- Email: [email protected]
The electroretinogram of the little owl (Athene noctua)
Vision Res. Vol. 29, No. 12, pp. 1693-1698,1989 Printed in Great Britain. All rights reserved
0042-6989/89$3.00+ 0.00 Copyright 0 1989Pergamon Press plc
THE ELECTRORETINOGRAM OF THE LITTLE (ATHENE NOCTUA) VI~~ORIO PORCIATTI, GIGLIOLA FONTANW
OWL
and PAOLA BAGNOLI
Department of Physiology and Biochemistry, University of Pisa and Institute of Neurophysiology, CNR, Via S. Zeno 31-51, Pisa 56100, Italy (Receioed 21 October
1988; in revised form 17 February 1989)
Abstract-Electroretinographic responses (ERGS) have been recorded from the cornea of the little owl (Athene noctua) in response to single light flashes and to alternating sinusoidal gratings (pattern) at different levels of light adaptation. Both flash- and pattern-evoked ERGS show scotopic as well as photopic components. The pattern evoked ERG is spatially tuned with tuning functions which shift towards lower frequencies by reducing the mean luminance. The retinal acuity is about 6c/deg at 2.3 logcd/m2 and decreases progressively by reducing the mean luminance. No pattern ERG can be recorded beyond -6.7 logcd/m* at any spatial frequency. The pattern ERG amplitude decreases progressively by reducing the contrast. The extrapolated contrast threshold is about 1%. Acuity and contrast sensitivity ERG values are in the range of those obtained by operant techniques in other species with duplex retinae such as owls and cats. Flash ERG
Pattern ERG
Little owl
Luminance levels
METHODS
INTRODUCTION
Among birds, owls possess eyes which are highly adapted for nocturnal viewing, with a well developed scotopic system (Fite, 1973). However, owl eyes also possess functional retinal cones numerous enough to mediate colour vision (Martin, 1974; Martin & Gordon, 1974a; Martin, Gordon & Cadee, 1975; Bowmaker & Martin, 1978). Among the Strigiformes, there are species that are strongly nocturnal, such as the tawny owl (Strix aluco). In contrast, the little owl (Athene noctua) which is common in Northern Italy and Sardinia, is primarily crepuscular with some diurnal activity. In the present study, the electroretinogram (ERG) of the little owl Athene noctua in response to light flashes under light and dark adaptation was recorded in order to evaluate the relative contribution of photopic and scotopic systems to retinal function. In addition, electroretinographic responses to sinusoidal gratings (Pattern ERGS, PERGs) of different contrast and spatial frequency were recorded under different luminance levels in order to determine the spatial domain at which the retina is operating.
Subjects Two little owls (Athene noctua) were used in this investigation. They were obtained from a local institution collecting protected birds with severe wing and leg damage. The owls underwent several experiments during the winter time at dusk. After each experiment, the two birds fully recovered from anaesthesia and were returned to their cages. Both animals were subsequently used in another investigation aimed at elucidating functional and anatomical properties of the geniculo-striate pathway (Casini, Fontanesi, Porciatti & Bagnoli, 1988). Electrophysiological techniques In the present study, the two animals (body weight 180 and 192 g, respectively) were anaesthetized with Chloral hydrate (0.2 ml/100 g body weight) injected into breast muscles. The anaesthetic dose was sufficient to render the birds unconscious during the 2 hr registration period. Owls were then mounted on a modified stereotaxis apparatus. The head was gently pressed against the beak holder by means of an elastic band. No pressure points were used. Pupils were dilated to a diameter of 6 mm by
1693
1694
VITTORIOPORCIATTIet al.
instillation of 1.O% cyclopentolate hydrochloride (Martin, Gordon & Cadee, 1975). The ERG was recorded simultaneously from both eyes by means of a hook-shaped gold lamina inserted into the lower fornix following instillation of oxibromocaine. A gentle downward traction of electrode wires served as a good substitute for blepharostat. The reference electrode was a subcutaneous stainless steel needle placed into the skin of the posterior part of the skull. A similar electrode placed in the upper part of the skull served as a ground. Visual stimulation
obtained by means of either retinoscopy or photorefraction (Murphy & Howland, 1983). Electrical signals were amplified (flash ERG : 10,000; PERG : 50,000), filtered (flash ERG; t-250 Hz; PERG: l-50 Hz, -6 dB/oct), sampled at different frequencies according to resolution and time base needed (0.5-5 KHz) and averaged (10 bit resolution, flash ERG: 20 events; PERG: 250 events). Averaged “noise” responses were obtained for any stimulus conditions by occluding the visual stimuli. Fourier analysis (resolution: 1 Hz) of the averaged responses was performed off line in order to establish the ERG flicker fusion (signal-to-noise ratio & 1.5).
Single light flashes were delivered by a strobe lamp (Energy: 5 J; flash duration*: 1.5 msec; RESULTS colour temperature: 5.9oO”K) placed inside a Ganzfeld bowl (45 cm dia.) at a rate of O.l/sec Feats ERG for the scotopic ERG or l/set for the photopic Figure IA, B shows examples of ERGS ERG. When measuring ERG flicker fusion frequency (FFF) the strobe lamp was driven with recorded in response to flashes of different a lower power (Energy: 0.2 J; flash duration*: intensities under light (A) and dark (B) adap175 psec; colour temperature: 5900°K) in such a tation. ERGS recorded simultaneously from way as its intensity was independent of flicker both eyes are shown superimposed. Since ERG rate in the range l-60 flasheslsec. Neutral den- waveforms were comparable in the two eyes at sity filters were placed in front of the strobe the different intensity levels, amplitude and lamp to attenuate flash intensity. Before record- latency values of a- and b-waves have been ings animals were kept 45 min in a darkened or averaged. As shown in Fig. lC, D the b-wave lightened (250 lx at the eye) room in order to amplitude as well as a- and b-wave latencies preadapt them to dark and light, respectively. change as a function of light intensity. It can be Recordings were obtained after either a 15 min seen that the b-wave increases in amplitude as period of dark adaptation or 3min period of the flash intensity is increased, under both dark light and light adaptation, and tends to saturate at light adaptation to background (100 cd/m*). A check was made that adap~tion the highest intensities (Fig. 1C). The difference periods to dark and light were sufficient to between the photopic and scotopic threshold is obtain maximal scotopic and photopic ERGS, of the order of 8 log units. As shown in Fig. lD, respectively. Vertical sinusoidal gratings of under light adaptation the latency of a- and different contrasts and spatial frequencies were b-waves tends to increase as the flash intensity generated by a microcomputer on a Joyce TV is decreased. In contrast, under dark adapraster display (white phosphor 200 cd/m2 mean tation, a striking increase in latency at low flash luminance, 20 deg field size at 57 cm distance). intensities for both a- and b-waves can be The stimulus display was surrounded by a large observed. white cardboard whose integrated luminance In Fig. 2, ERG flicker fusion frequency (FFF) was kept at the same level as the stimulus. as a function of flash intensity is shown. In the Refraction was measured objectively by placing range of low flash intensities the FFF progrestrials lens before the eyes and determining the sively increases as the flash intensity is increased trial lenses at which the PERG amplitude and tends to saturate at a value of about 20 (2c/deg, 95% contrast, 200cd/m’) was largest. flashes/set. Further increases in flash intensity Usually no lenses or + 1 sph. diopters were induce a rapid rise in FFF. At the highest necessary to obtain maximal PERG responses intensities the FFF again tends to saturate. This double branched plot is typical of duplex retinal. at 57 cm distance. This objective refraction agreed with previous measurements in owls Pattern ERG *Taken at l/3 of peak amplitude.
Figure 3A shows examples of PERGs in response to sinusoidal gratings (95% contrast)
Owl ERG
1695
PHOTOPIC
-8.19
+
. 1 -10
-9
-8
-7
-6
LOG RELATIVE
-8.59
D
+
-5 FLASH
-4
-3
-2
-1
0
INTENSITY
210
1
-1.38
7 4
-8.8g-bP+ -919
-269
w
.
110.
s -I -9.59
.
.
l l
8s l* .=..
-
oooooo
! 125 pi’
00
IO* -10
204msec
, -9
, -8
, -7
LOG RELATIVE
, -6
, -5 FLASH
. -4
%no 4
-3
-2
-1
0
INTENSITY
Fig. 1. Examples of ERGs recorded s~~taneously from both eyes in response to light Rashes of different relative intensity (indicated by numbers to the Ieft of each pair of tracings} under condition of light (A) and dark (B) adaptation (unattenuated flash energy 5 J, flash duration; 1.5 msec; colour temperature: 5900°K, background light: 100 cd/m*). Variation in amplitude (C) and peak latency (D) of the major ERG components (circles: b-wave; squares: a-wave) as a function of flash intensity under conditions of dark (solid symbols) and light (open symbols) adaptation.
of different spatial frequencies and mean luminances, reversed in their spatial phase (squarewave) at 2 Hz. At low spatial frequency and high luminance, the response consists of a small negative wave and a main positive complex
0
0
0
0 0
0
0 0
LOG
00
0
00
oooo
RELATIVE
FLASH
INTENSITY
Fig. 2. Relationship between ERG tlicker fusion frequency (FFF) and flash intensity. Note the double branched aspect of the plot which is typical of duplex retinae (unattenuated flash energy: 0.2 J, flash duration 17544~; colour temperature: 5900°K). VR 29/l2-D
which slowly returns to the base line. The main positive wave can be approximately divided in two sub-components (early and late) which display different characteristics as a function of spatial frequency and luminance. It can be noticed that both early and late PERG positive components progressively decrease in amplitude as a function of spatial frequency. However, the relative cont~bution of the two response components depends on mean luminance. At 200cd/m2 the BERG is dominated by the early component whereas at 20 and 2 cd/m* the PERG is dominated by the late component. These characteristics suggest that the early component is photopic in origin whereas the late component is scotopic. Further support for a scotopic origin of the late PERG component comes from the observation that this is not present in the PERG of diurnal birds (Bagnoli, Porciatti, Francesconi & Barsellotti, 1984; Bagnoli, Porciatti & Francesconi, 1985). As shown in Fig. 3B the amplitude of the photopic PERG component changes as a func-
VITTORIO PORCIATTI et al.
1696
s
200
,lfy
x
-.I------
zz
I~SJN 204msec
6.
5 Y y
I
0
0 2
0
-1.5
-1
2
00
---
-0.5
LOG
SPATIAL
0 --______ 20 0
0
_
FREQUENCY
(
---
0.5 (cldegl
1
Fig. 3. (A) Examplesof PERGs in response to sinusoidal gratings (95% contrast; 2 Hz alternation frequency) of differential spatial frequencies (indicated by numbers to the left of each tracing) and mean luminances (indicated by numbers to the top of each column). Note the complex waveform and the slow decay of the owl PERG. Also note that the early PERG positive component (photopic) decreases in amplitude by increasing the stimulus spatial frequency and reducing the mean luminance. The late PERG positive component (scotopic) dominates the response at low luminance levels (20 and 2cd/m2). (B) Amplitude variation of the photopic component of the PERG as a function of the stimulus spatial frequency and mean luminance (indicated by numbers close to each curve). Note that the spatial frequency function has low pass like characteristics and shifts to the left by reducing the mean luminance. Concurrently, the high spatial frequency threshold (retinal acuity) decreases. The dashed line represents the noise level.
tion of spatial frequency and mean luminance in the photopic range. At 200 and 20cd/m2 the PERG amplitude is approximately constant for spatial frequencies ranging between 0.125 c/deg and 1 c/deg. For spatial frequencies higher than 1 c/deg the PERG amplitude shows a marked decrease. At 2 cd/m2 the PERG monotonically decreases in amplitude as the spatial frequency increases. It can be noticed that the high spatial frequency threshold decreases by reducing the mean luminance. Threshold responses were considered those whose peak-to-trough amplitude was twice as large as the noise level. In Fig. 4 changes in high spatial frequency threshold (retinal acuity) as a function of mean luminance are reported. Retinal acuity progressively decreases by reducing the mean hnninance over 9 log units. In the photopic range, however, the reduction in retinal acuity is more marked than in the scotopic one. At luminances lower than - 6.7 log cd/m2 no response can be recorded at any spatial frequency. Steady-state (8.3 Hz) PERGs were also recorded in response to sinusoidal gratings (0.5 c/deg, 200 cd/m2) of different contrast (not shown in figures). The steady state PERG decreased progressively by reducing the contrast and reliable responses could be recorded up to 5% contrast. The contrast threshold, obtained by extrapolating to 0 V the PERG amplitude as a function of log contrast was about 1%.
DISCUSSION As shown by our results, ERG responses to flash stimuli can be recorded from the eye of the little owl, Athene noctua, in conditions of both dark and light adaptation. Photopic and scotopic responses differ in their relative amplitude and intensity threshold (8 log units). These findings suggest that the little owl possesses a duplex retina in which rod and cone receptors exist in large enough number to contribute to the retinal responses in agreement with previous studies obtained in other owl species (Fite, 1973; 1.00 035 +s d ; L’)
1
a50 0.25
-8
-7
-6
-5
-4
LUMINANCE
-3
-2 (LOG
-1
0
1
2
3
cdlsqm)
Fig. 4. Relationship between the PERG high spatial frequency threshold (retinal acuity) and stimulus mean luminance. Note that the retinal acuity decreases by reducing the mean luminance. The slope of retinal acuity variation with mean luminance appears steeper in the photopic range. Contrast: 95%, alternation frequency: 2 Hz.
Owl ERG
Martin, 1974; Martin, 1977). Further indication for retinal duality comes from the doublebranched aspect of ERG FFF function, which suggests a larger contribution of scotopic components as compared to photopic ones. Differentiation between rod- and cone-driven activity by flicker electroretinography has been performed in different classes of animals. In this respect, the ERG FFF function of the little owl is similar to that of nocturnal animals like cats (Dodt, 1954) and differs from that of diurnal animals such as pigeons (Dodt & Wirth, 1953). In addition to ERG responses to luminance changes, ERGS to patterned stimuli presented at constant mean luminance can also be recorded from the little owl eye. The PERG, as the flash ERG, shows a consistent contribution of both scotopic and photopic components which seems to depend upon the mean luminance levels. Similarly, rods and cones contribute to the cat PERG whereas the primate PERG shows only photopic components (Hess, Baker, Zrenner & Schwarzer, 1986). As shown by our results the spatial tuning function, which shows low pass characteristics, shifts towards low spatial frequencies by decreasing the mean luminance level. Concurrently, retinal acuity decreases with an apparently steeper decay in the photopic range as compared to the scotopic one. Comparable acuity values have been determined in the photopic range both electrophysiologically at retinal level in pigeons (Bagnoli et al., 1984; Bagnoli, Porciatti, Fontanesi & Sebastiani, 1987) and behaviorally either in pigeons (Hodos, Leibowitz & Bonbright, 1976; Hodos, Bessette, Macko & Weiss, 1985) or owls (Fite, 1973; Martin & Gordon, 1974b). The values of retinal acuity measured in the scotopic range presumably permits a limited degree of pattern vision at low luminance levels. The luminance level beyond which no PERG can be recorded, is in the range of behaviorally determined absolute visual thresholds of other owl species (Fite, 1973; Martin, 1977) as well as of cats and man (Martin, 1982). The PERG contrast threshold corresponds to that determined in cats either behaviorally (Bisti & Maffei, 1974) or electrophysiologically (Campbell, Maffei & Piccolino, 1973). No similar data are available in owls at the behavioral level. In conclusion, the analysis of the owl ERG to flash and pattern stimuli permits to evaluate the relative contribution of photopic and scotopic
1697
components, and to establish the range of spatio-temporal frequencies, contrasts and luminances at which the retina is operating. Most of these electrophysiological data have their counterpart in behaviorally determined thresholds. Therefore, the ERG may represent a useful tool to predict owl visual behaviour under different environmental conditions. Electroretinography as compared to operant techniques, may have the advantage of an easier experimental approach which also allows to explore the suprathreshold parameter space. work was supported by a grant of the Consiglio Nazionale delle Ricerche (Ctb.87.029.3 1.04) and a grant from the Minister0 della Pubblica Istruzione (60% 1987). We thank A. Bertini, M. Vaghni and C. Pucci for excellent technical assistance and Dr V . P. Bingman for revising the manuscript. Acknowledgements-This
REFERENCES Bagnoli, P., Porciatti, V., Francesconi, W. & Barsellotti, R. (1984). Pigeon pattern electroretinogram: A response unaffected by chronic section of the optic nerve. ~xperimental Brain Research,
55, 253-262.
Bagnoli, P., Prociatti, V. & Francesconi, W. (1985). Retinal and tectal responses to alternating gratings are unaffected by monocular deprivation in pigeons. Brain Research, 338, 341-345.
Bagnoli, P., Porciatti, V., Fontanesi, G. & Sebastiani, L. (1987). Morphological and functional changes in the retinotectal system of the pigeon during the early posthatching period. Journal of Comparative Neurology, 256, 40&411. Bisti, S. & MatTei, L. (1974). Behavioural contrast sensitivity of the cat in various visual meridians. Journal of Physiology, Lona’on, 241, 201-210.
Bowmaker, J. K. & Martin, G. R. (1978). Visual pigments and colour vision in a nocturnal bird, Strix aluco (tawny owl). Brain Research, 18, 112~1130. Campbell, F. W., Maffei, L. & Piccolino, M. (1973). The contrast sensitivity of the cat. Journal of Physiology, London, 229, 719731.
Casini, G., Fontanesi, G., Porciatti, V. & Bagnoli, P. (1988). Anatomical and functional organization of the visual wulst in the owl Athene noctua. European Journal of Neuroscience, (Suppl.), 39, 12-15. Dodt, E. (1954). Ergebnisse der Flimmem Elektroretinographie. Experientia, 10, 330-338. Dodt, E. Jr Wirth, A. (1953). Differentiation between rods and cones by flicker electroretinography in pigeon and guinea pig. Acta Physiologica Scandinavia, 30, 80-89. Fite, K. V. (1973). Anatomical and behavioral correlates of visual acuity in the great horned owl. Vision Research, 13, 219-229.
Hess, R. F., Baker, C. L., Zrenner, E. & Schwarxer, J. (1986). Differences between electroretinograms of cat and primate. Journal of Neurophysiology, 56, 747-768. Hodos, W., Leibowitz, R. W. & Bonbright, J. C. Jr (1976). Near-field visual acuity of pigeons: Effect of head location and stimulus luminance. Journal of Experimental Analysts of Behavior, 25, 129-141.
1698
VITTORIO PORCIATTI et al.
Hodos, W., Bessette, B. B., Macko, K. A. &Weiss, S. R. B. (1985). Normative data for pigeon vision. Vision Research, 25, 1525-1527.
Martin, G. R. (1974). Color vision in the tawny owl (Srrix aluco). Journal of Comparative Physiological Psychology, 86, 133-142. Martin, G. R. (1977). Absolute visual threshold and scotopic spectral sensitivity in the tawny owl Strix aluco. Nature, London, 268, 636638.
Martin, G. R. (1982). The owl’s eye: Schematic optics and visual performance in Strix aluco L. Journal of Comparative Physiology, 145, 341-349.
Martin, G. R. & Gordon, I. E. (1974a). Increment threshold spectral sensitivity in the tawny owl (Strix aluco). Vision Research, 14, 615620.
Martin, G. R. & Gordon, 1. E. (1974b). Visual acuity in the tawny owl (Strix aluco). Vision Research, 14, 1393-1397. Martin, G. R., Gordon, I. E. & Cadee, D. R. (1975). Electroretinographically determined spectral sensitivity in the tawny owl (Strix aluco). Journal of Comparative Physiological Psychology, 89, 72-78.
Murphy, C. J. & Howland, H. C. (1983). Owl eyes: Accomodation, cornea1 curvature and refractive state. Journal of Comparative Physiology, 151, 277-284.
0042-6989/89$3.00+ 0.00 Copyright 0 1989Pergamon Press plc
THE ELECTRORETINOGRAM OF THE LITTLE (ATHENE NOCTUA) VI~~ORIO PORCIATTI, GIGLIOLA FONTANW
OWL
and PAOLA BAGNOLI
Department of Physiology and Biochemistry, University of Pisa and Institute of Neurophysiology, CNR, Via S. Zeno 31-51, Pisa 56100, Italy (Receioed 21 October
1988; in revised form 17 February 1989)
Abstract-Electroretinographic responses (ERGS) have been recorded from the cornea of the little owl (Athene noctua) in response to single light flashes and to alternating sinusoidal gratings (pattern) at different levels of light adaptation. Both flash- and pattern-evoked ERGS show scotopic as well as photopic components. The pattern evoked ERG is spatially tuned with tuning functions which shift towards lower frequencies by reducing the mean luminance. The retinal acuity is about 6c/deg at 2.3 logcd/m2 and decreases progressively by reducing the mean luminance. No pattern ERG can be recorded beyond -6.7 logcd/m* at any spatial frequency. The pattern ERG amplitude decreases progressively by reducing the contrast. The extrapolated contrast threshold is about 1%. Acuity and contrast sensitivity ERG values are in the range of those obtained by operant techniques in other species with duplex retinae such as owls and cats. Flash ERG
Pattern ERG
Little owl
Luminance levels
METHODS
INTRODUCTION
Among birds, owls possess eyes which are highly adapted for nocturnal viewing, with a well developed scotopic system (Fite, 1973). However, owl eyes also possess functional retinal cones numerous enough to mediate colour vision (Martin, 1974; Martin & Gordon, 1974a; Martin, Gordon & Cadee, 1975; Bowmaker & Martin, 1978). Among the Strigiformes, there are species that are strongly nocturnal, such as the tawny owl (Strix aluco). In contrast, the little owl (Athene noctua) which is common in Northern Italy and Sardinia, is primarily crepuscular with some diurnal activity. In the present study, the electroretinogram (ERG) of the little owl Athene noctua in response to light flashes under light and dark adaptation was recorded in order to evaluate the relative contribution of photopic and scotopic systems to retinal function. In addition, electroretinographic responses to sinusoidal gratings (Pattern ERGS, PERGs) of different contrast and spatial frequency were recorded under different luminance levels in order to determine the spatial domain at which the retina is operating.
Subjects Two little owls (Athene noctua) were used in this investigation. They were obtained from a local institution collecting protected birds with severe wing and leg damage. The owls underwent several experiments during the winter time at dusk. After each experiment, the two birds fully recovered from anaesthesia and were returned to their cages. Both animals were subsequently used in another investigation aimed at elucidating functional and anatomical properties of the geniculo-striate pathway (Casini, Fontanesi, Porciatti & Bagnoli, 1988). Electrophysiological techniques In the present study, the two animals (body weight 180 and 192 g, respectively) were anaesthetized with Chloral hydrate (0.2 ml/100 g body weight) injected into breast muscles. The anaesthetic dose was sufficient to render the birds unconscious during the 2 hr registration period. Owls were then mounted on a modified stereotaxis apparatus. The head was gently pressed against the beak holder by means of an elastic band. No pressure points were used. Pupils were dilated to a diameter of 6 mm by
1693
1694
VITTORIOPORCIATTIet al.
instillation of 1.O% cyclopentolate hydrochloride (Martin, Gordon & Cadee, 1975). The ERG was recorded simultaneously from both eyes by means of a hook-shaped gold lamina inserted into the lower fornix following instillation of oxibromocaine. A gentle downward traction of electrode wires served as a good substitute for blepharostat. The reference electrode was a subcutaneous stainless steel needle placed into the skin of the posterior part of the skull. A similar electrode placed in the upper part of the skull served as a ground. Visual stimulation
obtained by means of either retinoscopy or photorefraction (Murphy & Howland, 1983). Electrical signals were amplified (flash ERG : 10,000; PERG : 50,000), filtered (flash ERG; t-250 Hz; PERG: l-50 Hz, -6 dB/oct), sampled at different frequencies according to resolution and time base needed (0.5-5 KHz) and averaged (10 bit resolution, flash ERG: 20 events; PERG: 250 events). Averaged “noise” responses were obtained for any stimulus conditions by occluding the visual stimuli. Fourier analysis (resolution: 1 Hz) of the averaged responses was performed off line in order to establish the ERG flicker fusion (signal-to-noise ratio & 1.5).
Single light flashes were delivered by a strobe lamp (Energy: 5 J; flash duration*: 1.5 msec; RESULTS colour temperature: 5.9oO”K) placed inside a Ganzfeld bowl (45 cm dia.) at a rate of O.l/sec Feats ERG for the scotopic ERG or l/set for the photopic Figure IA, B shows examples of ERGS ERG. When measuring ERG flicker fusion frequency (FFF) the strobe lamp was driven with recorded in response to flashes of different a lower power (Energy: 0.2 J; flash duration*: intensities under light (A) and dark (B) adap175 psec; colour temperature: 5900°K) in such a tation. ERGS recorded simultaneously from way as its intensity was independent of flicker both eyes are shown superimposed. Since ERG rate in the range l-60 flasheslsec. Neutral den- waveforms were comparable in the two eyes at sity filters were placed in front of the strobe the different intensity levels, amplitude and lamp to attenuate flash intensity. Before record- latency values of a- and b-waves have been ings animals were kept 45 min in a darkened or averaged. As shown in Fig. lC, D the b-wave lightened (250 lx at the eye) room in order to amplitude as well as a- and b-wave latencies preadapt them to dark and light, respectively. change as a function of light intensity. It can be Recordings were obtained after either a 15 min seen that the b-wave increases in amplitude as period of dark adaptation or 3min period of the flash intensity is increased, under both dark light and light adaptation, and tends to saturate at light adaptation to background (100 cd/m*). A check was made that adap~tion the highest intensities (Fig. 1C). The difference periods to dark and light were sufficient to between the photopic and scotopic threshold is obtain maximal scotopic and photopic ERGS, of the order of 8 log units. As shown in Fig. lD, respectively. Vertical sinusoidal gratings of under light adaptation the latency of a- and different contrasts and spatial frequencies were b-waves tends to increase as the flash intensity generated by a microcomputer on a Joyce TV is decreased. In contrast, under dark adapraster display (white phosphor 200 cd/m2 mean tation, a striking increase in latency at low flash luminance, 20 deg field size at 57 cm distance). intensities for both a- and b-waves can be The stimulus display was surrounded by a large observed. white cardboard whose integrated luminance In Fig. 2, ERG flicker fusion frequency (FFF) was kept at the same level as the stimulus. as a function of flash intensity is shown. In the Refraction was measured objectively by placing range of low flash intensities the FFF progrestrials lens before the eyes and determining the sively increases as the flash intensity is increased trial lenses at which the PERG amplitude and tends to saturate at a value of about 20 (2c/deg, 95% contrast, 200cd/m’) was largest. flashes/set. Further increases in flash intensity Usually no lenses or + 1 sph. diopters were induce a rapid rise in FFF. At the highest necessary to obtain maximal PERG responses intensities the FFF again tends to saturate. This double branched plot is typical of duplex retinal. at 57 cm distance. This objective refraction agreed with previous measurements in owls Pattern ERG *Taken at l/3 of peak amplitude.
Figure 3A shows examples of PERGs in response to sinusoidal gratings (95% contrast)
Owl ERG
1695
PHOTOPIC
-8.19
+
. 1 -10
-9
-8
-7
-6
LOG RELATIVE
-8.59
D
+
-5 FLASH
-4
-3
-2
-1
0
INTENSITY
210
1
-1.38
7 4
-8.8g-bP+ -919
-269
w
.
110.
s -I -9.59
.
.
l l
8s l* .=..
-
oooooo
! 125 pi’
00
IO* -10
204msec
, -9
, -8
, -7
LOG RELATIVE
, -6
, -5 FLASH
. -4
%no 4
-3
-2
-1
0
INTENSITY
Fig. 1. Examples of ERGs recorded s~~taneously from both eyes in response to light Rashes of different relative intensity (indicated by numbers to the Ieft of each pair of tracings} under condition of light (A) and dark (B) adaptation (unattenuated flash energy 5 J, flash duration; 1.5 msec; colour temperature: 5900°K, background light: 100 cd/m*). Variation in amplitude (C) and peak latency (D) of the major ERG components (circles: b-wave; squares: a-wave) as a function of flash intensity under conditions of dark (solid symbols) and light (open symbols) adaptation.
of different spatial frequencies and mean luminances, reversed in their spatial phase (squarewave) at 2 Hz. At low spatial frequency and high luminance, the response consists of a small negative wave and a main positive complex
0
0
0
0 0
0
0 0
LOG
00
0
00
oooo
RELATIVE
FLASH
INTENSITY
Fig. 2. Relationship between ERG tlicker fusion frequency (FFF) and flash intensity. Note the double branched aspect of the plot which is typical of duplex retinae (unattenuated flash energy: 0.2 J, flash duration 17544~; colour temperature: 5900°K). VR 29/l2-D
which slowly returns to the base line. The main positive wave can be approximately divided in two sub-components (early and late) which display different characteristics as a function of spatial frequency and luminance. It can be noticed that both early and late PERG positive components progressively decrease in amplitude as a function of spatial frequency. However, the relative cont~bution of the two response components depends on mean luminance. At 200cd/m2 the BERG is dominated by the early component whereas at 20 and 2 cd/m* the PERG is dominated by the late component. These characteristics suggest that the early component is photopic in origin whereas the late component is scotopic. Further support for a scotopic origin of the late PERG component comes from the observation that this is not present in the PERG of diurnal birds (Bagnoli, Porciatti, Francesconi & Barsellotti, 1984; Bagnoli, Porciatti & Francesconi, 1985). As shown in Fig. 3B the amplitude of the photopic PERG component changes as a func-
VITTORIO PORCIATTI et al.
1696
s
200
,lfy
x
-.I------
zz
I~SJN 204msec
6.
5 Y y
I
0
0 2
0
-1.5
-1
2
00
---
-0.5
LOG
SPATIAL
0 --______ 20 0
0
_
FREQUENCY
(
---
0.5 (cldegl
1
Fig. 3. (A) Examplesof PERGs in response to sinusoidal gratings (95% contrast; 2 Hz alternation frequency) of differential spatial frequencies (indicated by numbers to the left of each tracing) and mean luminances (indicated by numbers to the top of each column). Note the complex waveform and the slow decay of the owl PERG. Also note that the early PERG positive component (photopic) decreases in amplitude by increasing the stimulus spatial frequency and reducing the mean luminance. The late PERG positive component (scotopic) dominates the response at low luminance levels (20 and 2cd/m2). (B) Amplitude variation of the photopic component of the PERG as a function of the stimulus spatial frequency and mean luminance (indicated by numbers close to each curve). Note that the spatial frequency function has low pass like characteristics and shifts to the left by reducing the mean luminance. Concurrently, the high spatial frequency threshold (retinal acuity) decreases. The dashed line represents the noise level.
tion of spatial frequency and mean luminance in the photopic range. At 200 and 20cd/m2 the PERG amplitude is approximately constant for spatial frequencies ranging between 0.125 c/deg and 1 c/deg. For spatial frequencies higher than 1 c/deg the PERG amplitude shows a marked decrease. At 2 cd/m2 the PERG monotonically decreases in amplitude as the spatial frequency increases. It can be noticed that the high spatial frequency threshold decreases by reducing the mean luminance. Threshold responses were considered those whose peak-to-trough amplitude was twice as large as the noise level. In Fig. 4 changes in high spatial frequency threshold (retinal acuity) as a function of mean luminance are reported. Retinal acuity progressively decreases by reducing the mean hnninance over 9 log units. In the photopic range, however, the reduction in retinal acuity is more marked than in the scotopic one. At luminances lower than - 6.7 log cd/m2 no response can be recorded at any spatial frequency. Steady-state (8.3 Hz) PERGs were also recorded in response to sinusoidal gratings (0.5 c/deg, 200 cd/m2) of different contrast (not shown in figures). The steady state PERG decreased progressively by reducing the contrast and reliable responses could be recorded up to 5% contrast. The contrast threshold, obtained by extrapolating to 0 V the PERG amplitude as a function of log contrast was about 1%.
DISCUSSION As shown by our results, ERG responses to flash stimuli can be recorded from the eye of the little owl, Athene noctua, in conditions of both dark and light adaptation. Photopic and scotopic responses differ in their relative amplitude and intensity threshold (8 log units). These findings suggest that the little owl possesses a duplex retina in which rod and cone receptors exist in large enough number to contribute to the retinal responses in agreement with previous studies obtained in other owl species (Fite, 1973; 1.00 035 +s d ; L’)
1
a50 0.25
-8
-7
-6
-5
-4
LUMINANCE
-3
-2 (LOG
-1
0
1
2
3
cdlsqm)
Fig. 4. Relationship between the PERG high spatial frequency threshold (retinal acuity) and stimulus mean luminance. Note that the retinal acuity decreases by reducing the mean luminance. The slope of retinal acuity variation with mean luminance appears steeper in the photopic range. Contrast: 95%, alternation frequency: 2 Hz.
Owl ERG
Martin, 1974; Martin, 1977). Further indication for retinal duality comes from the doublebranched aspect of ERG FFF function, which suggests a larger contribution of scotopic components as compared to photopic ones. Differentiation between rod- and cone-driven activity by flicker electroretinography has been performed in different classes of animals. In this respect, the ERG FFF function of the little owl is similar to that of nocturnal animals like cats (Dodt, 1954) and differs from that of diurnal animals such as pigeons (Dodt & Wirth, 1953). In addition to ERG responses to luminance changes, ERGS to patterned stimuli presented at constant mean luminance can also be recorded from the little owl eye. The PERG, as the flash ERG, shows a consistent contribution of both scotopic and photopic components which seems to depend upon the mean luminance levels. Similarly, rods and cones contribute to the cat PERG whereas the primate PERG shows only photopic components (Hess, Baker, Zrenner & Schwarzer, 1986). As shown by our results the spatial tuning function, which shows low pass characteristics, shifts towards low spatial frequencies by decreasing the mean luminance level. Concurrently, retinal acuity decreases with an apparently steeper decay in the photopic range as compared to the scotopic one. Comparable acuity values have been determined in the photopic range both electrophysiologically at retinal level in pigeons (Bagnoli et al., 1984; Bagnoli, Porciatti, Fontanesi & Sebastiani, 1987) and behaviorally either in pigeons (Hodos, Leibowitz & Bonbright, 1976; Hodos, Bessette, Macko & Weiss, 1985) or owls (Fite, 1973; Martin & Gordon, 1974b). The values of retinal acuity measured in the scotopic range presumably permits a limited degree of pattern vision at low luminance levels. The luminance level beyond which no PERG can be recorded, is in the range of behaviorally determined absolute visual thresholds of other owl species (Fite, 1973; Martin, 1977) as well as of cats and man (Martin, 1982). The PERG contrast threshold corresponds to that determined in cats either behaviorally (Bisti & Maffei, 1974) or electrophysiologically (Campbell, Maffei & Piccolino, 1973). No similar data are available in owls at the behavioral level. In conclusion, the analysis of the owl ERG to flash and pattern stimuli permits to evaluate the relative contribution of photopic and scotopic
1697
components, and to establish the range of spatio-temporal frequencies, contrasts and luminances at which the retina is operating. Most of these electrophysiological data have their counterpart in behaviorally determined thresholds. Therefore, the ERG may represent a useful tool to predict owl visual behaviour under different environmental conditions. Electroretinography as compared to operant techniques, may have the advantage of an easier experimental approach which also allows to explore the suprathreshold parameter space. work was supported by a grant of the Consiglio Nazionale delle Ricerche (Ctb.87.029.3 1.04) and a grant from the Minister0 della Pubblica Istruzione (60% 1987). We thank A. Bertini, M. Vaghni and C. Pucci for excellent technical assistance and Dr V . P. Bingman for revising the manuscript. Acknowledgements-This
REFERENCES Bagnoli, P., Porciatti, V., Francesconi, W. & Barsellotti, R. (1984). Pigeon pattern electroretinogram: A response unaffected by chronic section of the optic nerve. ~xperimental Brain Research,
55, 253-262.
Bagnoli, P., Prociatti, V. & Francesconi, W. (1985). Retinal and tectal responses to alternating gratings are unaffected by monocular deprivation in pigeons. Brain Research, 338, 341-345.
Bagnoli, P., Porciatti, V., Fontanesi, G. & Sebastiani, L. (1987). Morphological and functional changes in the retinotectal system of the pigeon during the early posthatching period. Journal of Comparative Neurology, 256, 40&411. Bisti, S. & MatTei, L. (1974). Behavioural contrast sensitivity of the cat in various visual meridians. Journal of Physiology, Lona’on, 241, 201-210.
Bowmaker, J. K. & Martin, G. R. (1978). Visual pigments and colour vision in a nocturnal bird, Strix aluco (tawny owl). Brain Research, 18, 112~1130. Campbell, F. W., Maffei, L. & Piccolino, M. (1973). The contrast sensitivity of the cat. Journal of Physiology, London, 229, 719731.
Casini, G., Fontanesi, G., Porciatti, V. & Bagnoli, P. (1988). Anatomical and functional organization of the visual wulst in the owl Athene noctua. European Journal of Neuroscience, (Suppl.), 39, 12-15. Dodt, E. (1954). Ergebnisse der Flimmem Elektroretinographie. Experientia, 10, 330-338. Dodt, E. Jr Wirth, A. (1953). Differentiation between rods and cones by flicker electroretinography in pigeon and guinea pig. Acta Physiologica Scandinavia, 30, 80-89. Fite, K. V. (1973). Anatomical and behavioral correlates of visual acuity in the great horned owl. Vision Research, 13, 219-229.
Hess, R. F., Baker, C. L., Zrenner, E. & Schwarxer, J. (1986). Differences between electroretinograms of cat and primate. Journal of Neurophysiology, 56, 747-768. Hodos, W., Leibowitz, R. W. & Bonbright, J. C. Jr (1976). Near-field visual acuity of pigeons: Effect of head location and stimulus luminance. Journal of Experimental Analysts of Behavior, 25, 129-141.
1698
VITTORIO PORCIATTI et al.
Hodos, W., Bessette, B. B., Macko, K. A. &Weiss, S. R. B. (1985). Normative data for pigeon vision. Vision Research, 25, 1525-1527.
Martin, G. R. (1974). Color vision in the tawny owl (Srrix aluco). Journal of Comparative Physiological Psychology, 86, 133-142. Martin, G. R. (1977). Absolute visual threshold and scotopic spectral sensitivity in the tawny owl Strix aluco. Nature, London, 268, 636638.
Martin, G. R. (1982). The owl’s eye: Schematic optics and visual performance in Strix aluco L. Journal of Comparative Physiology, 145, 341-349.
Martin, G. R. & Gordon, I. E. (1974a). Increment threshold spectral sensitivity in the tawny owl (Strix aluco). Vision Research, 14, 615620.
Martin, G. R. & Gordon, 1. E. (1974b). Visual acuity in the tawny owl (Strix aluco). Vision Research, 14, 1393-1397. Martin, G. R., Gordon, I. E. & Cadee, D. R. (1975). Electroretinographically determined spectral sensitivity in the tawny owl (Strix aluco). Journal of Comparative Physiological Psychology, 89, 72-78.
Murphy, C. J. & Howland, H. C. (1983). Owl eyes: Accomodation, cornea1 curvature and refractive state. Journal of Comparative Physiology, 151, 277-284.