Documenta Ophthalmologica 80: 171-181, t992. 1992 Kluwer Academic Publishers. Printed in the Netherlands.
9
Visual evoked cortical potentials from transient dark and bright stimuli Selective 'on' and 'off-pathway' testing? E R K A N MUTLUKAN, I M I C H A E L B R A D N A M , 2 DAVID KEATING 2 & B E R T I L E. D A M A T O 1 tTennent Institute of Ophthalmology, University of Glasgow, 38 Church Street, Glasgow Gll 6NT, Scotland; ZWest of Scotland Health Boards Department of Clinical Physics and Bio-Engineering, 11 West Graham Street, Glasgow G4 9LF, Scotland Accepted 12 April 1992
Key words: on-pathway, off-pathway, parallel pathways, retinal ganglion cell, visual cortex, contrast, electrophysiology, VECP, cortical potentials, visual evoked response Abstract. The superior hemifields of five normal left eyes were stimulated by novel equal and opposite contrast pattern onset stimuli which were generated on a cathode ray tube. Patterns consisted of 144 discs, each subtending 60 rain arc at the viewing distance of 40 cm and were separated by a distance equal to their diameter. Equal and opposite light changes were created by presenting the disc patterns with different luminance values on a uniform constant background (20cd/m2). Transient visual evoked cortical potentials to the appearance and disappearance of the patterns were recorded separately and analysed. Significant amplitude differences between the responses to bright and dark stimuli were observed with light increment responses being 36-53% larger than the light decrement responses for pattern on-set and 54-80% larger for pattern off-set respectively. This fnding is attributed to the difference in the input to the ON and OFF Parallel Pathways which are known to carry light increment and decrement information respectively, as well as differences in the metabolic and discharge rates of these pathways.
Introduction The most commonly employed stimuli for measuring pattern visual evoked cortical potentials (VECPs) are checkerboard and grating patterns, which consist of alternating dark and light elements. With such patterns, light increments and decrements occur simultaneously at each reversal, so that the average luminance is steady. The combined visual response to both the bright and dark components of the stimulus pattern is therefore recorded simultaneously. It is believed that light increments (bright stimulus on dark background) and decrements (dark stimulus on bright background) are processed by two separate channels in the visual system, namely 'On' and 'Off'-Pathways, which exist between the retinal bipolar cell layer and visual cortex, and which display morphological, physiological, pharmacological and psycho-
172
Fig. 1 (a-f). Black, grey and white pattern stimuli. The positions of the discs in one pattern
173
corresponded to the interdisc areas of the other pattern to decrease the on/off time ratio by alternate presentations.
174 physical differences [1-8]. In view of these differences, it is possible that 'On' and 'Off' Pathways also differ in their susceptibility to damage in glaucoma and other neuro-ophthalmic diseases. It may therefore be clinically useful to test 'On' and 'Off' pathways separately when performing electrophysiological examinations. We have measured visual evoked cortical responses to bright and dark stimuli presented sequentially against a constant background. The superior hemifield was examined because this part of the visual field tends to be lost relatively early in conditions such as glaucoma and chiasmal compression. In this article, we describe the bright and dark stimuli we have developed and report the results of a preliminary study which shows differences between responses obtained with such stimuli.
Materials and methods
Five normal subjects were examined, three male and two female, ranging in age from 26 to 35 year (mean 30). All subjects had a corrected visual acuity of 6/6 or better, normal pupils (3-6 mm) and normal visual fields. The stimulus and recording system was based on a personal computer (IBM Model 30-286). Monocular stimulation of the left eye was used with the stimulus presented on a cathode ray tube (CRT) with a 70 Hz refresh rate (IBM Model 8512). Only the superior hemifield was stimulated with the stimulus subtending a 24 x 24 degree field at the viewing distance of 40 cm. The stimulus consisted of 144 discs, each subtending a visual angle of 60 minutes and separated by a distance equal to their diameter (Fig. 1). The subjects maintained fixation by viewing a letter at the inferior edge of the stimulus field. Measurements were made in a darkened room with a pattern background luminance of 20 cd/m 2 (Minolta nt-1 photometer). Prior to making recordings, there was a three minute period for retinal light adaptation to the CRT background luminance. In order to investigate the responses to equal and opposite changes in stimulus contrast, three different stimulus luminance values were chosen; 0.1cd/m 2 (black), 10cd/m 2 (grey) and 40cd/m 2 (white) giving Weber Contrasts of - 1 0 0 % , - 5 0 % and +100% and, Michaelson Contrasts of - 1 0 0 % , - 3 0 % and +30% respectively (Fig. 2) [9]. In order to separate stimulus on- and off-set responses, the stimulus pattern was presented on the screen (on-time) for 400 miliseconds. This was followed by stimulus disappearance and uniform background (off-time) for 2 seconds. The stimulus 'pattern on-time : off-time ratio' was therefore 1 : 5. However, by alternate presentation of two patterns with the positions of the discs in one pattern corresponding to the interdisc areas of the other pattern, an effective "pattern on-time/off-time ratio" of 1 : 11 was achieved in order to minimise adaptation to luminance changes and to prevent distortion of spatial selectivity (Fig. 1) [10, 11].
175 Pattern
1 - o n : o f f ratio 1:11
L (cd/m 2)
On
40=
3O off
10 0
Pattern
2 - o n : o f f ratio
1:11
L (cd/m 2) 40
i
white
30 20
i
10
or g r e y
or black Stimulus
- o n : o f f ratio 1:5
L (cd/m 2)
40
V-
a0
r
I
10 0
400 ms
2s
Time
Fig. 2. Stimulus luminance (L) values for black, grey and white patterns. By alternate presentations of the patterns with the same contrast for 400 miliseconds (ms) with 2 second (s) intervals, an effective 'pattern on-time/off-time ratio' of 1 : 11 was achieved. (Weber's Contrast=Stimulus Luminance-Background Luminance/Background Luminance; Michaelson's Contrast = Stimulus Luminance-Background Luminance/Stimulus Luminance + Background Luminance. cd/m 2= candelas per squaremeter.)
The VECPs were recorded with silver-silver chloride electrodes and measured between the mid-occipital location, 5 cm above the inion on the midline and relative to the midfrontal location 12 cm above the nasion (i.e. Queen Square System) [12]. A guard electrode was positioned on the scalp within the hairline. A physiological amplifier (PA 400, Biodata Ltd, Manchester, UK) was used to amplify and filter the signals. The filter 3 dB
176 cut-off points were set at 0.8 Hz and 100 Hz. The VECPs were digitised at a rate of 320Hz and 800ms epochs were averaged. The averaging was triggered from stimulus on-set, the on-set (pattern appearance) response was contained in the first 400 ms of the recording and the off-set (pattern disappearance) response was contained in the second 400 ms of the recordings. One hundred pattern-onset and pattern-offset responses were recorded for each stimulus type. Recordings were made as two sets of 50 responses to check signal reproducibility. The two recordings were summed and filtered with a digital low-pass filter to attenuate residual high frequency noise. The recordings were evaluated by identifying their common reproducible features which were then quantified by measuring their latency and relative amplitude. A statistical evaluation of the recordings was made by using the "paired t-test".
Results
Both the on-set and off-set responses showed a similar series of two negative and positive peaks (Fig. 3). The recordings were similar in all subjects. The full results are presented in Table 1. The P125-N170 and N170-P225 amplitudes of the on-set responses to the bright stimulus pattern were significantly larger than those to the dark stimulus patterns ( p < 0.01) (Fig. 4). The average P125-N170 amplitude to white stimulus was 36% larger than that to the black, and was 46% larger than that to the grey. Similarly, the average N170-P225 amplitude to white stimulus was 53% larger than that to the black and grey stimuli. The
10
Amplitude (pV) P125 P515
_' , Pe.,o.j
P225
-5 I
N75 -10 0
l 100
N170 i 200
L 300
400
Time tstimulus on-set
i
i
I
500
600
700
800
(ms)
t stimulus off-set
Fig. 3. On-set and Off-set responses from the superior hemifield to three different contrast pattern stimuli showed similar series of negative and positive peaks in all subjects.
Pattern contrast - 100% -50% + 100%
Pattern contrast -100% -50% +100% 473.8• 478.2 • 22.9 467.5•
7.3• 6.8• 7.2•
9.2• 8.3• 12.5•
7.3• 7.3• 11.2•
5.7• 4.2• 3.7•
580.4• 584.2 • 16.6 577.3-+8.5
8.3• 6.3• 4.6•
P515-N580
514.2• 516.7 +- 15.4 515.5•
Off-Set Responses N170-P225
Amplitudes (pN)
213.4• 234.2 • 25.6 231.7•
N470-P515
P125-N170
169.1• 174.8 • 10.6 165.9•
On-Set Responses
122.4• 129.3 • 11.8 124.9•
Off-Set Responses
Latencies (Ins)
N75-P125
80,1• 81.4 • 8.7 70.2•
On-Set Responses
5.4• 4.3• 4.9•
N580-P640
641.2• 634.9 • 22.9 641.6•
Table 1. The latencies and amplitudes of visual evoked cortical potential recordings to black ( - 1 0 0 % ) , grey ( - 5 0 % ) and white (+ 100%) pattern stimuli in five normal subjects (Mean • Standard Deviation, ~V= microvolt, ms = milisecond).
.-,a ---,i
178 JV 20
15
T
10
N75-P125
BGW
BGW
GW i
P125-N170
M e a n _+ SD
i N170-P225
B:black G:grey W:white
Fig. 4. V E C P amplitudes of On-set responses (n = 5, Mean -+ Standard Deviation) to black, grey and white pattern stimuli (IxV = microvolt).
N75-P125 amplitude, however, showed no significant difference between the responses to the three stimuli (p >0.05). There were also no significant differences in the latency of the responses (p >0.05). The off-set of both dark stimuli produced a larger response than the off-set of the bright stimulus (Fig. 5). This finding is consistent with the ,pV
15
I
10= @
M e a n * SD
5-
B GW N470-P515
i
B G W P515-N580
BGW N580-P640
B:black G:grey W:white
Fig. 5. V E C P amplitudes of Off-set responses (n = 5, Mean -+ Standard Deviation) to black, grey and white pattern stimuli (IxV = microvolt).
179 results from the on-set stimuli because off-set of a dark pattern creates a light increment which acts as a bright stimulus and vice versa. The average N470-P515 amplitude from the off-set of the black pattern was significantly (54%) larger than that from the white ( p < 0.01). Similarly, the off-set of the grey pattern gave significantly (14%) larger N470-P515 recordings than for the white ( p < 0.05). Also, the P515-N580 amplitudes to black and grey were 80% and 37% larger respectively. The N580-P640 amplitude showed no significant difference in the responses from the three different contrasts ( p > 0.05). As with the on-set stimuli, there were also no significant latency differences in the off-set responses ( p > 0.05).
Discussion
The dark stimulus (light decrement), by definition, excites the Off-Pathway. In this study, we demonstrate that visual evoked cortical responses to light decrements are significantly smaller than responses to light increments in normal subjects. Sensitivity of the retina to light stimuli decreases towards the periphery of the visual field ('Hill of Vision'). This change in retinal light sensitivity is explained by the fact that the number of retinal ganglion cells decreases with increasing eccentricity in such a way as to cause an increased convergence of photoreceptor cells onto ganglion cells. The 1 : 1 photoreceptor/ganglion cell ratio at the macular region becomes 100 : 1 or more peripherally and this results in enlargement of the size of retinal receptive fields with reduced overlap and consequent sensitivity loss [13]. The smaller amplitudes of the responses to light decrements may be explained in a similar way. Off-Centre ganglion cells are known to be less numerous than the On-Centre ganglion cells in the monkey retina, with the Off/On-Centre type cell ratio being 1/3-2/3 depending on the eccentricity [14, 15]. If this difference also exists in the human eye, it may lead to the convergence of a larger number of photoreceptor cells on to Off-Centre retinal ganglion cells than on to On-Centre cells, which would result in larger dendritic receptive fields with reduced overlap and consequent less sensitivity to a given stimulus size. It has previously been suggested that all rod bipolars are 'On' type and cone bipolars are both 'On' and 'Off' type [16, 17]. Therefore, it is possible that the rod contribution to the 'On-Pathway' may also, to some extent, account for the amplitude differences between responses. Additionally, a higher metabolic and discharge activity rate may also exist in the On-Pathway of the human visual system which might contribute to the discrepancy in the magnitude of the responses from both pathways [18]. If 'Off-Centre' and 'On-Centre' ganglion cells and 'Off' and 'On' neural pathways are equally sensitive to several forms of injury, e.g. ocular hypertension, compression, demyelination, ischemia etc., then the functional reserve of 'Off' cells may become depleted before any deficit of the 'On'
180 system is detectable; in addition, the signs of such a depletion may be more prominent in the 'Off' Pathway. In glaucoma, it is known that a subclinical period exists between the onset of retinal ganglion cell damage and the appearance of reproducible visual field defects. Hence, it has been suggested that psychophysical tests should take into account the 'functional reserves' of the group of the cells being tested [19]. If dark stimuli preferentially test the 'Off' pathway, and if the 'Off' pathway has a smaller functional reserve than the 'On' pathway, then the sensitivity of electrophysiological and perimetric examinations may be increased by the use of dark stimuli.
Acknowledgement E.M. was supported by a fellowship from The Royal National Institute for the Blind. M.B. and the electrophysiology equipment were funded by the Scottish Office Home and Health Department, Principal Grant Holder Dr. D. Montgomery. Photographs by Mr John McCormick.
References 1. Wassle H, Boycott BB. Functional architecture of the mammalian retina. Physiological Rev 1991; 71(2): 447-480. 2. Kageyama GH, Wong-Riley MTT. The histochemical localisation of cyto-chrome oxidase in the retina and lateral geniculate nucleus of the ferret, cat, and monkey, with particular reference to retinal mosaics and on/off-center visual channels. The Journal of Neuroscience 1984; 4: 2445-2459. 3. Peichl L. Zur Organisation der Netzhaut: Struktur/Funktions-Beziehungen und ein Spezievergleich retinaler Ganglienzellen. Fortschr Ophthalmol 1989; 86: 47-53. 4. Perry VH, Silveria LCL. Functional lamination in the ganglion cell layer of the macaque's retina. Neuroscience 1988; 25: 217-224. 5. Shechter S, Hochstein S. On and Off-Pathway contributions to apparent motion perception. Vision Res 1990; 30: 1189-1204. 6. Schiller PH, Sandell JH, Maunsell JHR. Functions of the On and Off channels of the visual system. Nature (London) 1986; 322: 824-825. 7. Mutlukan E, Damato BE. The dark perimetric stimulus. Brit J Ophthalmol 1992; 76: 264-267. 8. Schiller PH. Central connections of the retinal ON and OFF pathways. Nature 1982; 297: 580-583. 9. Westheimer G. The oscilloscopic view: Retinal illuminance and contrast of point and line targets. Vision Res 1985; 25: 1097-2003. 10. Barber C, Galloway NR. Adaptation effects in the transient visual evoked potential. In: Lehmann D, Calway E, eds. Human Evoked Potentials: Applications and Problems. New York: Plenum Press, 1979: 17-30. 11. Drasdo N, Thompson D. Distortion of spatial selectivity by pattern onset stimulation. Doc Ophthalmol 1989; 72: 1-8. 12. Blumhardt LD, Barett G, Halliday AM. The asymmetrical visual evoked potential to pattern reversal in one half field and its significance for the analysis of visual field defects. Br J Ophthatmol 1977; 61: 454-461.
181 13. Tate GW. The physiological basis for perimetry. In: Drance SM, Anderson DA, eds. Automatic Perimetry in Glaucoma. Orlando, FL: Grunne-Stratton, 1985: 1-28. 14. de Monesterio FM, Gouras P. Functional properties of ganglion cells of the rhesus monkey retina. J Physiol (London) 1975; 25l: 167-195. 15. Schiller PH. The central visual system. Vision Res 1986; 26: 1351-1388. 16. Wassle H. Sampling of visual space by retinal ganglion cells. Invest Ophthalmol Vis Sc, Suppl. 1988; 29: 117. 17. Mfiller F, Wassle H, Voigt T. Pharmacological manipulation of the rod pathway in the cat retina. Neuro-Physiology 1988; 59: 1657-1672. 18. Liu S, Wong-Riley M. Quantitative light and electron-microscopic analysis of cytochromeoxidase distribution in neurons of the lateral geniculate nucleus of the adult monkey. Visual Neuro-Science 1990; 4: 269-287. 19. Glovinsky Y, Quigley HA, Dunkelberger GR. Retinal ganglion cell loss is size dependent in experimental glaucoma. Invest Ophthalmol Vis Sc, 1991; 32: 484-491.
Address for correspondence: Erkan Mutlukan, MD, Tennent Institute of Ophthalmology, University of Glasgow, 38 Church Street, Glasgow G l l 6NT, Scotland. Phone: (41) 339 8822, Ext 4034; Fax: (41) 339 7485.
9
Visual evoked cortical potentials from transient dark and bright stimuli Selective 'on' and 'off-pathway' testing? E R K A N MUTLUKAN, I M I C H A E L B R A D N A M , 2 DAVID KEATING 2 & B E R T I L E. D A M A T O 1 tTennent Institute of Ophthalmology, University of Glasgow, 38 Church Street, Glasgow Gll 6NT, Scotland; ZWest of Scotland Health Boards Department of Clinical Physics and Bio-Engineering, 11 West Graham Street, Glasgow G4 9LF, Scotland Accepted 12 April 1992
Key words: on-pathway, off-pathway, parallel pathways, retinal ganglion cell, visual cortex, contrast, electrophysiology, VECP, cortical potentials, visual evoked response Abstract. The superior hemifields of five normal left eyes were stimulated by novel equal and opposite contrast pattern onset stimuli which were generated on a cathode ray tube. Patterns consisted of 144 discs, each subtending 60 rain arc at the viewing distance of 40 cm and were separated by a distance equal to their diameter. Equal and opposite light changes were created by presenting the disc patterns with different luminance values on a uniform constant background (20cd/m2). Transient visual evoked cortical potentials to the appearance and disappearance of the patterns were recorded separately and analysed. Significant amplitude differences between the responses to bright and dark stimuli were observed with light increment responses being 36-53% larger than the light decrement responses for pattern on-set and 54-80% larger for pattern off-set respectively. This fnding is attributed to the difference in the input to the ON and OFF Parallel Pathways which are known to carry light increment and decrement information respectively, as well as differences in the metabolic and discharge rates of these pathways.
Introduction The most commonly employed stimuli for measuring pattern visual evoked cortical potentials (VECPs) are checkerboard and grating patterns, which consist of alternating dark and light elements. With such patterns, light increments and decrements occur simultaneously at each reversal, so that the average luminance is steady. The combined visual response to both the bright and dark components of the stimulus pattern is therefore recorded simultaneously. It is believed that light increments (bright stimulus on dark background) and decrements (dark stimulus on bright background) are processed by two separate channels in the visual system, namely 'On' and 'Off'-Pathways, which exist between the retinal bipolar cell layer and visual cortex, and which display morphological, physiological, pharmacological and psycho-
172
Fig. 1 (a-f). Black, grey and white pattern stimuli. The positions of the discs in one pattern
173
corresponded to the interdisc areas of the other pattern to decrease the on/off time ratio by alternate presentations.
174 physical differences [1-8]. In view of these differences, it is possible that 'On' and 'Off' Pathways also differ in their susceptibility to damage in glaucoma and other neuro-ophthalmic diseases. It may therefore be clinically useful to test 'On' and 'Off' pathways separately when performing electrophysiological examinations. We have measured visual evoked cortical responses to bright and dark stimuli presented sequentially against a constant background. The superior hemifield was examined because this part of the visual field tends to be lost relatively early in conditions such as glaucoma and chiasmal compression. In this article, we describe the bright and dark stimuli we have developed and report the results of a preliminary study which shows differences between responses obtained with such stimuli.
Materials and methods
Five normal subjects were examined, three male and two female, ranging in age from 26 to 35 year (mean 30). All subjects had a corrected visual acuity of 6/6 or better, normal pupils (3-6 mm) and normal visual fields. The stimulus and recording system was based on a personal computer (IBM Model 30-286). Monocular stimulation of the left eye was used with the stimulus presented on a cathode ray tube (CRT) with a 70 Hz refresh rate (IBM Model 8512). Only the superior hemifield was stimulated with the stimulus subtending a 24 x 24 degree field at the viewing distance of 40 cm. The stimulus consisted of 144 discs, each subtending a visual angle of 60 minutes and separated by a distance equal to their diameter (Fig. 1). The subjects maintained fixation by viewing a letter at the inferior edge of the stimulus field. Measurements were made in a darkened room with a pattern background luminance of 20 cd/m 2 (Minolta nt-1 photometer). Prior to making recordings, there was a three minute period for retinal light adaptation to the CRT background luminance. In order to investigate the responses to equal and opposite changes in stimulus contrast, three different stimulus luminance values were chosen; 0.1cd/m 2 (black), 10cd/m 2 (grey) and 40cd/m 2 (white) giving Weber Contrasts of - 1 0 0 % , - 5 0 % and +100% and, Michaelson Contrasts of - 1 0 0 % , - 3 0 % and +30% respectively (Fig. 2) [9]. In order to separate stimulus on- and off-set responses, the stimulus pattern was presented on the screen (on-time) for 400 miliseconds. This was followed by stimulus disappearance and uniform background (off-time) for 2 seconds. The stimulus 'pattern on-time : off-time ratio' was therefore 1 : 5. However, by alternate presentation of two patterns with the positions of the discs in one pattern corresponding to the interdisc areas of the other pattern, an effective "pattern on-time/off-time ratio" of 1 : 11 was achieved in order to minimise adaptation to luminance changes and to prevent distortion of spatial selectivity (Fig. 1) [10, 11].
175 Pattern
1 - o n : o f f ratio 1:11
L (cd/m 2)
On
40=
3O off
10 0
Pattern
2 - o n : o f f ratio
1:11
L (cd/m 2) 40
i
white
30 20
i
10
or g r e y
or black Stimulus
- o n : o f f ratio 1:5
L (cd/m 2)
40
V-
a0
r
I
10 0
400 ms
2s
Time
Fig. 2. Stimulus luminance (L) values for black, grey and white patterns. By alternate presentations of the patterns with the same contrast for 400 miliseconds (ms) with 2 second (s) intervals, an effective 'pattern on-time/off-time ratio' of 1 : 11 was achieved. (Weber's Contrast=Stimulus Luminance-Background Luminance/Background Luminance; Michaelson's Contrast = Stimulus Luminance-Background Luminance/Stimulus Luminance + Background Luminance. cd/m 2= candelas per squaremeter.)
The VECPs were recorded with silver-silver chloride electrodes and measured between the mid-occipital location, 5 cm above the inion on the midline and relative to the midfrontal location 12 cm above the nasion (i.e. Queen Square System) [12]. A guard electrode was positioned on the scalp within the hairline. A physiological amplifier (PA 400, Biodata Ltd, Manchester, UK) was used to amplify and filter the signals. The filter 3 dB
176 cut-off points were set at 0.8 Hz and 100 Hz. The VECPs were digitised at a rate of 320Hz and 800ms epochs were averaged. The averaging was triggered from stimulus on-set, the on-set (pattern appearance) response was contained in the first 400 ms of the recording and the off-set (pattern disappearance) response was contained in the second 400 ms of the recordings. One hundred pattern-onset and pattern-offset responses were recorded for each stimulus type. Recordings were made as two sets of 50 responses to check signal reproducibility. The two recordings were summed and filtered with a digital low-pass filter to attenuate residual high frequency noise. The recordings were evaluated by identifying their common reproducible features which were then quantified by measuring their latency and relative amplitude. A statistical evaluation of the recordings was made by using the "paired t-test".
Results
Both the on-set and off-set responses showed a similar series of two negative and positive peaks (Fig. 3). The recordings were similar in all subjects. The full results are presented in Table 1. The P125-N170 and N170-P225 amplitudes of the on-set responses to the bright stimulus pattern were significantly larger than those to the dark stimulus patterns ( p < 0.01) (Fig. 4). The average P125-N170 amplitude to white stimulus was 36% larger than that to the black, and was 46% larger than that to the grey. Similarly, the average N170-P225 amplitude to white stimulus was 53% larger than that to the black and grey stimuli. The
10
Amplitude (pV) P125 P515
_' , Pe.,o.j
P225
-5 I
N75 -10 0
l 100
N170 i 200
L 300
400
Time tstimulus on-set
i
i
I
500
600
700
800
(ms)
t stimulus off-set
Fig. 3. On-set and Off-set responses from the superior hemifield to three different contrast pattern stimuli showed similar series of negative and positive peaks in all subjects.
Pattern contrast - 100% -50% + 100%
Pattern contrast -100% -50% +100% 473.8• 478.2 • 22.9 467.5•
7.3• 6.8• 7.2•
9.2• 8.3• 12.5•
7.3• 7.3• 11.2•
5.7• 4.2• 3.7•
580.4• 584.2 • 16.6 577.3-+8.5
8.3• 6.3• 4.6•
P515-N580
514.2• 516.7 +- 15.4 515.5•
Off-Set Responses N170-P225
Amplitudes (pN)
213.4• 234.2 • 25.6 231.7•
N470-P515
P125-N170
169.1• 174.8 • 10.6 165.9•
On-Set Responses
122.4• 129.3 • 11.8 124.9•
Off-Set Responses
Latencies (Ins)
N75-P125
80,1• 81.4 • 8.7 70.2•
On-Set Responses
5.4• 4.3• 4.9•
N580-P640
641.2• 634.9 • 22.9 641.6•
Table 1. The latencies and amplitudes of visual evoked cortical potential recordings to black ( - 1 0 0 % ) , grey ( - 5 0 % ) and white (+ 100%) pattern stimuli in five normal subjects (Mean • Standard Deviation, ~V= microvolt, ms = milisecond).
.-,a ---,i
178 JV 20
15
T
10
N75-P125
BGW
BGW
GW i
P125-N170
M e a n _+ SD
i N170-P225
B:black G:grey W:white
Fig. 4. V E C P amplitudes of On-set responses (n = 5, Mean -+ Standard Deviation) to black, grey and white pattern stimuli (IxV = microvolt).
N75-P125 amplitude, however, showed no significant difference between the responses to the three stimuli (p >0.05). There were also no significant differences in the latency of the responses (p >0.05). The off-set of both dark stimuli produced a larger response than the off-set of the bright stimulus (Fig. 5). This finding is consistent with the ,pV
15
I
10= @
M e a n * SD
5-
B GW N470-P515
i
B G W P515-N580
BGW N580-P640
B:black G:grey W:white
Fig. 5. V E C P amplitudes of Off-set responses (n = 5, Mean -+ Standard Deviation) to black, grey and white pattern stimuli (IxV = microvolt).
179 results from the on-set stimuli because off-set of a dark pattern creates a light increment which acts as a bright stimulus and vice versa. The average N470-P515 amplitude from the off-set of the black pattern was significantly (54%) larger than that from the white ( p < 0.01). Similarly, the off-set of the grey pattern gave significantly (14%) larger N470-P515 recordings than for the white ( p < 0.05). Also, the P515-N580 amplitudes to black and grey were 80% and 37% larger respectively. The N580-P640 amplitude showed no significant difference in the responses from the three different contrasts ( p > 0.05). As with the on-set stimuli, there were also no significant latency differences in the off-set responses ( p > 0.05).
Discussion
The dark stimulus (light decrement), by definition, excites the Off-Pathway. In this study, we demonstrate that visual evoked cortical responses to light decrements are significantly smaller than responses to light increments in normal subjects. Sensitivity of the retina to light stimuli decreases towards the periphery of the visual field ('Hill of Vision'). This change in retinal light sensitivity is explained by the fact that the number of retinal ganglion cells decreases with increasing eccentricity in such a way as to cause an increased convergence of photoreceptor cells onto ganglion cells. The 1 : 1 photoreceptor/ganglion cell ratio at the macular region becomes 100 : 1 or more peripherally and this results in enlargement of the size of retinal receptive fields with reduced overlap and consequent sensitivity loss [13]. The smaller amplitudes of the responses to light decrements may be explained in a similar way. Off-Centre ganglion cells are known to be less numerous than the On-Centre ganglion cells in the monkey retina, with the Off/On-Centre type cell ratio being 1/3-2/3 depending on the eccentricity [14, 15]. If this difference also exists in the human eye, it may lead to the convergence of a larger number of photoreceptor cells on to Off-Centre retinal ganglion cells than on to On-Centre cells, which would result in larger dendritic receptive fields with reduced overlap and consequent less sensitivity to a given stimulus size. It has previously been suggested that all rod bipolars are 'On' type and cone bipolars are both 'On' and 'Off' type [16, 17]. Therefore, it is possible that the rod contribution to the 'On-Pathway' may also, to some extent, account for the amplitude differences between responses. Additionally, a higher metabolic and discharge activity rate may also exist in the On-Pathway of the human visual system which might contribute to the discrepancy in the magnitude of the responses from both pathways [18]. If 'Off-Centre' and 'On-Centre' ganglion cells and 'Off' and 'On' neural pathways are equally sensitive to several forms of injury, e.g. ocular hypertension, compression, demyelination, ischemia etc., then the functional reserve of 'Off' cells may become depleted before any deficit of the 'On'
180 system is detectable; in addition, the signs of such a depletion may be more prominent in the 'Off' Pathway. In glaucoma, it is known that a subclinical period exists between the onset of retinal ganglion cell damage and the appearance of reproducible visual field defects. Hence, it has been suggested that psychophysical tests should take into account the 'functional reserves' of the group of the cells being tested [19]. If dark stimuli preferentially test the 'Off' pathway, and if the 'Off' pathway has a smaller functional reserve than the 'On' pathway, then the sensitivity of electrophysiological and perimetric examinations may be increased by the use of dark stimuli.
Acknowledgement E.M. was supported by a fellowship from The Royal National Institute for the Blind. M.B. and the electrophysiology equipment were funded by the Scottish Office Home and Health Department, Principal Grant Holder Dr. D. Montgomery. Photographs by Mr John McCormick.
References 1. Wassle H, Boycott BB. Functional architecture of the mammalian retina. Physiological Rev 1991; 71(2): 447-480. 2. Kageyama GH, Wong-Riley MTT. The histochemical localisation of cyto-chrome oxidase in the retina and lateral geniculate nucleus of the ferret, cat, and monkey, with particular reference to retinal mosaics and on/off-center visual channels. The Journal of Neuroscience 1984; 4: 2445-2459. 3. Peichl L. Zur Organisation der Netzhaut: Struktur/Funktions-Beziehungen und ein Spezievergleich retinaler Ganglienzellen. Fortschr Ophthalmol 1989; 86: 47-53. 4. Perry VH, Silveria LCL. Functional lamination in the ganglion cell layer of the macaque's retina. Neuroscience 1988; 25: 217-224. 5. Shechter S, Hochstein S. On and Off-Pathway contributions to apparent motion perception. Vision Res 1990; 30: 1189-1204. 6. Schiller PH, Sandell JH, Maunsell JHR. Functions of the On and Off channels of the visual system. Nature (London) 1986; 322: 824-825. 7. Mutlukan E, Damato BE. The dark perimetric stimulus. Brit J Ophthalmol 1992; 76: 264-267. 8. Schiller PH. Central connections of the retinal ON and OFF pathways. Nature 1982; 297: 580-583. 9. Westheimer G. The oscilloscopic view: Retinal illuminance and contrast of point and line targets. Vision Res 1985; 25: 1097-2003. 10. Barber C, Galloway NR. Adaptation effects in the transient visual evoked potential. In: Lehmann D, Calway E, eds. Human Evoked Potentials: Applications and Problems. New York: Plenum Press, 1979: 17-30. 11. Drasdo N, Thompson D. Distortion of spatial selectivity by pattern onset stimulation. Doc Ophthalmol 1989; 72: 1-8. 12. Blumhardt LD, Barett G, Halliday AM. The asymmetrical visual evoked potential to pattern reversal in one half field and its significance for the analysis of visual field defects. Br J Ophthatmol 1977; 61: 454-461.
181 13. Tate GW. The physiological basis for perimetry. In: Drance SM, Anderson DA, eds. Automatic Perimetry in Glaucoma. Orlando, FL: Grunne-Stratton, 1985: 1-28. 14. de Monesterio FM, Gouras P. Functional properties of ganglion cells of the rhesus monkey retina. J Physiol (London) 1975; 25l: 167-195. 15. Schiller PH. The central visual system. Vision Res 1986; 26: 1351-1388. 16. Wassle H. Sampling of visual space by retinal ganglion cells. Invest Ophthalmol Vis Sc, Suppl. 1988; 29: 117. 17. Mfiller F, Wassle H, Voigt T. Pharmacological manipulation of the rod pathway in the cat retina. Neuro-Physiology 1988; 59: 1657-1672. 18. Liu S, Wong-Riley M. Quantitative light and electron-microscopic analysis of cytochromeoxidase distribution in neurons of the lateral geniculate nucleus of the adult monkey. Visual Neuro-Science 1990; 4: 269-287. 19. Glovinsky Y, Quigley HA, Dunkelberger GR. Retinal ganglion cell loss is size dependent in experimental glaucoma. Invest Ophthalmol Vis Sc, 1991; 32: 484-491.
Address for correspondence: Erkan Mutlukan, MD, Tennent Institute of Ophthalmology, University of Glasgow, 38 Church Street, Glasgow G l l 6NT, Scotland. Phone: (41) 339 8822, Ext 4034; Fax: (41) 339 7485.
