Sex-Specific Lateralization of Contraction Anisocoria in Transient Pupillary Light Reflex

Abstract

purpose. Contraction anisocoria describes a phenomenon in which the pupil of a directly illuminated eye constricts more than the pupil of the consensual (not illuminated) eye. The purpose of this study was to investigate the lateralization of contraction anisocoria in young female and male subjects.

methods. Infrared binocular pupillography was used to measure pupillary light reflex (PLR) in 44 healthy children (23 girls, 21 boys) from 6 to 16 years of age. Measurements were conducted in both light-adapted and dark-adapted conditions with different stimulus intensities. Relative constriction amplitude was obtained by dividing the maximal pupil area change by the initial static pupil area. Contraction anisocoria was calculated by subtracting relative constriction amplitude in the consensual eye from that of the direct eye. Values of contraction anisocoria obtained by stimulating a subject’s right or left eye were compared to determine any potential lateralization.

results. It was found that stimulating the right eye led to larger contraction anisocoria than stimulating the left eye. Such right-side lateralization of contraction anisocoria is much greater in males than in females. In addition, the effects of sex were related to the ambient light level and stimulus intensity.

conclusions. These results provide evidence that contraction anisocoria is more laterally asymmetric in males than in females.

Sex-specific dimorphism is observed in many brain structures, such as the corpus callosum, splenium, and neocortex, as revealed in positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) studies. A recent study by Amunts et al. ¹ shows the same sexually dimorphic pattern exists in human visual cortex. They reported that the male visual cortices are more asymmetrical than female and that they are lateralized toward the right hemisphere. Similarly, hemispheric lateralization and sex effects have been observed in autonomic cardiovascular control. ^{2 3} Such sex-specific trends in lateralization may be a common feature for both cortical and subcortical physiological functions. 

In this study, we investigated sex-specific lateralization of the pupillary light reflex (PLR), the phenomenon by which pupil size changes with luminance. Pupil size is controlled by two antagonistic muscles, the sphincter and the dilator, which are innervated by the parasympathetic and sympathetic nervous systems. ⁴ In the classic view, the constriction process involved in PLR is mediated by the parasympathetic pathway, involving the Edinger-Westphal (EW) nucleus, the ciliary ganglion, and the short ciliary nerves. The pupilloconstrictor neurons in the EW nucleus act as a signal generator whose output is modulated by a number of inputs; the major one is the afferent projection from the retina through the pretectal olivary nucleus (PON) in midbrain. As an output of multiple neuronal systems, PLR is a precise measure of neural network functions, and it has been applied in many neurology and psychophysiology studies. ^{5 6 7 8 9 10}  

Here, we focus our study on the phenomenon of contraction anisocoria in healthy male and female subjects. Contraction anisocoria refers to the difference between the pupillary constrictions in the stimulated eye (direct) and the contralateral eye (consensual). It was initially considered as an abnormality in the reflex pathway, ^{11 12} but subsequent studies have proved that a small yet statistically significant difference in constriction amplitude between the direct and consensual eyes exists in healthy subjects. ^{13 14 15 16 17} The small contraction anisocoria in healthy subjects using full-field stimulation has been hypothetically attributed to the asymmetries at the two decussations, ^{13 14 15 16 17} as illustrated in Figure 1 . The first occurs at the optical chiasm, where the ratio of crossed and uncrossed optic nerve fibers is estimated to be approximately 53:47, as reported in a histologic study in humans. ¹⁸ The second decussation takes place in the midbrain, where each PON projects predominantly to the contralateral side. ^{19 20 21} Overall, the EW nucleus at the same side of the stimulated (direct) eye receives a stronger signal, thus inducing a greater constriction. However, the specific distribution of the pupil-serving retinal ganglion cells and the associated decussations at the chiasm and further in the midbrain are still not thoroughly known. ¹⁶  

To investigate any potential lateralization in contraction anisocoria, we compared contraction anisocoria obtained by stimulating the left and right eyes separately. The effects of sex and the side of the stimulated eye were analyzed to determine whether contraction anisocoria has a sex-specific pattern in the lateralization of the response. 

Methods

Subjects

Forty-four healthy children (23 girls, 21 boys) participated in this study. The subjects ranged from 6 to 16 years of age (10.4 ± 2.7 [mean ± SD]). The detailed age distribution is shown in Figure 2 . All subjects had normal vision, including visual acuity or correctable visual acuity at 20/20 or better. All were in good physical health and lacked any history of eye disorders. Most of the participants (19 girls, 18 boys) were tested between 2:00 pm and 6:30 pm, and the other seven (3 boys, 4 girls) were tested between 10:00 am and 12:00 pm. 

All participants were recruited from Columbia, Missouri, or its surrounding areas. The study adhered to the tenets of the Declaration of Helsinki. The experimental protocol was approved by the university institutional review board. All participants and their parents signed informed consent forms before the tests. 

PLR Apparatus

As shown in Figure 3 , the pupillogram recording system consists of two independent pupil-imaging channels for each eye. Each channel has an illumination port and an imaging port. The two ports are combined with a cold mirror that reflects visible light and transmits infrared light. 

A green LED of 530 nm (LXK2-PD12-R00; Lumileds Lighting, San Jose, CA) was used in each channel as the stimulation light source. The LEDs illuminated an optical diffuser to provide a uniform on-axis illumination with a visual field of 5.7°. A wide range of stimulation light intensity from 10^7.4 to 10^13.6 photons/cm²/s was achieved by controlling the LED current and using neutral density filters. Two different intensities of 10^9.3 and 10^12.5 photons/cm²/s were actually used in this study. All intensities were calibrated with a photograph diode (818-ST-UV/CM; Newport Corp., Irvine, CA) coupled with a power meter (2835-C; Newport Corp.) located at the same location as the subject’s eye. The pulse width (100 ms) of the stimulation was programmed through a custom computer interface. 

An infrared LED of 880 nm wavelength (604-WP34SF4C; Mouser Electronics, Mansfield, TX) was installed in each imaging channel to illuminate the subject’s pupil. Video cameras (PC164CEX-2; Supercircuits Inc., Austin, TX) with high sensitivity in the infrared range were used. The magnification of the imaging system was calibrated by imaging a standard calibrated ruler. Pupil images were captured by a video card at 30 frames/s. The two cameras (one in each channel) were synchronized by use of a common trigger signal. 

Procedure

Measurements were conducted in light-adapted and dark-adapted conditions. Subjects were first seated in the testing room for at least 5 minutes. The room had an illumination level of 30 cd/m². When instructed, each subject placed his or her head into a chin-and-forehead rest located on the height-adjustable platform. Before each measurement, the subjects had 10 to approximately 15 seconds to look into the imaging channels for adaptation. To minimize eye movement during tests, subjects were asked to look at pictures (cartoon images) shown on a computer screen 4 feet away through the imaging channels in our PLR device (Fig. 3) . The pictures on the screen were switched every 30 seconds. The color of the screen was adjusted to pure red by disabling the green and blue channels, and its brightness and contrast were reduced (0.01∼0.02 cd/m²) to avoid affecting pupil size. After finishing all the light-adapted tests, all lights in the testing room were turned off. Subjects stayed in the dark room for 20 minutes for dark adaptation before the tests were repeated. 

Accommodation effects were not monitored in this study. Subjects were instructed to stare at the monitor during the tests. Because of the short light flash (100 ms) applied, the chance of the fixation on the flash may not be significant. Indeed, the pupillary constriction amplitude was strongly dependent on stimulus intensities and adaptation in our tests, suggesting our measured constrictions were induced by the stimulation light. 

In the light-adapted tests, a stimulus intensity of 10^12.5 photons/cm²/s was applied. In the dark-adapted tests, two stimulus intensities of 10^9.3 and 10^12.5 photons/cm²/s were applied. Only one eye was stimulated in each test. Under each stimulus condition, the left eye was stimulated first, and then the right eye was stimulated. PLRs of the direct and consensual eyes were acquired at the same time. Image recording started 1 second before the stimulus and lasted for 10 seconds. Four measurements were conducted at each testing condition, and the interval between two consecutive measurements was 1.5 minutes. Between the consecutive measurements, the subject was free to move his or her head from the testing device. 

PLR Parameters

Pupil size was automatically computed with the use of custom software from saved image stacks. Briefly, the pixels located at the pupil boundary in each image were first determined using a histogram-based segmentation method. A fast elliptic-fitting algorithm was then applied to fit the entire pupil boundary, and the pupil area was calculated as the area of the fitted ellipse. Average pupil diameter could be computed if the pupil area was approximated as a circle. 

The initial static pupil size was obtained by averaging the pupil sizes calculated from the 30 images acquired before the stimulus onset. Anisocoria was calculated by subtracting the left pupil diameter from the right pupil diameter. Relative constriction amplitude was obtained by dividing the maximal pupil area change (i.e., the difference between the initial static pupil area and the minimum pupil area after stimulus) by the initial static pupil area. Although some previous studies ^{13 16} used the raw pupil size to calculate contraction anisocoria, our data suggested that using the relative percentage constriction can minimize any variations caused by the difference in the initial pupil size. Contraction anisocoria was then derived by subtracting relative constriction amplitude in the consensual eye from that of the direct eye. Therefore, negative contraction anisocoria indicated the relative constriction amplitude was larger in the consensual eye than in the direct eye. 

Statistical Analysis

Distributions of relative constriction amplitude, anisocoria, and contraction anisocoria obtained in all participants were verified to conform to normal distributions using the Kolmogorov-Smirnov test. A mixed-factor analysis of variance (ANOVA) with repeated measures was then applied to study sex as the between-subjects factor. In addition, ambient illumination level was used as a within-subjects factor for anisocoria, whereas the eye stimulated and the stimulus condition (including ambient light level and stimulus intensity) were used as within-subjects factors to study contraction anisocoria. To analyze relative constriction amplitude, the eye measured (direct or consensual eye) was combined with the factor of eye stimulated and was included as a within-subject factor in the ANOVA model, and the interactions were examined for all major factors. ANOVA was carried out with mixed procedure (SAS version 9.1.3; SAS Institute Inc., Cary, NC). Follow-up post hoc t-tests were then carried out in all the interested pairwise comparisons. Statistics from multiple comparison procedures were Bonferroni corrected by multiplying the obtained P values by the number of pairwise comparisons. P < 0.05 was considered significant. 

Results

Anisocoria

Table 1summarizes the average pupil diameter of the two eyes and the calculated anisocoria values. All results are shown in the form of mean ± SE. To show the data distribution, all measurements including outliers are shown in a box plot in Figure 4A . Because measurements obtained when stimulating the left or the right eye can be used to calculate initial pupil size, the total number of measurements used in each subject to calculate anisocoria were 8 and 16 at light-adapted and dark-adapted conditions, respectively. Of all measurements in the 44 subjects, only three single measurements, each from a different subject, were excluded from the calculation because they fell more than three times the interquartile range above the third quartile or below the first quartile. ²²  

Mean anisocoria was not significantly greater than zero in dark-adapted or light-adapted conditions. ANOVA results indicated no significant effects from sex, ambient light level, or their interaction (P = 0.0999, P = 0.6788, and P = 0.2936, respectively). Figure 4Bshows the distribution of anisocoria measured in light-adapted and dark-adapted conditions. The interquartile ranges (25th–75th percentiles) of the standard deviation for individual subject (within-subject variance) were 0.23 mm to 0.38 mm. It can be seen that the measurements obtained at the two different adaptation conditions were highly correlated (r = 0.88). 

Constriction Amplitude

Relative constriction amplitude depends on stimulus intensity and adaptation. As expected, relative constriction amplitude was larger in the dark-adapted conditions than in the light-adapted conditions, and constriction increased with stimulus intensity. Box plots of all measured constriction amplitudes are shown in Figure 5 . Interquartile ranges (25th–75th percentiles) of standard deviations for individual subjects were 2.71% to 4.85%. Table 2summarizes the values (in the form of mean ± SE) of direct and consensual relative constriction amplitude measured in female and male subjects. The main effect of stimulus condition was highly significant (P < 0.0001). 

The “effect of eye” (a combination of the eye stimulated and the eye measured) was significant (P < 0.0001). Multiple pairwise comparisons were performed to investigate the difference in constriction amplitude observed by stimulating different eyes. Constriction amplitudes in the direct eye (stimulated) were significantly larger in the right eyes than in the left eyes (P < 0.0011). Conversely, constriction amplitudes showed no statistical difference in the consensual eyes (P = 0.9578). These two P values were Bonferroni corrected by multiplying the raw P values by 2. 

We also observed a significant interaction of sex by stimulus condition (P < 0.0001) in ANOVA. However, post hoc pairwise t-tests indicated no significant sex difference in any of the three stimulus conditions (P = 1.0000, P = 1.0000, and P = 0.1601 for light-adapted 10^12.5, dark-adapted 10^9.3, and 10^12.5 photons/cm²/s, respectively). All P values were Bonferroni corrected by multiplying by 3. 

As discussed, relative constriction amplitude was used in our analysis to reduce any variations caused by potential differences in initial pupil size. As a quantitative comparison, the obtained constrictions were 0.86 ± 0.38 mm, 0.76 ± 0.31 mm, and 2.12 ± 0.31 mm (mean ± SD) at light-adapted 10^12.5, dark-adapted 10^9.3, and 10^12.5 photons/cm²/s, respectively. Relative variations (the ratio of the SD and the mean value) were 41.6%, 40.5%, and 14.6% at these three conditions. If the relative constriction amplitude was used, the calculated relative variations were reduced to 33.1%, 31.4%, and 11.3%, respectively. Similar results were obtained for pupil area calculations. Nevertheless, the same statistical conclusion was reached when using the absolute constriction amplitude in our ANOVA. 

Contraction Anisocoria

Contraction anisocoria measured in boys and girls by stimulating left and right eyes at three stimulus conditions are summarized in Table 3 . Values of contraction anisocoria are shown in the form of mean ± SE. P values indicate how significantly the measurements are different from zero. Four measurements of each subject were used to calculate the contraction anisocoria for each stimulated eye at each of the three test conditions. In all measurements of the 44 subjects, three single measurements from two subjects were excluded because the measurements were extreme outliers. ²²  

We found the mean of contraction anisocoria was not significantly greater than zero in all subjects when the left eye was stimulated at light-adapted 10^12.5 photons/cm²/s (P = 0.0606) and in male subjects at dark-adapted 10^9.3 photons/cm²/s (P = 0.4105). In all other conditions, contraction anisocoria was significant, and constrictions in the direct eye were approximately 0.4% to 2.8% larger than in the consensual eye. 

Each subject could be classified into one of three categories based on contraction anisocoria. ¹³ According to the criterion described by Smith et al., ¹³ contraction anisocoria exists when the difference of pupil diameter change in the two eyes is greater than twice the spatial resolution, which was 0.06 mm in our imaging system. A subject has bilateral contraction anisocoria if contraction anisocoria always exists regardless of which eye is stimulated. Unilateral contraction anisocoria means contraction anisocoria existed only when one eye was stimulated. If the contraction anisocoria obtained by stimulating neither side was greater than the threshold, the subject was classified as having no contraction anisocoria. Results are shown in Table 4 . Overall, 61.4%, 75.0%, and 79.5% subjects showed contraction anisocoria at the three stimulus conditions of light-adapted 10^12.5 photons/cm²/s, dark-adapted 10^9.3 photons/cm²/s, and 10^12.5 photons/cm²/s, respectively. 

Sex Difference in Contraction Anisocoria

Figure 6shows the distributions of contraction anisocoria in each subject obtained by stimulating different eyes at three stimulus conditions. The interquartile range (25th-75th percentile) of the measurement SD for individual subjects was 1.04% to 2.36%. In girls, the distribution of contraction anisocoria was similar under dark-adapted conditions regardless of which eye was stimulated, which is also shown by the population distribution in Table 4 . In contrast, asymmetry of contraction anisocoria was readily observed in boys, especially at dark-adapted 10^9.3 photons/cm²/s. Overall, contraction anisocoria in boys was larger than in girls. F statistics in ANOVA indicated that the effect of sex (P = 0.0301), the interaction of sex by stimulated eye (P < 0.0001), and the interaction of sex by stimulated eye by stimulus condition (P = 0.0088) were all significant. 

To further illustrate the sex-related differences, the averaged values (Table 3)of contraction anisocoria in boys and girls are shown in Figure 7 . There was no significant difference between eyes in contraction anisocoria in girls. In boys, stimulating the right eye produced a significantly larger contraction anisocoria than stimulating the left eye under a dark-adapted stimulus of 10^9.3 photons/cm²/s (right vs. left, 2.90% vs. −0.26%; P < 0.0001). Such an asymmetry pattern in boys also appeared at the other two conditions—light-adapted 10^12.5 photons/cm²/s (right vs. left, 1.40% vs. 0.42%) and dark-adapted 10^12.5 photons/cm²/s (right vs. left, 2.12% vs. 1.21%), though the difference was not statistically significant (Fig. 7) . Furthermore, under dark-adapted conditions, contraction anisocoria in boys was significantly larger than in girls when the right eyes were stimulated (at 10^9.3 photons/cm²/s: 2.90% [boys] vs. 0.90% [girls], P = 0.0002; at 10^12.5 photons/cm ² /s: 2.12% [boys] vs. 0.73% [girls], P = 0.0232). All these P values were Bonferroni corrected by multiplying the raw P values by 12. Such a right-over-left preponderance in boys can also be seen in Table 4 . 

Discussion

Several studies have revealed that contraction anisocoria can be detected in healthy subjects. Smith et al. ¹³ were the first to report a small but significant contraction anisocoria in healthy subjects (∼6.1% of constriction amplitude). In their study of 72 subjects, 84.7% showed contraction anisocoria. They also found that contraction anisocoria was not influenced by stimulus intensity when expressed as a percentage of constriction amplitude. 

Our stimulus configuration provided illumination over the macular region only. By stimulating the right and left eyes separately, we made several interesting observations. First, contraction anisocoria seemed to be lateralized to the right eye because its mean value was always significantly larger than zero when the right eye was stimulated, but this was not true when the left eye was stimulated (Table 3 , summary). Under most stimulus conditions, stimulating the right eye led to larger contraction anisocoria than did stimulating the left eye (Tables 3 4) . In addition, we found that such right-side lateralization of contraction anisocoria was much greater in male subjects than in female subjects. 

Right lateralization of constriction amplitude in the direct eye (stimulated) was also observed by Bär et al. ²³ The same results were obtained in the present study (Table 2) . In addition, our results indicated that such a difference was not present in the consensual eye, which was not measured in the study of Bär et al. ²³ Because pupillary constriction is primarily controlled by the parasympathetic component of the autonomic nervous system, these results imply a lateralization of autonomic functions. In fact, several studies ^{2 23} have confirmed the hemispheric lateralization of autonomic controls in cardiovascular functions. 

Our results cannot be explained by the existing knowledge on contraction anisocoria because neither sex- nor stimulus intensity-related lateralization has been described at the two decussations at the optical chiasm and the PON. However, the potential involvement of cortical level pathways in PLR ^{24 25} is worth some discussion. Barbur ⁴ suggested that there might be a projection from the primary visual cortex (V1) to the PON and a projection from extrastriate visual cortical areas to the EW nucleus. A recent anatomic study ¹ has shown that the male visual cortex is rightward biased, whereas the female visual cortex is more symmetric. Such a difference in the visual cortex may be related to the asymmetric contraction anisocoria observed in male subjects. In addition, the topographic segregation of the signal input from rods and cones at human visual cortex caused by retinotopic mapping has been reported. ²⁶ It is known that rods are active at dark-adapted conditions but are suppressed by high-intensity stimuli. On the other hand, cones become active under light-adapted conditions or when stimulated by high-intensity light. Coincidentally, our observed sex effect seems related to the ambient light level and stimulus intensity. At dark-adapted 10^9.3 photons/cm²/s, the sexually dimorphic pattern of contraction anisocoria is significant. However, this trend is largely reduced at dark-adapted 10^12.5 photons/cm²/s and light-adapted 10^12.5 photons/cm²/s. If the cortical areas receiving signals from rods and cones make different contributions to the PLR pathway, it may explain the observed illuminance-dependent difference in contraction anisocoria. 

We would like to point out that the dominant eye effect was not examined in our subjects. It is not likely that our testing results were biased by such effects. Several comprehensive studies have shown that the distribution of the dominant eye has no correlation with sex or age ^{27 28} and that the dominant eye effect may have no specific function in vision. ²⁹ In addition, we did not monitor subjects’ fixation in this study though subjects were instructed to stare at the screen during the tests. However, as described in section 2.3, the influence of accommodation-induced near pupil response was unlikely to be significant in our results. 

In summary, our study revealed sex-specific lateralization patterns in contraction anisocoria. However, further functional and anatomic studies are needed for full understanding of the biological basis of this phenomenon. 

 Supported in part by a grant from the Wallace H. Coulter Foundation.

 Submitted for publication May 23, 2008; revised August 3 and September 16, 2008; accepted January 14, 2009.

 Disclosure: X. Fan, None; J.H. Miles, None; N. Takahashi, None; G. Yao, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

An illustration of the parasympathetic pupillary pathways. Only the afferent signals from the right eye are illustrated. Arrows: direction of signal transduction. Thick line: stronger signal (not in proportion). CG, ciliary ganglions; EW, Edinger-Westphal nucleus; OC, optic chiasm; PON, pretectal olivary nucleus.

Figure 2.

 

Age distribution of all participants. Histograms are grouped by sex.

Figure 3.

 

Schematic diagram of the experimental setup for PLR recording. LED, stimulus light sources; IR LED, illumination light sources; BS, beam splitter; NDF, neutral density filter; HM, hot mirror; D, diffuser; CCD, near-infrared charge-coupled device camera.

Table 1.

 

Average Pupil Size, and Anisocoria of Subjects Measured in Light- and Dark-adapted Conditions

Sex	Condition	Pupil Diameter (mm)			Anisocoria (mm)
				Left		Right
Female	Light-adapted	6.25 ± 0.11	6.22 ± 0.11	−0.04 ± 0.05
	Dark-adapted	6.85 ± 0.11	6.80 ± 0.11	−0.05 ± 0.05
Male	Light-adapted	6.27 ± 0.12	6.32 ± 0.12	0.05 ± 0.06
	Dark-adapted	6.74 ± 0.11	6.75 ± 0.11	0.001 ± 0.06

 Results are given as mean ± SE.

Figure 4.

 

Pupil size and anisocoria. (A) Box plot of pupil size in terms of side of eye, sex, and ambient light level. Boxes indicate the 25th and 75th percentiles, and whiskers show 1.5 times the interquartile range. The bar at the center of each box indicates the median value. (×) The maximal and the minimal data points. (B) The scatter plot of anisocoria measured under dark adaptation and light adaptation. Data points shown are mean values measured in each subject.

Figure 5.

 

Box plots of relative constriction amplitude of both sexes at three stimulus conditions. (A) Light-adapted 10^12.5 photons/cm²/s, (B) dark-adapted 10^9.3 photons/cm²/s, and (C) dark-adapted 10^12.5 photons/cm²/s.

Table 2.

 

Relative Constriction Amplitudes of Subjects Measured by Stimulating Different Eyes at Three Stimulus Conditions

Sex	Direct Eye	Light Adapted, %			Dark Adapted, %
				1012.5 photons/cm2/s			109.3 photons/cm2/s			1012.5 photons/cm2/s
				Direct	Consensual	Direct	Consensual	Direct	Consensual
Female	Left	26.00 ± 1.38	25.34 ± 1.38	20.81 ± 1.37	20.05 ± 1.37	50.98 ± 1.37	50.33 ± 1.38
	Right	26.77 ± 1.37	25.51 ± 1.37	22.32 ± 1.38	21.25 ± 1.38	51.94 ± 1.38	51.15 ± 1.37
Male	Left	24.82 ± 1.44	24.39 ± 1.44	20.22 ± 1.44	19.67 ± 1.44	55.04 ± 1.44	53.89 ± 1.44
	Right	25.14 ± 1.43	23.75 ± 1.43	22.69 ± 1.44	19.90 ± 1.44	55.93 ± 1.44	54.03 ± 1.44

 Results are given as group mean ± SE.

Table 3.

 

Contraction Anisocoria Measured in Subjects at Three Stimulus Conditions

Sex	Direct Eye	Light Adapted			Dark Adapted
				1012.5 photons/cm2/s			109.3 photons/cm2/s			1012.5 photons/cm2/s
				Mean ± SE	P	Mean ± SE	P	Mean ± SE	P
Female	Left	0.56 ± 0.32	0.0816	0.84 ± 0.30	0.0062*	0.96 ± 0.30	0.0018*
	Right	1.19 ± 0.30	0.0001*	0.90 ± 0.31	0.0045*	0.73 ± 0.30	0.0145*
Male	Left	0.42 ± 0.31	0.1836	0.26 ± 0.32	0.4105	1.21 ± 0.32	0.0002*
	Right	1.40 ± 0.31	<0.0001*	2.90 ± 0.32	<0.0001*	2.12 ± 0.32	<0.0001*

 Results are given as group mean ± SE.

*  Contraction anisocoria was significantly larger than zero.

Table 4.

 

Number and Percentage of Subjects Classified into Each of Three Categories of Contraction Anisocoria at Different Stimulus Conditions

Sex	Category	Light Adapted			Dark Adapted
				1012.5 photons/cm2/s			109.3 photons/cm2/s			1012.5 photons/cm2/s
				No.	%	No.	%	No.	%
Female	None	10	43.5	7	30.4	7	30.4
	Bilateral	3	13.0	2	8.7	2	8.7
	Unilateral (left)	3	13.0	6	26.1	7	30.4
	Unilateral (right)	7	30.4	8	34.8	7	30.4
Male	None	7	33.3	4	19.0	2	9.5
	Bilateral	2	9.5	2	9.5	4	19.0
	Unilateral (left)	3	14.3	1	4.8	5	23.8
	Unilateral (right)	9	42.9	14	66.7	10	47.6

 The unilateral category was further divided into unilateral left and unilateral right.

Figure 6.

 

Scatter plots of contraction anisocoria of both sexes as found by stimulating different eyes at three stimulus conditions: (A) light-adapted 10^12.5 photons/cm²/s, (B) dark-adapted 10^9.3 photons/cm²/s, and (C) dark-adapted 10^12.5 photons/cm²/s. Data points shown are mean values measured in each subject.

Figure 7.

 

Contraction anisocoria obtained by stimulating different eyes at three stimulus conditions in (A) female subjects and (B) male subjects. The error bars indicate the SE. LA, light adapted, DA, dark adapted.