Chapter 6 Results and Analysis
6.3 Analysis
6.3.2 Analysis of the Subjective Indicator (Questionnaire Method)
Fig. 6-1 shows the 10 subjects’ questionnaire results. After 4 experiments, the mean difference score for each subject in each mode is calculated by taking the 15 scores for the symptoms after viewing the display minus the before scores and dividing them by 4. Because the individual subjective cognition is quite different, as shown in Fig. 6-1, it is hard to draw a conclusion or compare the different visual fatigue levels with each mode. Thus, the subjects’ total score ranking for each symptom in each mode is applied to the following analyses on the relationship between individual differences and visual fatigue. In the figure, the red line labels the largest scores in each mode for S3D, P3D and LP at above four. A maximum point exists in P3D.
-1 0 1 2 3 4 5 6 7 S3D
S2D
P3D
P2D
LP
A B C D E F G H I J
Figure 6-1: The result of the questionnaires from 10 subjects. The outer
point means that discomfort level caused by the mode is more serious.Fig. 6-2 shows the mean scores for symptoms from the ten subjects caused by each mode. The red line labels the highest scores for each symptom, (1) eyestrain, (2) dry eyes which are much more serious and common, (3) too bright, (5) feeling of pressure in the eyes, (7) blurred vision, (11) dazed feeling, (14) sleepy feeling and (15) difficulty concentrating, are also common symptoms.
However, it is still hard to compare the effects of each variable. So the total ranking of mean scores for each symptom in the modes was used in the following analyzes of the relationship between each mode and visual fatigue.
-0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 symptoms of numbers are: (1) eyestrain, (2) dry eyes, (3) too bright, (4) eyelid twitching, (5) feeling of pressure in the eyes, (6) ache behind the eyes, (7) blurred vision, (8) headache (9) head feels heavy, (10) head hurts when shaken (11) dazed feeling, (12) irritated feeling, (13) stiff shoulders, (14) sleepy feeling and (15) difficulty concentrating.
The total ranking for mean scores of each symptom in the modes are ranked 5 (high) to 1 (low) for all 5 modes, and the total ranking for each mode is derived by summing up the 15 ranks, minus 15x2.5=37.5 (average), which was used to compare the effect on visual fatigue caused by different variables.
Fig. 6-3 shows the total ranking for each mode. The rankings for S3D and P3D are higher than the S2D, P2D and LP. Then S2D, LP and P2D are put in descending order from high to low. But there are no differences between S3D and P3D.
Figure 6-3: The total ranking for mean scores of each mode.
Fig. 6-4 shows the mean total rankings in the 3D and 2D modes, the LCD and DLP, and the shutter and polarized. The mean total rank in the 3D mode is higher than the 2D mode, but there are no differences between the LCD and DLP, or the shutter and polarized. It shows that 3D videos cause more visual fatigue than 2D ones do.
Figure 6-4: The mean total ranking for mean scores under 3D/2D mode
(white), LCD/DLP (gray) and shutter/polarized (black).To compare the relationship between individual differences and visual fatigue in each mode, the 10 subjects’ total score for each symptom in each mode is ranked from 10 (high) to 1 (low), and each subject’s total for each mode sums up the 15 ranks in each mode minus 15x5=75 (average).
Fig. 6-5 shows the mean total ranking for all subjects, and subjects under the 3D and 2D modes grouped by chromatic discrimination. The mean total ranking of subjects with normal chromatic discrimination and the ranking in the 3D mode are higher than those subjects with good chromatic discrimination.
Subjects with good and normal chromatic discrimination power in the 2D mode are ranked almost the same. But there are no obvious results of comparison due to large standard errors.
Figure 6-5: The mean total ranking for scores of all subjects (white),
subjects under 3D mode (gray) and 2D mode (dark gray) grouped by chromatic discrimination.Fig. 6-6 shows the subjects’ mean total ranking under the shutter and polarized systems when grouped by chromatic discrimination. The ranking of subjects with normal chromatic discrimination under the shutter system is higher than those with good chromatic discrimination, but there is no obvious difference between subjects with either good or normal chromatic discrimination power when viewing 3D videos with polarized glasses. But there is also no obvious result of comparison due to large standard errors.
Figure 6-6: The mean total ranking for scores of subjects under shutter
system (white) and polarized system (gray) grouped by chromatic discrimination.Fig. 6-7 shows the mean total rankings for subjects in the LCD and DLP grouped by chromatic discrimination. The rankings of subjects with good chromatic discrimination under the DLP is higher than those with normal chromatic discrimination, while the opposite is true with the LCD where the ranking of subjects with good chromatic discrimination is lower than those with normal chromatic discrimination power. Due to large standard errors, there are no obvious results of comparison, too.
Figure 6-7: The mean total ranking for scores of subjects under LCD
(white) and DLP (gray) grouped by chromatic discrimination power.According above those results, it is hard to conclude that the effects of conditions on viewers grouped by chromatic discrimination because of large standard errors. So, changing analysis method might is needed to improve the effectiveness of the questionnaire.
It is noticed that among the 15 symptoms on the questionnaire only 8 are obvious: (1) eyestrain, (2) dry eyes are much more serious and common, (3) too bright, (5) feeling of pressure in the eyes, (7) blurred vision, (11) feeling dazed, (14) feeling sleepy and (15) difficulty concentrating. After evaluation by their mean scores compared to the total mean score, 6 major symptoms,(1) eyestrain, (2) dry eyes are much more serious and common, (3) too bright, (7) blurred vision, (14) feeling sleepy and (15) difficulty concentrating, are chosen to evaluate the effects on viewers by different variables, as shown as Fig. 6-8.
Figure 6-8: The comparison between the mean score of each symptom
and the mean scores of 15 symptoms. The symptoms of numbers are: (1) eyestrain, (2) dry eyes, (3) too bright, (4) eyelid twitching, (5) feeling of pressure in the eyes, (6) ache behind the eyes, (7) blurred vision, (8) headache (9) head feels heavy, (10) head hurts when shaken (11) dazed feeling, (12) irritated feeling, (13) stiff shoulders, (14) sleepy feeling and (15) difficulty concentrating.The total ranking for mean score of 6 major symptoms in the modes are ranked 5 (high) to 1 (low) for all 5 modes, and the total ranking for each mode is derived by summing up the 6 ranks, minus 6x2.5=15 (average), which was used to compare the effect on visual fatigue caused by different variables.
Fig. 6-9 shows the mean total rankings in the 3D and 2D modes, the LCD and DLP, and the shutter and polarized by 6 major symptoms. The mean total rank in the 3D mode is higher than the 2D mode, in the DLP is higher than LCD, and in the shutter is higher than polarized. It shows that 3D videos cause more visual fatigue than 2D ones do. The DLP systems cause more visual fatigue than LCD and the shutter cause more visual fatigue than the polarized does.
Figure 6-9: The mean total ranking for mean scores under 3D/2D mode
(white), LCD/DLP (gray) and shutter/polarized (black) by 6 major symptoms.To compare the relationship between individual differences and visual fatigue in each mode, the 10 subjects’ total score for 6 major symptoms in each mode is ranked from 10 (high) to 1 (low), and each subject’s total for each mode sums up the 6 ranks in each mode minus 6x5=30 (average).
Fig. 6-10 shows the mean total ranking by 6 major symptoms for all subjects, and subjects under the 3D and 2D modes grouped by chromatic discrimination. The mean total ranking of subjects with normal chromatic discrimination (p=0.13) and the ranking in the 3D (p=0.18) and 2D (p=0.27) mode are higher than those subjects with good chromatic discrimination. Thus, when viewing 3D and 2D videos, subjects with normal chromatic discrimination power generally suffer more visual fatigue than those with good chromatic discrimination.
Figure 6-10: The mean total ranking for scores by 6 major symptoms of
all subjects (white), subjects under 3D mode (gray) and 2D mode (dark gray) grouped by chromatic discrimination.Fig. 6-11 shows the subjects’ mean total ranking by 6 major symptoms under the shutter and polarized systems grouped by chromatic discrimination.
The ranking of subjects with normal chromatic discrimination under the shutter system is higher than those with good chromatic discrimination (p=0.11), but there is no obvious difference between subjects with either good or normal chromatic discrimination when viewing 3D videos with polarized glasses (p=0.49).
Figure 6-11: The mean total ranking for scores by 6 major symptoms of
subjects under shutter system (white) and polarized system (gray) grouped by chromatic discrimination.Fig. 6-12 shows the mean total rankings by 6 major symptoms for subjects in the LCD and DLP grouped by chromatic discrimination power. The rankings of subjects with good chromatic discrimination under the DLP is higher than those with normal chromatic discrimination (p=0.19), while there is no obvious different effect by the LCD on subjects with normal chromatic discrimination and those with good chromatic discrimination (p=0.43).
Thus, in general subjects with normal chromatic discrimination suffer more visual fatigue than those with good chromatic discrimination. Besides, subjects with normal chromatic discrimination suffer more from visual fatigue while viewing 3D videos with shutter glasses and viewing 2D videos on the DLP system.