- Email: [email protected]
The standing potential of the human eye reflects differences between upper and lower retinal areas
THE STANDING POTENTIAL OF THE HUMAN EYE REFLECTS DIFFERENCES BETWEEN UPPER AND LOWER RETINAL AREAS WOLFGANG SKRANDIESand MONIKA BAIER
Department
of Experimental
Ophthalmology, Max-Planck-Institute for Physiological Research. 6350 Bad Nauheim. F.R.G.
(Receked 7 March 1985; in retiredfortn
and Clinical
23 September 1985)
Abstract-In It healthy subjects the “light peak” of the electrooculogram was measured following localized stimulation of various retinal locations. Significant differences in “light peak” amplitudes were found between central and peripheral stimulation, and at 10 deg eccentricity the “light peak“ amplitudes were significantly larger following upper retinal stimulation than those elicited by lower retinal stimuli. in addition, the “light peak” amplitude produced by upper or lower retinal stimulation behaved differently when test light intensity increased. The upper retinal areas showed consistently a higher sensitivity to light intensity changes than the lower retinal areas. The “light peak” of the EOG is believed to index the rate of retinal metabolism elicited by light stimuli. Our findings show that upper retinal areas display a higher level of light-induced activity reflecting the interaction between the photoreceptors and the retinal pigment epitheiium than lower retinal areas. The results are interpreted as a superiority of the upper over the lower retina and are related to other electrophysjological and functional differences between upper and lower retinal areas of man. Retinal pigment epithelium
Human electrooculography
INTRODUCTION
In the vertebrate, the eye can be regarded as an electrical dipole with the cornea positive with respect to the fundus of the eye. This is the physiological basis of eye movement recordings (electrooculogram, EOG). If eye movements of a constant amplitude ate made and d.c. amplification is used, the EOG can also be used to detect changes of the intraocular standing potential (Arden and Ketsey, 1962a, b). The onset of illumination of the eye is followed by the waves of the electroretinogram which are of short latency (in the order of milliseconds for the a- and b-waves or in the order of seconds for the c-wave; Granit, 1947; Armington, 1974), and by a large positive deflection occurring after severat minutes called the “fight peak”. Both clinical studies (Arden and Kelsey, 1962b; TQumer, 1967) and animal studies (Valeton and van Norren, 1982; Linsenmeyer and Steinberg, 1982; Griff and Steinberg, 1982) showed that the retinal pigment epithelium is necessary to produce the “light peak”. Intraocular microPiease address correspondence: Wolfgang Skrandies, Ph.D., Max-Pianck-Institute Res. Lab., Department of Ophthalmology, University Hospitals, 6000 Frankfurt 70. F.R.G. 577
Upper/lower
retina
Electrophysiology
electrode recordings (Valeton and van Norren, 1982; Linsenmeyer and Steinberg, 1982; Griff and Steinberg, 1982) suggest that the source of the “light peak” is located in the basal membranes of the retinal pigment epithelium. The measurement of the standing potential of the eye, as recorded in man indirectly by the EOG, gives some indication on the functional interaction between the pigment epithelium and the retinal layer of photoreceptors. Different retinal areas show electrophysiological and functional differences. These differences are found not only between the nasal and temporal side of the retina, but also along the vertical meridian where differences between upper and lower retinal areas have been reported in human evoked potential studies (Eason et al., 1967; Lehmann and Skrandies, 1979; Skrandies et oi., 1980; Skrandies, 1984). Shorter latencies of visual evoked potential components were reported for the upper than the lower hemiretina, and there is supporting evidence that upper retinal areas are also functionally superior as reflected by regional differences in motor reaction time (Hall and von Kries, 1879; Payne, 1967), in temporal sensitivity (Skrandies, 1985a), in visual acuity (Landolt and Hummelsheim, 1904; Low, 1943;
Millodot and Lamont. 1974). and in contrast sensitivity (Skrandies. 1985b). It appeared of interest to investigate whether such differences are restricted to higher functional characteristics of the upper and lower hemiretina systems or whether already at a basic retinal level differences can be detected. In the present paper the activity of the pigment epithelium as quantified by electrooculographic measurements of upper and lower retinal regions will be compared in a subject population. In addition, the different reactivity of upper and lower retinal areas to physical stimulus changes will be reported. METHODS
Twelve healthy subjects between 19 and 32 years of age participated in the experiments which extended over five different sessions lasting about 2 hours each. In order to minimize diurnal variations, all experiments were performed at the same time of day (late in the afternoon). The pupils were maximally dilated by application of Mydriaticum Roche and Neosynephrine. Hellige Ag-AgCI electrodes were fastened at the lateral and medial canthii of each eye. During an adaptation phase of at least 60min, the subject was exposed to a white adaptation light of - I .77 log cd/m’, projected as a Ganzfeld onto a half-cylindric screen. After about I hr the baseline was checked for stability. The subject performed at 1 min intervals horizontal eye movements with an excursion of 40 deg, and the amplitude of the standing potential was indirectly measured with the EOG. About 8 min later the standing potential had reached a steady state, and a circular white test light (IO deg diameter, 2.73 log cd/m’) was backprojected at the center of the Ganzfeld. The adaptation light remained on throughout the session. Central, and peripheral upper and lower retinal areas at IO and 40 deg eccentricity along the vertical meridian were stimulated in five different experimental sessions by changing the location of the fixation marks. EOG was The recorded using a “Tonnies-EOG-System” with a d.c. amplifier (low pass filter at 20 Hz, amplification 1.5. IO’) and an analog computer which integrated the amplitudes of 10 consecutive eye movements. Individual EOGs as well as the integrated values were plotted online on a y--f-plotter. The mean amplitude of the last eight recordings of the adaptation phase were used as a baseline.
In order to discriminate the ..light peak” from spontaneous fluctuations of the dark value. the 95% confidence limit of the values obtained in the adaptation phase was computed for each eye separately using the r-distribution. Peak values exceeding this limit were considered reliable. and mean values of the right and left eye were used for further analysis. Only two out of the 120 measures obtained (I 2 subjects. 2 eyes. 5 retinal locations) had to be excluded. The “light peak” amplitudes were expressed as percentages of the adaptation values. Since for percentage values a normal distribution cannot be assumed, and in order to eliminate intersubject variations. Wilcoxon tests were used for comparing experimental conditions. Three subjects participated in a second series of experiments consisting of 15 sessions each in which the reactivity of upper or lower retinal areas to different intensities of the test light was investigated. The same procedures as described above were used, but measurements were restricted to central, and upper and lower retinal areas at IOdeg eccentricity. The test light intensity varied between 2.43 and 3.63 logcd/m’ in steps of 0.5 log cd/m? yielding five light rise values for each location. The relationship between EOG amplitude and test light intensity was established by determining linear regressions. RESULTS
Figure 1 shows the original EOGs of one subject measured at five different retinal locations. In all recordings the “light peak” occurs after about 10 min. The peak latencies were similar for all retinal locations averaging 9.5 min in the subject population. There were also no differences observed when comparing central and peripheral “light peak” latencies. Due to the small test light, the “light peak” amplitudes were in the order of only l&20%. The highest amplitudes of the “light peak” were observed centrally (median over all subjects: 16.12%) which were larger than those obtained peripherally (median difference: 2.47%). This difference was significant in the subject population (P < 0.05). When comparing corresponding peripheral areas, it turned out that upper retinal areas had higher amplitudes of the “light peak” than lower retinal areas. However, these differences were significant only at 10 deg eccentricity where upper retinal areas showed a median light rise of 15.26% while for lower retinal areas the median “light peak” amplitude was
Standing
potential
RETINAL LOCATION
i kj z a I
110 100
90 1
!z
io
20
MIN
LIGHT
Fig. I. Original local EOG measurements of one subject obtained at various retinal locations along the vertical meridian. Retinal locations refer to the center of the IO deg arc test light and range from 40deg on the upper retina (4O‘U). via IO deg on the upper (IOU). the center (0°C). and IOdeg on the lower retina (IO’L) to 40deg on the lower retina (40-L). A rise in potential peaking after approximately IOmin was measured with reference to the base value. Note that there are no differences in latency but in amplitudes of the EOG (see text).
12.44% (median difference: 2.82%, P c 0.025). At 40 deg eccentricity the same tendency was observed (12.06% peak amplitude for the upper, 11.76% amplitude for the lower retinal areas), but the amplitudes were very variable was not statistically and the difference significant (median difference: 0.3%, ns). Figure 2 summarizes the data and shows the
of human
579
eye
median “light peak” amplitude distribution of the subject population at the five retinal locations examined. Central stimuli yield the highest amplitudes, and the upper retinal areas have higher amplitudes than the corresponding lower retinal areas. We note that due to the interindividual variation of these physiological measures, group means are only a rough indication of the experimental effects, and only the consistent differences revealed by paired statistical tests help to reach meaningful conclusions. With increasing intensity of the test light the “light peak” amplitudes increased. A linear relationship was found in all subjects at all retinal locations investigated. The correlation coefficients between the EOG amplitude and the logarithm of the test light intensity averaged 0.75. The resulting slopes of the regression lines computed individually for each subject and retinal location are illustrated in Fig. 3. The relationship between test light intensity and the “light peak” amplitudes is reflected by the slope of the regression lines, and in order to enable a direct comparison between the retinal locations amplitude independent of the absolute differences the regression lines were normalized: Fig. 3 shows the “light peak” amplitudes relative to the amplitudes obtained with a light RETINAL LOCATION 0%
3or
Sl
/
,
10% 1O”L
20 .
E
10.
iii 2
o-
d
4 20-
H
10.
w 1 3
o-
it! 20
1O”U
/
/
k s2
0°C 1OOL
1O~U
s3
0°C 1OlL 10 0I
/ I
L
2.5
30
3.5
TEST LIGHT INTENSITY [LOG CC&~] L
40” UPPER
lo”
RETINAL
0”
10’
LOWER 40”
LOCATION
Fig. 2. Distribution of local EOG amplitudes in a population of 12 subjects. Median values are presented; the differences between upper and lower areas at IO deg eccentricity were statistically significant.
Fig. 3. Slopes of regressions lines indicating the relationship between test light intensity and relative EOG amplitude obtained from three different subjects with stimuli presented in the center(O’C) and IO deg above (IO’U) or below (IO’L) the fovea. Measurements were made at five intensities between 2.43 and 3.63 log cd/m’. The values at 2.5 log cd/m! are set to 0% to enable a comparison independent of absolute “light peak” amplitudes.
intensity of 2.5 log cd m’ at the respectiv,e retinal location. Different slopes are found consistently for upper and lower retinal stimuli in all subjects vvith upper stimuli producing a steeper slope than lower retinal stimuli. This suggests a greater sensitivity to light increase of the upper than the lovv-er retinal areas. The behaviour of ‘-light peak” amplitudes cbtained with central stimuli displayed more intersubject variation: for subject I (Sl) central stimuli resulted in the highest increase while for subjects 2 (52) and 3 (S3) the amplitude increases elicited by central stimuli lay in between those of upper or lower retinal stimuli. In addition, in subject 3 (S3) the difference between central and lower retinal areas was very small (see Fig. 3). DiSCUSSION The “light peak” amplitudes in the order of less than 200/a of the baseline value are in agreement with the results of other studies in which small localized test stimuli were used (Aschoff, I98 1). Fovea1 stimuli yielded slightly higher EOG than perifoveal stimuli. The amplitudes difference averaged about 2.5% and achieved statistical significance. This finding is in line with reports on the contribution of cone activity to the “light peak” amplitude (Afanador and Andrews, 1978), and with data obtained from congenital achromats (Elenius and Aantaa, 1973). Similar differences were observed when using central and peripheral test lights (Aschoff, 1981: Dodt and Baier, 1984). The small differences between the upper and lower perifoveal retinal areas were consistent and statistically significant in the subject population. Thus, differences of “light peak” amplitudes in the various test sessions caused by intrasubject variation can be ruled out in our data. In addition, variation of electrode placement causing different EOG amplitudes cannot account for the effects of retinal location because the “light peak” amplitudes were always referred to the respective EOG values obtained in the baseline condition at the end of the dark adaptation periods. We also note that a very conservative statistical method was employed in order to take care of the low signal to noise ration: only “light peak” amplitudes exceeding the t-transformed 95% confidence limit were used for further analysis. Upper retinal areas had consistently higher “light peak” amplitudes than corresponding lower retinal areas. These differences were sta-
tistically significant only at IO deg eccentricity. At 10 dep eccentricity similar upper, lower differences vvere found but the variability of the “light peak” amplitudes was high. and the did not differences achieve statistical significance. The higher variability of the “light peak” amplitudes at 30 deg eccentricity may be explained by the fact that a smaller number of photoreceptors is activated by a test light located betbceen 35 and 45 deg eccentricity than by stimuli located between 5 and I5 deg eccentricity (Osterberg. 1935). This might well result in more consistent differences at perifoveal than at peripheral retinal areas. In addition to the absolute -‘light peak” amplitude differences between upper and lower retinal areas which were observed in a population of I2 subjects we also found relative differences between upper and lower retinal areas when the intensity of the test light was varied. As expected from earlier results (Arden and Kelsey, 1962b) the amplitude of the “light peak” was proportional to the logarithm of the test light intensity. Upper and lower retinal “light peaks” behaved differently when the test light intensity was changed. The relative “light peak” amplitude increase following upper retinal stimuli was much larger than that following lower retinal stimuli. This effect was consistent in all subjects studied (see Fig. 3), and it suggests that the upper retinal areas are more sensitive to changes of the test light intensity. The response of the EOG to light requires contact between the photoreceptors and the pigment epithelium, and the “light peak” amplitude is believed to reflect the rate of metabolism elicited by light stimufi (Arden and Kelsey, 1962b; Marmor and Lurie, 1979). In animal studies it was found that the source of the ‘*light peak” presumably is located in the basal membranes of the pigment epithelium (Valeton and van Norren, 1982; Linsenmeyer and Steinberg, 1982; Griff and Steinberg, 1982), and our findings suggest that light stimuli presented to upper retinal areas are followed by a higher level of activity in the interaction between the photoreceptors and the retinal pigment epithelium than stimuli presented to lower retinal areas. This interpretation is supported by anatomical differences in the regional distribution of rods and cones in the human retina where a higher density of photoreceptors was reported for upper retinal areas (Msterberg, 1935). The amplitude of the “light peak” may reflect the rate of metabolism elicited by light stimuli, and thus
581
Standing potential of human eye
correlates with the anatomically determined photoreceptor density. The present findings are in general agreement with the reported functional differences in regional visual acuity (Landolt and Hummelsheim, 1904: Low, 1943; Millodot and Lamont, 1974). Data on contrast sensitivity measured over the whole range from very low to very high spatial frequencies suggest a global functional superiority of the upper over the lower perifoveal retinal areas which is not restricted to individual soatial freauencv channels (Skrandies, 198Sb). * a _ In electrophysiologica1 studies on component latencies of the visual evoked potential stimulation of the lower hemiretina showed consistently longer latencies than stimulation of upper retinal areas (Lehmann and Skrandies, 1979; Skrandies et al., 1980; Adachi-Usami and Lehmann, 1983; Skrandies, 1984). This as well as data on temporal resolution as measured by critical flicker fusion and double flash discrimination (Skrandies, t985a), and on motor reaction time (Hall and von Kries, 1879; Payne, 1967) indicates a functional superiority of the upper retina which is responsive to visual objects below the horizontal meridian of the visual field. Our present data show that differences between upper and lower retinal areas can be demonstrated aiready at a very basic, retinal level at the location of the first stage of visual information processing. Acknowledgements-We thank Professor E. Dodt for his continuing support, and Professor D. Lehmann and Dr L. Peichl for invaluable comments on an earlier version of the manuscript.
REFERENCES Adachi-Usami E. and Lehmann D. (1983) Monocular and binocular evoked average potential field topography: upper and lower hemiretinal stimuli. Expi Bruin Res. 50, 34 l-346. Afanador A. 1. and Andrews C. E. (1978) Rod and cone contribution to the EOG ratio. Am. J. Optom. 55, 101-107. Arden G. B. and Kelsey J. H. (1962a) Changes produced by light in the standing potential of the human eye. J. Physiol. 161, 189-204. Arden G. B. and Kelsey 1. H. (1962b) Some observations on the relationship between the standing potential of the human eye and bleaching and regeneration of visual purple. J. Physioi. 161, 205-226. Armington J. C. (1974) The Electrorefinogram. Academic Press, New York.
Aschoffu. (1981) Skotopische und photopische Anteile der Hell- und Dunkeischwingung im Elcktrookulo~ramm. Dol. Ophrhal. 4, 149-166. Dodt E. and Bairr M. (1984) Area-luminance relationship for a constant light peak of the standing potential in the human eye. Ophrhalmologica 188. 232-238. Eason R. G.. White C. T. and Oden D. (1967) Averaged occipital responses to stimulation of sites in the upper and lower halves of the retina. Percepr. &choph_vs. 2, 4234’5. Elenius V. and Aantaa E. (I 973) Light-induced increase in amplitude of electro-oculogram. Archs OphthaL 90, 60-63. Granit R. (1947) Sensory Mechanisms of fhe Rerina. Hafner, New York. Griff E. R. and Steinberg R. H. (1982) Origin of the light peak: in rirro study of Gekko gekko. J. Ph,rsio/. 331, 631-652.
Hall G. S. and von Kries J. (1879) tiber die Abhlngigkeit der Reaktionszeit vom Ort des Reizes. Archs Anat. fhysioi. Suppl., Leipzig, l-10. Landolt E. and Hummelsheim E. (1904) Die Untersuchung der Funktionen des excentrischen Netzhautgebietes. In Graefe-Saemisch
Handbuch
der gesamten
Augenheilkunde.
(Ediied by Saemisch T.), Vol. 4,pp. 503-583: Engelmann, Leipzig. Lehmann D. and Skrandies W. (1979) multichannel evoked potential fields show different properties of human upper and lower hemi-retinal systems. E,vpl Brain Res. 35, 151-159. Linsenmeyer R. A. and Steinberg R. H. (1982) Origin and sensitivity of the light peak in the intact cat eye. J. Physiol. 331, 6.53-673. Low F. N. (1943) The peripheral visual acuity of IO0 subjects. Am. J. Physiol. 140, 83-88. Marmor M. F. and Lurie M. (1979) Light-induced electrical responses of the retinal pigment epithelium. In The Rerinal Pigment Ep~fhelium (Edited by Zinn K. M. and Marmor M. F.) Harvard Univ. Press, Cambridge, Mass., pp. 226-244. Millodot M. and Lamont A. (1974) Peripheral visual acuity in the vertical plane. Vision Res. 14, 1497-1498. Osterberg G. (1935) Topography of the layer of rods and cones in the human retina. Acta ophthat., suppl. 6, I-102. Payne W. H. (1965) Visual reaction times on a circle about the fovea. Science 155, 481-482. Skrandies W. (1984) Scalp potential fields evoked by grating stimuli: effects of spatial frequency and orientation. E~e~Iroen~eph.
clin. iVeurophysjo~.
58, 325-332.
Skrandies W. (l98Sa) Critical flicker fusion and double flash discrimination in different parts of the visual field. Inl. J. Neurosci. 2.5, 225-23 I. Skrandies W. (1985b) Human contrast sensitivity: regional retinal differences. Human ~eurobioi. 4, 97-99. Skrandies W., Richter M. and Lehmann D. (1980) Checkerboard evoked potentials: Topography and latency for onset, offset. and reversal. Prog. Brain Res. 54, 291-295. TIumer R. (Editor) (1976) Electra-Oculography-Its Clinical Importance. Biblrhca Ophthal., Vol. 85, Karger, Basel. Valeton J. M. and van Norren D. (1982) Intraretinal recording of slow electrical responses to steady illumination in monkey: isolation of receptor responses and the origin of the light peak. Vision Res. 22, 393-399.
Department
of Experimental
Ophthalmology, Max-Planck-Institute for Physiological Research. 6350 Bad Nauheim. F.R.G.
(Receked 7 March 1985; in retiredfortn
and Clinical
23 September 1985)
Abstract-In It healthy subjects the “light peak” of the electrooculogram was measured following localized stimulation of various retinal locations. Significant differences in “light peak” amplitudes were found between central and peripheral stimulation, and at 10 deg eccentricity the “light peak“ amplitudes were significantly larger following upper retinal stimulation than those elicited by lower retinal stimuli. in addition, the “light peak” amplitude produced by upper or lower retinal stimulation behaved differently when test light intensity increased. The upper retinal areas showed consistently a higher sensitivity to light intensity changes than the lower retinal areas. The “light peak” of the EOG is believed to index the rate of retinal metabolism elicited by light stimuli. Our findings show that upper retinal areas display a higher level of light-induced activity reflecting the interaction between the photoreceptors and the retinal pigment epitheiium than lower retinal areas. The results are interpreted as a superiority of the upper over the lower retina and are related to other electrophysjological and functional differences between upper and lower retinal areas of man. Retinal pigment epithelium
Human electrooculography
INTRODUCTION
In the vertebrate, the eye can be regarded as an electrical dipole with the cornea positive with respect to the fundus of the eye. This is the physiological basis of eye movement recordings (electrooculogram, EOG). If eye movements of a constant amplitude ate made and d.c. amplification is used, the EOG can also be used to detect changes of the intraocular standing potential (Arden and Ketsey, 1962a, b). The onset of illumination of the eye is followed by the waves of the electroretinogram which are of short latency (in the order of milliseconds for the a- and b-waves or in the order of seconds for the c-wave; Granit, 1947; Armington, 1974), and by a large positive deflection occurring after severat minutes called the “fight peak”. Both clinical studies (Arden and Kelsey, 1962b; TQumer, 1967) and animal studies (Valeton and van Norren, 1982; Linsenmeyer and Steinberg, 1982; Griff and Steinberg, 1982) showed that the retinal pigment epithelium is necessary to produce the “light peak”. Intraocular microPiease address correspondence: Wolfgang Skrandies, Ph.D., Max-Pianck-Institute Res. Lab., Department of Ophthalmology, University Hospitals, 6000 Frankfurt 70. F.R.G. 577
Upper/lower
retina
Electrophysiology
electrode recordings (Valeton and van Norren, 1982; Linsenmeyer and Steinberg, 1982; Griff and Steinberg, 1982) suggest that the source of the “light peak” is located in the basal membranes of the retinal pigment epithelium. The measurement of the standing potential of the eye, as recorded in man indirectly by the EOG, gives some indication on the functional interaction between the pigment epithelium and the retinal layer of photoreceptors. Different retinal areas show electrophysiological and functional differences. These differences are found not only between the nasal and temporal side of the retina, but also along the vertical meridian where differences between upper and lower retinal areas have been reported in human evoked potential studies (Eason et al., 1967; Lehmann and Skrandies, 1979; Skrandies et oi., 1980; Skrandies, 1984). Shorter latencies of visual evoked potential components were reported for the upper than the lower hemiretina, and there is supporting evidence that upper retinal areas are also functionally superior as reflected by regional differences in motor reaction time (Hall and von Kries, 1879; Payne, 1967), in temporal sensitivity (Skrandies, 1985a), in visual acuity (Landolt and Hummelsheim, 1904; Low, 1943;
Millodot and Lamont. 1974). and in contrast sensitivity (Skrandies. 1985b). It appeared of interest to investigate whether such differences are restricted to higher functional characteristics of the upper and lower hemiretina systems or whether already at a basic retinal level differences can be detected. In the present paper the activity of the pigment epithelium as quantified by electrooculographic measurements of upper and lower retinal regions will be compared in a subject population. In addition, the different reactivity of upper and lower retinal areas to physical stimulus changes will be reported. METHODS
Twelve healthy subjects between 19 and 32 years of age participated in the experiments which extended over five different sessions lasting about 2 hours each. In order to minimize diurnal variations, all experiments were performed at the same time of day (late in the afternoon). The pupils were maximally dilated by application of Mydriaticum Roche and Neosynephrine. Hellige Ag-AgCI electrodes were fastened at the lateral and medial canthii of each eye. During an adaptation phase of at least 60min, the subject was exposed to a white adaptation light of - I .77 log cd/m’, projected as a Ganzfeld onto a half-cylindric screen. After about I hr the baseline was checked for stability. The subject performed at 1 min intervals horizontal eye movements with an excursion of 40 deg, and the amplitude of the standing potential was indirectly measured with the EOG. About 8 min later the standing potential had reached a steady state, and a circular white test light (IO deg diameter, 2.73 log cd/m’) was backprojected at the center of the Ganzfeld. The adaptation light remained on throughout the session. Central, and peripheral upper and lower retinal areas at IO and 40 deg eccentricity along the vertical meridian were stimulated in five different experimental sessions by changing the location of the fixation marks. EOG was The recorded using a “Tonnies-EOG-System” with a d.c. amplifier (low pass filter at 20 Hz, amplification 1.5. IO’) and an analog computer which integrated the amplitudes of 10 consecutive eye movements. Individual EOGs as well as the integrated values were plotted online on a y--f-plotter. The mean amplitude of the last eight recordings of the adaptation phase were used as a baseline.
In order to discriminate the ..light peak” from spontaneous fluctuations of the dark value. the 95% confidence limit of the values obtained in the adaptation phase was computed for each eye separately using the r-distribution. Peak values exceeding this limit were considered reliable. and mean values of the right and left eye were used for further analysis. Only two out of the 120 measures obtained (I 2 subjects. 2 eyes. 5 retinal locations) had to be excluded. The “light peak” amplitudes were expressed as percentages of the adaptation values. Since for percentage values a normal distribution cannot be assumed, and in order to eliminate intersubject variations. Wilcoxon tests were used for comparing experimental conditions. Three subjects participated in a second series of experiments consisting of 15 sessions each in which the reactivity of upper or lower retinal areas to different intensities of the test light was investigated. The same procedures as described above were used, but measurements were restricted to central, and upper and lower retinal areas at IOdeg eccentricity. The test light intensity varied between 2.43 and 3.63 logcd/m’ in steps of 0.5 log cd/m? yielding five light rise values for each location. The relationship between EOG amplitude and test light intensity was established by determining linear regressions. RESULTS
Figure 1 shows the original EOGs of one subject measured at five different retinal locations. In all recordings the “light peak” occurs after about 10 min. The peak latencies were similar for all retinal locations averaging 9.5 min in the subject population. There were also no differences observed when comparing central and peripheral “light peak” latencies. Due to the small test light, the “light peak” amplitudes were in the order of only l&20%. The highest amplitudes of the “light peak” were observed centrally (median over all subjects: 16.12%) which were larger than those obtained peripherally (median difference: 2.47%). This difference was significant in the subject population (P < 0.05). When comparing corresponding peripheral areas, it turned out that upper retinal areas had higher amplitudes of the “light peak” than lower retinal areas. However, these differences were significant only at 10 deg eccentricity where upper retinal areas showed a median light rise of 15.26% while for lower retinal areas the median “light peak” amplitude was
Standing
potential
RETINAL LOCATION
i kj z a I
110 100
90 1
!z
io
20
MIN
LIGHT
Fig. I. Original local EOG measurements of one subject obtained at various retinal locations along the vertical meridian. Retinal locations refer to the center of the IO deg arc test light and range from 40deg on the upper retina (4O‘U). via IO deg on the upper (IOU). the center (0°C). and IOdeg on the lower retina (IO’L) to 40deg on the lower retina (40-L). A rise in potential peaking after approximately IOmin was measured with reference to the base value. Note that there are no differences in latency but in amplitudes of the EOG (see text).
12.44% (median difference: 2.82%, P c 0.025). At 40 deg eccentricity the same tendency was observed (12.06% peak amplitude for the upper, 11.76% amplitude for the lower retinal areas), but the amplitudes were very variable was not statistically and the difference significant (median difference: 0.3%, ns). Figure 2 summarizes the data and shows the
of human
579
eye
median “light peak” amplitude distribution of the subject population at the five retinal locations examined. Central stimuli yield the highest amplitudes, and the upper retinal areas have higher amplitudes than the corresponding lower retinal areas. We note that due to the interindividual variation of these physiological measures, group means are only a rough indication of the experimental effects, and only the consistent differences revealed by paired statistical tests help to reach meaningful conclusions. With increasing intensity of the test light the “light peak” amplitudes increased. A linear relationship was found in all subjects at all retinal locations investigated. The correlation coefficients between the EOG amplitude and the logarithm of the test light intensity averaged 0.75. The resulting slopes of the regression lines computed individually for each subject and retinal location are illustrated in Fig. 3. The relationship between test light intensity and the “light peak” amplitudes is reflected by the slope of the regression lines, and in order to enable a direct comparison between the retinal locations amplitude independent of the absolute differences the regression lines were normalized: Fig. 3 shows the “light peak” amplitudes relative to the amplitudes obtained with a light RETINAL LOCATION 0%
3or
Sl
/
,
10% 1O”L
20 .
E
10.
iii 2
o-
d
4 20-
H
10.
w 1 3
o-
it! 20
1O”U
/
/
k s2
0°C 1OOL
1O~U
s3
0°C 1OlL 10 0I
/ I
L
2.5
30
3.5
TEST LIGHT INTENSITY [LOG CC&~] L
40” UPPER
lo”
RETINAL
0”
10’
LOWER 40”
LOCATION
Fig. 2. Distribution of local EOG amplitudes in a population of 12 subjects. Median values are presented; the differences between upper and lower areas at IO deg eccentricity were statistically significant.
Fig. 3. Slopes of regressions lines indicating the relationship between test light intensity and relative EOG amplitude obtained from three different subjects with stimuli presented in the center(O’C) and IO deg above (IO’U) or below (IO’L) the fovea. Measurements were made at five intensities between 2.43 and 3.63 log cd/m’. The values at 2.5 log cd/m! are set to 0% to enable a comparison independent of absolute “light peak” amplitudes.
intensity of 2.5 log cd m’ at the respectiv,e retinal location. Different slopes are found consistently for upper and lower retinal stimuli in all subjects vvith upper stimuli producing a steeper slope than lower retinal stimuli. This suggests a greater sensitivity to light increase of the upper than the lovv-er retinal areas. The behaviour of ‘-light peak” amplitudes cbtained with central stimuli displayed more intersubject variation: for subject I (Sl) central stimuli resulted in the highest increase while for subjects 2 (52) and 3 (S3) the amplitude increases elicited by central stimuli lay in between those of upper or lower retinal stimuli. In addition, in subject 3 (S3) the difference between central and lower retinal areas was very small (see Fig. 3). DiSCUSSION The “light peak” amplitudes in the order of less than 200/a of the baseline value are in agreement with the results of other studies in which small localized test stimuli were used (Aschoff, I98 1). Fovea1 stimuli yielded slightly higher EOG than perifoveal stimuli. The amplitudes difference averaged about 2.5% and achieved statistical significance. This finding is in line with reports on the contribution of cone activity to the “light peak” amplitude (Afanador and Andrews, 1978), and with data obtained from congenital achromats (Elenius and Aantaa, 1973). Similar differences were observed when using central and peripheral test lights (Aschoff, 1981: Dodt and Baier, 1984). The small differences between the upper and lower perifoveal retinal areas were consistent and statistically significant in the subject population. Thus, differences of “light peak” amplitudes in the various test sessions caused by intrasubject variation can be ruled out in our data. In addition, variation of electrode placement causing different EOG amplitudes cannot account for the effects of retinal location because the “light peak” amplitudes were always referred to the respective EOG values obtained in the baseline condition at the end of the dark adaptation periods. We also note that a very conservative statistical method was employed in order to take care of the low signal to noise ration: only “light peak” amplitudes exceeding the t-transformed 95% confidence limit were used for further analysis. Upper retinal areas had consistently higher “light peak” amplitudes than corresponding lower retinal areas. These differences were sta-
tistically significant only at IO deg eccentricity. At 10 dep eccentricity similar upper, lower differences vvere found but the variability of the “light peak” amplitudes was high. and the did not differences achieve statistical significance. The higher variability of the “light peak” amplitudes at 30 deg eccentricity may be explained by the fact that a smaller number of photoreceptors is activated by a test light located betbceen 35 and 45 deg eccentricity than by stimuli located between 5 and I5 deg eccentricity (Osterberg. 1935). This might well result in more consistent differences at perifoveal than at peripheral retinal areas. In addition to the absolute -‘light peak” amplitude differences between upper and lower retinal areas which were observed in a population of I2 subjects we also found relative differences between upper and lower retinal areas when the intensity of the test light was varied. As expected from earlier results (Arden and Kelsey, 1962b) the amplitude of the “light peak” was proportional to the logarithm of the test light intensity. Upper and lower retinal “light peaks” behaved differently when the test light intensity was changed. The relative “light peak” amplitude increase following upper retinal stimuli was much larger than that following lower retinal stimuli. This effect was consistent in all subjects studied (see Fig. 3), and it suggests that the upper retinal areas are more sensitive to changes of the test light intensity. The response of the EOG to light requires contact between the photoreceptors and the pigment epithelium, and the “light peak” amplitude is believed to reflect the rate of metabolism elicited by light stimufi (Arden and Kelsey, 1962b; Marmor and Lurie, 1979). In animal studies it was found that the source of the ‘*light peak” presumably is located in the basal membranes of the pigment epithelium (Valeton and van Norren, 1982; Linsenmeyer and Steinberg, 1982; Griff and Steinberg, 1982), and our findings suggest that light stimuli presented to upper retinal areas are followed by a higher level of activity in the interaction between the photoreceptors and the retinal pigment epithelium than stimuli presented to lower retinal areas. This interpretation is supported by anatomical differences in the regional distribution of rods and cones in the human retina where a higher density of photoreceptors was reported for upper retinal areas (Msterberg, 1935). The amplitude of the “light peak” may reflect the rate of metabolism elicited by light stimuli, and thus
581
Standing potential of human eye
correlates with the anatomically determined photoreceptor density. The present findings are in general agreement with the reported functional differences in regional visual acuity (Landolt and Hummelsheim, 1904: Low, 1943; Millodot and Lamont, 1974). Data on contrast sensitivity measured over the whole range from very low to very high spatial frequencies suggest a global functional superiority of the upper over the lower perifoveal retinal areas which is not restricted to individual soatial freauencv channels (Skrandies, 198Sb). * a _ In electrophysiologica1 studies on component latencies of the visual evoked potential stimulation of the lower hemiretina showed consistently longer latencies than stimulation of upper retinal areas (Lehmann and Skrandies, 1979; Skrandies et al., 1980; Adachi-Usami and Lehmann, 1983; Skrandies, 1984). This as well as data on temporal resolution as measured by critical flicker fusion and double flash discrimination (Skrandies, t985a), and on motor reaction time (Hall and von Kries, 1879; Payne, 1967) indicates a functional superiority of the upper retina which is responsive to visual objects below the horizontal meridian of the visual field. Our present data show that differences between upper and lower retinal areas can be demonstrated aiready at a very basic, retinal level at the location of the first stage of visual information processing. Acknowledgements-We thank Professor E. Dodt for his continuing support, and Professor D. Lehmann and Dr L. Peichl for invaluable comments on an earlier version of the manuscript.
REFERENCES Adachi-Usami E. and Lehmann D. (1983) Monocular and binocular evoked average potential field topography: upper and lower hemiretinal stimuli. Expi Bruin Res. 50, 34 l-346. Afanador A. 1. and Andrews C. E. (1978) Rod and cone contribution to the EOG ratio. Am. J. Optom. 55, 101-107. Arden G. B. and Kelsey J. H. (1962a) Changes produced by light in the standing potential of the human eye. J. Physiol. 161, 189-204. Arden G. B. and Kelsey 1. H. (1962b) Some observations on the relationship between the standing potential of the human eye and bleaching and regeneration of visual purple. J. Physioi. 161, 205-226. Armington J. C. (1974) The Electrorefinogram. Academic Press, New York.
Aschoffu. (1981) Skotopische und photopische Anteile der Hell- und Dunkeischwingung im Elcktrookulo~ramm. Dol. Ophrhal. 4, 149-166. Dodt E. and Bairr M. (1984) Area-luminance relationship for a constant light peak of the standing potential in the human eye. Ophrhalmologica 188. 232-238. Eason R. G.. White C. T. and Oden D. (1967) Averaged occipital responses to stimulation of sites in the upper and lower halves of the retina. Percepr. &choph_vs. 2, 4234’5. Elenius V. and Aantaa E. (I 973) Light-induced increase in amplitude of electro-oculogram. Archs OphthaL 90, 60-63. Granit R. (1947) Sensory Mechanisms of fhe Rerina. Hafner, New York. Griff E. R. and Steinberg R. H. (1982) Origin of the light peak: in rirro study of Gekko gekko. J. Ph,rsio/. 331, 631-652.
Hall G. S. and von Kries J. (1879) tiber die Abhlngigkeit der Reaktionszeit vom Ort des Reizes. Archs Anat. fhysioi. Suppl., Leipzig, l-10. Landolt E. and Hummelsheim E. (1904) Die Untersuchung der Funktionen des excentrischen Netzhautgebietes. In Graefe-Saemisch
Handbuch
der gesamten
Augenheilkunde.
(Ediied by Saemisch T.), Vol. 4,pp. 503-583: Engelmann, Leipzig. Lehmann D. and Skrandies W. (1979) multichannel evoked potential fields show different properties of human upper and lower hemi-retinal systems. E,vpl Brain Res. 35, 151-159. Linsenmeyer R. A. and Steinberg R. H. (1982) Origin and sensitivity of the light peak in the intact cat eye. J. Physiol. 331, 6.53-673. Low F. N. (1943) The peripheral visual acuity of IO0 subjects. Am. J. Physiol. 140, 83-88. Marmor M. F. and Lurie M. (1979) Light-induced electrical responses of the retinal pigment epithelium. In The Rerinal Pigment Ep~fhelium (Edited by Zinn K. M. and Marmor M. F.) Harvard Univ. Press, Cambridge, Mass., pp. 226-244. Millodot M. and Lamont A. (1974) Peripheral visual acuity in the vertical plane. Vision Res. 14, 1497-1498. Osterberg G. (1935) Topography of the layer of rods and cones in the human retina. Acta ophthat., suppl. 6, I-102. Payne W. H. (1965) Visual reaction times on a circle about the fovea. Science 155, 481-482. Skrandies W. (1984) Scalp potential fields evoked by grating stimuli: effects of spatial frequency and orientation. E~e~Iroen~eph.
clin. iVeurophysjo~.
58, 325-332.
Skrandies W. (l98Sa) Critical flicker fusion and double flash discrimination in different parts of the visual field. Inl. J. Neurosci. 2.5, 225-23 I. Skrandies W. (1985b) Human contrast sensitivity: regional retinal differences. Human ~eurobioi. 4, 97-99. Skrandies W., Richter M. and Lehmann D. (1980) Checkerboard evoked potentials: Topography and latency for onset, offset. and reversal. Prog. Brain Res. 54, 291-295. TIumer R. (Editor) (1976) Electra-Oculography-Its Clinical Importance. Biblrhca Ophthal., Vol. 85, Karger, Basel. Valeton J. M. and van Norren D. (1982) Intraretinal recording of slow electrical responses to steady illumination in monkey: isolation of receptor responses and the origin of the light peak. Vision Res. 22, 393-399.