor analyses can yield problematic results. The article presents a survey of 2 prominent journals that suggests that researchers routinely conduct analyses using such questionable methods. The implications of these practices for psychological research are discussed, and the reasons for current practices are reviewed. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Full-text available
3D display technologies have been linked to visual discomfort and fatigue. In a lab-based study with a between-subjects design, 433 viewers aged from 4 to 82 years watched the same movie in either 2D or stereo 3D (S3D), and subjectively reported on a range of aspects of their viewing experience. Our results suggest that a minority of viewers, around 14%, experience adverse effects due to viewing S3D, mainly headache and eyestrain. A control experiment where participants viewed 2D content through 3D glasses suggests that around 8% may report adverse effects which are not due directly to viewing S3D, but instead are due to the glasses or to negative preconceptions about S3D (the ‘nocebo effect'). Women were slightly more likely than men to report adverse effects with S3D. We could not detect any link between pre-existing eye conditions or low stereoacuity and the likelihood of experiencing adverse effects with S3D. Practitioner Summary: Stereoscopic 3D (S3D) has been linked to visual discomfort and fatigue. Viewers watched the same movie in either 2D or stereo 3D (between-subjects design). Around 14% reported effects such as headache and eyestrain linked to S3D itself, while 8% report adverse effects attributable to 3D glasses or negative expectations.
Article
Existing stereoscopic imaging algorithms can create static stereoscopic images with perceived depth control function to ensure a compelling 3D viewing experience without visual discomfort. However, current algorithms do not normally support standard Cinematic Storytelling techniques. These techniques, such as object movement, camera motion, and zooming, can result in dynamic scene depth change within and between a series of frames (shots) in stereoscopic cinematography. In this study, we empirically evaluate the following three types of stereoscopic imaging approaches that aim to address this problem. (1) Real-Eye Configuration: set camera separation equal to the nominal human eye interpupillary distance. The perceived depth on the display is identical to the scene depth without any distortion. (2) Mapping Algorithm: map the scene depth to a predefined range on the display to avoid excessive perceived depth. A new method that dynamically adjusts the depth mapping from scene space to display space is presented in addition to an existing fixed depth mapping method. (3) Depth of Field Simulation: apply Depth of Field (DOF) blur effect to stereoscopic images. Only objects that are inside the DOF are viewed in full sharpness. Objects that are far away from the focus plane are blurred. We performed a human-based trial using the ITU-R BT.500-11 Recommendation to compare the depth quality of stereoscopic video sequences generated by the above-mentioned imaging methods. Our results indicate that viewers' practical 3D viewing volumes are different for individual stereoscopic displays and viewers can cope with much larger perceived depth range in viewing stereoscopic cinematography in comparison to static stereoscopic images. Our new dynamic depth mapping method does have an advantage over the fixed depth mapping method in controlling stereo depth perception. The DOF blur effect does not provide the expected improvement for perceived depth quality control in 3D cinematography. We anticipate the results will be of particular interest to 3D filmmaking and real time computer games.
Conference Paper
We develop an algorithm that predicts the best presentation of a stereo 3D image in the sense of viewers' preference. The algorithm operates in three steps. First, the 3D image is classified as either a “foreground dominant” or “background dominant” image. Next, for “foreground dominant” images, a model of the stereoacuity function is used to optimize the perceptual 3D resolution; for “background dominant” images, the nearest surface is placed in the 3D plane of the display screen. A human study was conducted to assess the algorithm and showed that the proposed model produced 3D images which had the best 3D quality scores among several candidate algorithms.
Article
To assess the validity of questionnaire use in the self-identification of refractive status. Two hundred and forty adults (21-60 years of age) presenting for a routine eye examination at various optometric practices in Northern Ireland were asked to complete one of two questionnaires. Both questionnaires used identical questions to ascertain age, gender, current spectacle use, age of first spectacle use and level of education. For the identification of refractive status, Questionnaire 1 used layman's terminology whilst Questionnaire 2 combined optometric terminology with descriptive explanations. Current refractive status was identified by the examining optometrist who did not see the completed questionnaire. The spherical equivalent refractive error of the non-cycloplegic subjective refraction was used to categorise myopia as <0D and hyperopia as ≥+1.00D. Astigmatism was defined according to two different criteria: ≥0.50DC and ≥1.00DC. Questionnaire 1 had a sensitivity of 0.63 and a specificity of 0.90 for identifying myopia; a sensitivity of 0.58 and a specificity of 0.71 for identifying hyperopia; a sensitivity of 0.12 and a specificity of 0.98 for identifying astigmatism ≥0.50DC and a sensitivity of 0.19 and a specificity of 0.95 for identifying astigmatism ≥1.00DC. Questionnaire 2 had a sensitivity of 0.83 and a specificity of 0.93 for identifying myopia; a sensitivity of 0.45 and a specificity of 0.86 for identifying hyperopia; a sensitivity of 0.32 and a specificity of 0.88 for identifying astigmatism ≥0.50DC and a sensitivity of 0.50 and a specificity of 0.84 for identifying astigmatism ≥1.00DC. For both questionnaires, altering a positive self-identification of myopia to include only those who had worn spectacles prior to age 30 reduced the sensitivity and increased the specificity slightly. Questionnaires are a valid tool in self-identification of myopic refractive status. However, they are not an effective way of identifying hyperopia and astigmatism and objective or subjective refraction remains the most appropriate method of identifying such individuals.
Article
This paper investigates visual discomfort induced by fast motion of salient object in a stereoscopic video. We have conducted a subjective assessment to investigate the degree of visual discomfort caused by motion characteristics of a controlled graphics object in a video scene. As results of the subjective assessment, we observe the changes of the degree of visual discomfort with varying velocity and direction of object motion. In order to verify the acceptability of our observation for real stereoscopic 3D videos, we exploit the concept of visual saliency to define the salient object motion severely affecting the degree of visual discomfort in a video scene. The salient object motion feature is extracted and a visual comfort model is derived from our observation. Then we predict the degree of visual discomfort by using the extracted motion feature and the visual comfort model. We have conducted a subjective test to compare the predicted visual comfort score with actual subjective score. The experiment results show that the predicted visual comfort score correlates well with the actual subject score.
Article
Nowadays, crosstalk is probably one of the most annoying distortions in 3D displays. So far, display designers still have a relative lack of knowledge about the relevant subjective attributes of crosstalk and how they are combined in an overall 3D viewing experience model. The aim of the current experiment is to investigate three perceptually important attributes influencing the overall viewing experience: perceived image distortion, perceived depth, and visual strain. The stimulus material used in this experiment consisted of two natural scenes varying in depth (0, 4, and 12cm camera base distance) and crosstalk level (0, 5, 10, and 15%). Subjects rated the attributes according to the ITU BT.500–10 in a controlled experiment. Results show that image distortion ratings show a clear increase with increasing crosstalk and increasing camera base distance. Especially higher crosstalk levels are more visible at larger camera base distances. Ratings of visual strain and perceived depth only increase with increasing camera base distance and remain constant with increasing crosstalk (at least until 15% crosstalk).