- Email: [email protected]
The mechanics of pigeon eye movements — Are they like other vertebrates?
Behavioural Brain Research, 19 (1986) 117-121 Elsevier
117
BBR 00536
THE M E C H A N I C S OF PIGEON EYE M O V E M E N T S - ARE THEY LIKE OTHER VERTEBRATES?
ALISTAIR M c V E A N and JOHN STELLING
Department of Zoology, Royal Holloway and Bedford New College, University of London, Englefield Green, Surrey TW20 9TY (U.K.) (Received March 20th, 1985) (Revised version received November 25th, 1985) (Accepted November 26th, 1985)
Key words: pigeon - eye movement - characteristic frequency
The lateral rectus muscle of the pigeon eye was driven by stimulation of the abducens nucleus. Eye rotation was measured with an opto-electronic movement detector. Eye position was linearly related to stimulation frequency in the range 40-110 Hz and saturated at frequencies above 250 Hz. Maximum eye velocity of 240°/s was obtained with a stimulation frequency of 360 Hz. Stimulation with sinusoidally modulated pulse frequencies (40-110 Hz) over the modulation frequency range 0.01-6.0 Hz were used to determine the gain and phase relationships of the oculomotor system. The response approximates a first order low pass frequency model with a characteristic frequency of 0.45 Hz at high frequencies. There is an additional phase lag equivalent to a time delay of 9.8 ms. The results are compared with similar experiments performed on the dogfish and cat oculomotor systems.
INTRODUCTION
The mechanics of eye movements have been comprehensively studied in mammals 3,18, but much less extensively in non-mammalian vertebrates. Studies of eye movements have been extended to teleosts 4"5'8"13-15 and elasmobranchs 9'12. These have involved descriptions of the intact oculomotor system as well as isolated parts of the system. Recently, eye movement in pigeons has come under scrutiny2"11.16 to determine the quality of eye movement present in animals with lateral vision and highly mobile heads. Bloch et al. 1reported the presence of conjugate eye movements in response to objects appearing in the frontal visual field and independent eye movements stimulated by objects in the lateral viewing fields. Amplitudes of up to 17 ° were recorded using EOG. Somewhat earlier Nye ~6 had described eye movements of a very different
nature. Using lightly anaesthetized birds and an optical lever, Nye showed that the eyes were subject to short bursts of oscillations at a frequency of 30 Hz, peak to peak amplitude of 5 °, separated by intervals of about 5 s. Also present were short impulse increments with no associated oscillation. The impulses appeared to be made under the restraint of an overdamped control system while the oscillations suggested an impulse response of an inertial load suspended from a system of passive, underdamped elastic muscles. This explanation proved to be too simple, because in some instances the eye could continue oscillating after a momentary interruption and at other times the oscillations grew rapidly in amplitude. The maximum frequency at which eye movements of comparable amplitude can be made in man is no more than 2 Hz. (ref. 6). N y e 16 suggested that the combination of underdamped oscillations and overdamped impulses could be
Correspondence: A. McVean, Department of Zoology, Royal Holloway and Bedford New College, University of London, Bakeham Lane, Englefield Green, Surrey TW20 9TY, U.K. 0166-4328/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)
118 achieved either by variable but controlled damping, or the eye muscles must be capable of exerting a flexible variety of force patterns upon a permanently overdamped system. We decided to investigate the damping characteristics of the pigeon eye using a technique previously employed by Zuber 19, and Reinhart and Zuber 17 on cats and Montgomery 12 on dogfish, in which direct stimulation of oculomotor nerves or motomeurons combined with systems analysis has provided a mathematical description of the dynamic characteristics of the oculomotor plant. The above experiments confirmed that in the cat and dogfish the dynamics approximate to that of a first order, low pass filter model incorporating a fixed time delay. Considerable damping is present. That is not to say that the components of the cat and dogfish plant were similar in detail. Thus eye rotation in the dogfish was linearly related to stimulation frequency up to 20 Hz, whereas in the cat linearity holds up to 150 Hz. The gain of the system is much higher in the cat and the phase lag less than in the dogfish (see Fig. 3). The better performance of the cat oculomotor plant, driven in both species by the lateral rectus muscle, is reflected by a characteristic or comer frequency of 3.5 H z t7 compared with 0.23 Hz in the dogfish. We have examined the pigeon oculomotor system using the same method employed by Reinhart and Zuber 17 to provide a direct comparison with the mechanics of the cat and dogfish eye. MATERIALS AND METHODS
Five adult pigeons (Columbia livia) of both sexes, 300-500g, were initially anaesthetized with Saffan (30 mg/kg, Glaxovet). Anaesthesia was maintained with Urethane (1 g/kg) at a level where eyelid and corneal reflexes were absent. Both anaesthetics were administered intramuscularly. The upper and lower eyelids were sewn open and the nictitating membrane removed. Xylocaine (0.5 ~ solution) was applied to the surface of the eye, injected beneath the skin around the eyelids and around the ears. The pigeon was then placed in a stereotaxic head holder with the beak secured to a bar passing between the
mandibles. The calvarium was exposed by a longitudinal cut and xylocaine applied to the exposed surfaces. A 4-mm diameter hole was drilled through the skull directly above the abducens nucleus 1°, the dura mater removed from the region underlying the hole and any accompanying bleeding controlled by cautery. The temperature of the bird was maintained at 41 ° C. A concentric bipolar stainless steel electrode (Rhodes SNE- 100, electrode separation 0.5-1 mm) was lowered through the brain until a trial stimulus train elicited eye movements in a posterior (temporal) direction. Electrical stimuli were in the form of trains of square wave pulses (1 ms, 30-100/aA) delivered at a constant frequency in the range 0-300 Hz or sinusoidally modulated between 40-110 Hz at modulation frequencies in the range 0.01-6 Hz. After each experiment the position of the electrode in the nucleus abducens was confirmed histologically. Horizontal eye movements were recorded by an opto-electronic movement detector9a which provides a voltage whose magnitude is linearly related to the position of an illuminated patch of 'Scotchlite' reflective tape, 2 x 2 mm, attached to the cornea. Care was taken when fLxingthe reflective patch to the eye, so that it did not impede eye movements. The weight of the patch was less than 0.05 ~o of the weight of the eyeball. The movement detector was calibrated at the beginning of each experiment. The linear movement of the patch was converted into eye angular rotation using 0 = sin/(A/r), r being the distance between the cornea and the centre of rotation, estimated at 4.6 mm. Movements of less than 0.1 ° could be discerned. Pulse frequency and eye movements were monitored on a digital storage oscilloscope (Gould OS 1420) and stored on a Racal instrument recorder. Data obtained from 3 preparations was plotted from the recorder onto an xy plotter (Gould Bryans 60000). RESULTS
The relation between stimulation frequency and eye deviation away from the rest position was determined by pulse trains delivered at fixed frequencies (Fig. 1). To determine the current ampli-
119 10 Fm=O
01~4z
& -- 2'0
Zo L
J nII
.
'~15
2o~/."
.~/
. . . . . . . . . . 20 4.0 60 80
/
LIJ
/ ~
",, -~ 100
120
10
12
. . . . 140 160
-,o.
o Frequency
2;0
3;0
Hz)
Fig. 1. Frequency response curve. Maximum stable eye rotation induced by constant frequency pulse trains in the range 0-300 Hz. The response is linear in the 40-110 Hz range saturating at higher frequencies.
-
2.5.
>-
.~
2
/,
0-2
04
6
g
14
16
'f'2 ' t 4
16
•
a.°"'3'0'.
tude used in the frequency response measurements, eye deviation was plotted against current intensity. The current intensity adopted was that which caused maximum eye deviation where the relationship between eye deviation and current intensity remained linear. Each pulse train of a given frequency was maintained until the eye had reached a stable position. At 110 Hz this time was 0.3-0.8 s. Between 40 and 110 Hz the relationship between stimulus frequency and eye deviation from the rest position was approximately linear; above 110 Hz the response increment declined, saturating at frequencies above 300 Hz. Within the linear region, low frequency gain was 0.06°/pulse/s. There was a fixed time delay of 9.8 ms between onset of a new level of stimulation and the beginning of eye movement. With sinusoidally modulated pulse trains (40-110 Hz) the eye deviated from its primary position to oscillate around a position determined by the mean frequency. At low modulating frequencies (0.01 Hz) the sinusoidal stimulus trains elicited sinusoidal eye movements whose gain was assigned an arbitrary value of 1. As modulation frequency increased (Fig. 2) the gain declined and the phase lag between stimulation waveform and eye position, already evident at 0.01 Hz, increased. The Bode plot for gain and phase characteristics of the system are shown in Fig. 3. In Fig. 3A the solid line represents the best-fit firstorder model for gain with a characteristic or
2.5
2-o~ 06 08 1"0 TINE ~sec0nds)
Fig. 2. Records of eye movement in response to sinusoidally modulated pulse trains at 3 different frequencies. In each case pulse frequencies are from 64-104 Hz. Note the low frequency lag at 0.01 Hz.
corner frequency of 0.45 Hz. The data for dogfish ]4 and cat 17 are also given in Fig. 3 for comparison with the pigeon. The pigeon eye did not follow modulation frequencies above 6 Hz. From the above results the transfer function of the system 17 can be given as
G(S) =
0.06 (deg/pps)e- 0.0098 s 1 + 0.35 s
which will be valid for frequencies between 40 and 110 Hz and modulation frequencies in the range 0.01-6.0 Hz. DISCUSSION
The results and model presented in this study are essentially the same as those for the cat 17 and the dogfish ]2. In all 3 preparations the eye was
120 1"0:
• I
,., ; :
........
-.
0-5
o fi, h - .
"..
0,1 I
. . . . . .
'I
phase contributionof delay, pigeon . . . . . . . . : : :.....]................../ "~ .cat, to '["phase Lag
o _
I
30
oo,,o.
[
%."
",,
pig+on,
~t~? phase[ a/'g " " " . . . ' ~ ~ t h e ° r e f i c a l
1
pigeen f0fat
ase ',
150 t
•
\\\
",
1801 B
001
'
'
'I
01
'
'
' '
'"I
'
10 Frequency ( Hz}
'
"'F"I
10
'
'
20
Fig. 3. Bode plots of gain and phase response for frequencies in the range 0.01-6.0 Hz. Each dot represents the mean of 10 measurements from one bird at a particular modulation frequency. Ten min were allowed to elapse between trials. A: log gain vs log modulation frequency. The solid line represents the best fit first order model. B: phase lag vs log modulation frequency. The upper solid line represents the best fit theoretical first order model while the lower solid line includes the phase lag caused by the fixed time delay. Gain and phase data from Reinhart and Zuber ~7 for the cat and from Montgomery ~2 for the dogfish are also plotted. See text for details.
driven by contraction of the lateral rectus muscle, though Montgomery stimulated the abducens nerve directly with suction electrodes applied to the cut end of the abducens nerve, while the pigeon and cat lateral rectus muscle was driven by brainstem stimulation. The transfer characteristics in each case approximate to a first order low pass filter system with a fixed time delay, though all 3 species fail to conform well to the model at low modulation frequencies, a disparity attributed to the presence of slow muscle fibres by Montgomery 12. In all 3 species there is a range on the stimulus response curve in which eyeball rotation is approximately linear with respect to stimulation frequency. This region occurs at a much lower frequency range in the dogfish (0-20 Hz) than in the
cat (50-150 Hz) or the pigeon (40-110 Hz). Saturation frequencies show a similar trend. Eye rotation saturates above 40 Hz in the dogfish; in the cat, eye rotation continues up to frequencies of 200 Hz and probably beyond 19 while the pigeon eye reaches maximum rotation at a stimulation frequency in excess of 300 Hz. All 3 species exhibit significant fixed delays between stimulation and the onset of eye movement. The delay in the pigeon (9.8 ms) is slightly less than in the cat (12 ms); both are significantly quicker to respond than the dogfish where the delay is 50 ms. In the dogfish this long time delay probably has little influence on a system which may not operate at modulation frequencies above 1 Hz 12. The latency between stimulation and contraction, frequency range over which rotation is linear and saturating frequencies in the pigeon are similar to those of the cat. When the gain and phase delay of the pigeon eye at different modulation frequencies is compared with that of the dogfish and cat, the pigeon is found to have an intermediate performance. The pigeon's eye ceases to follow modulation frequencies greater than 6 Hz; below this the phase lag for the pigeon is intermediate between the dogfish and the cat while the characteristic frequency (0.45 Hz) is much closer to the dogfish (0.23 Hz) than the cat (3.5 Hz). Pigeons, like other small to intermediate sized birds, move their heads in a manner that is reminiscent of mammalian saccades and have been shown to follow striped drums with smooth pursuit movements 7. The need for rapid, nontremor eye movements might be doubted, but Bloch et al. 2 and Martinoya et al. 11 have clearly demonstrated the presence of organised eye movements in pigeons which appear to be coordinated with redirected visual attention. The results from this study suggest that eye movements in pigeons are mediated through an ocular motor plant that is essentially similar to that found in the dogfish and cat and which can be represented by a first order low pass filter model containing a significant amount of damping. The underdamping which might underlie the rapid eye oscillations measured by N y e 16 w a s not revealed in these experiments.
121 REFERENCES 1 Bloch, S., Rivaud, S. and Martinoya, C., Comparing frontal and lateral viewing in the pigeon. III. Different patterns of eye movements for binocular and monocular fixation, Behav. Brain Res., 13 (1984) 173-182. 2 Bloch, S., Martinoya, C. and Rivaud, S., Eye movements in pigeons: participation in binocular fixation and visual pursuit, J. Physiol. (London), 320 (1981) 20-21. 3 Carpenter, R.H.S., Movements of the Eyes, Pion Press, London, 1977. 4 Easter, S.S., Spontaneous eye movements in restrained goldfish, Vision Res., II (1971) 333-342. 5 Easter, S.S., Pursuit eye movements in goldfish (Carrassius auratus), Vision Res., 12 (1972) 673-688. 6 Fender, D.H. and Nye, P.W., An investigation of the mechanisms of eye movement control, Kybernetik, 1 (1961) 81-88. 7 Fite, K.V., Optokinetic nystagmus and the pigeon visual system. In A.M. Granada and J.H. Maxwell (Eds.), Neural Mechanisms of Behaviour in the Pigeon, Plenum Press, New York, 1978, pp. 395-407. 8 Gestrin, P. and Sterling, P., Anatomy and physiology of goldfish oculomotor system. II. Firing patterns of neurons in abducens nucleus and surrounding medulla and their relation to eye movements, J. Neurophysiol., 40 (1977) 573-588. 9 Harris, A.J., Eye movements of the dogfish Squalus acanthias L, J. Exp. Biol., 43 (1965) 107-130. 9a Helversen, O.N. and Eisner, N., The stridulatory movements of acridid grasshoppers recorded with an optoelectronic device, J. Comp. Physiol., 122 (1977) 53-64.
10 Karten, H.J. and Hodos, W., A Sterotaxic Atlas of the Brain of the Pigeon, (Columba livia), John Hopkins Press, Baltimore, 1967. 11 Martinoya, C., Houezec, and Bloch, S., Pigeon's eyes converge during feeding: evidence for frontal binocular fixation in a lateral-eyed bird, Neurosci. Lett., 45 (1984) 335-339. 12 Montgomery, J.C., Eye movement dynamics in the dogfish, J. Exp. BioL, 105 (1983) 297-303. 13 Montgomery, J.C. and Macdonald, J.A., Performance of motor systems in antarctic fishes, J. Comp. Physiol., A. 154 (1984) 241-248. 14 Montgomery, J.C., McVean, A.R. and McCarthy, D., The effects of lowered temperature on spontaneous eye movements in a teleost fish, Comp. Biochem. Physiol., 75 A (1983) 363-368. 15 Northmore, D.P.M., Williams, B. and Vanegas, H., The teleostean torus longitudinalis: responses related to eye movements visuotopic mapping, and functional relations with the optic tectum, J. Comp. Physiol., 150 (1983) 39-50. 16 Nye, P.W., The monocular eye movements of the pigeon, Vision Res., 9 (1969) 133-144. 17 Reinhart, R.J. and Zuber, B.L., Horizontal eye movements from abducens nerve stimulation in the cat, IEEE Trans. Biomed. Eng., BME-17 (I) (1970) 11-14. 18 Robinson, D.A., The use of control systems analysis in the neurophysiology of eye movements, Ann Rev. Neurosci., 4 (1981) 463-503. 19 Zuber, B.L., Sinusoidal eye movements from brain stimulation in the cat, 8 (1968) 1073-1079.
117
BBR 00536
THE M E C H A N I C S OF PIGEON EYE M O V E M E N T S - ARE THEY LIKE OTHER VERTEBRATES?
ALISTAIR M c V E A N and JOHN STELLING
Department of Zoology, Royal Holloway and Bedford New College, University of London, Englefield Green, Surrey TW20 9TY (U.K.) (Received March 20th, 1985) (Revised version received November 25th, 1985) (Accepted November 26th, 1985)
Key words: pigeon - eye movement - characteristic frequency
The lateral rectus muscle of the pigeon eye was driven by stimulation of the abducens nucleus. Eye rotation was measured with an opto-electronic movement detector. Eye position was linearly related to stimulation frequency in the range 40-110 Hz and saturated at frequencies above 250 Hz. Maximum eye velocity of 240°/s was obtained with a stimulation frequency of 360 Hz. Stimulation with sinusoidally modulated pulse frequencies (40-110 Hz) over the modulation frequency range 0.01-6.0 Hz were used to determine the gain and phase relationships of the oculomotor system. The response approximates a first order low pass frequency model with a characteristic frequency of 0.45 Hz at high frequencies. There is an additional phase lag equivalent to a time delay of 9.8 ms. The results are compared with similar experiments performed on the dogfish and cat oculomotor systems.
INTRODUCTION
The mechanics of eye movements have been comprehensively studied in mammals 3,18, but much less extensively in non-mammalian vertebrates. Studies of eye movements have been extended to teleosts 4"5'8"13-15 and elasmobranchs 9'12. These have involved descriptions of the intact oculomotor system as well as isolated parts of the system. Recently, eye movement in pigeons has come under scrutiny2"11.16 to determine the quality of eye movement present in animals with lateral vision and highly mobile heads. Bloch et al. 1reported the presence of conjugate eye movements in response to objects appearing in the frontal visual field and independent eye movements stimulated by objects in the lateral viewing fields. Amplitudes of up to 17 ° were recorded using EOG. Somewhat earlier Nye ~6 had described eye movements of a very different
nature. Using lightly anaesthetized birds and an optical lever, Nye showed that the eyes were subject to short bursts of oscillations at a frequency of 30 Hz, peak to peak amplitude of 5 °, separated by intervals of about 5 s. Also present were short impulse increments with no associated oscillation. The impulses appeared to be made under the restraint of an overdamped control system while the oscillations suggested an impulse response of an inertial load suspended from a system of passive, underdamped elastic muscles. This explanation proved to be too simple, because in some instances the eye could continue oscillating after a momentary interruption and at other times the oscillations grew rapidly in amplitude. The maximum frequency at which eye movements of comparable amplitude can be made in man is no more than 2 Hz. (ref. 6). N y e 16 suggested that the combination of underdamped oscillations and overdamped impulses could be
Correspondence: A. McVean, Department of Zoology, Royal Holloway and Bedford New College, University of London, Bakeham Lane, Englefield Green, Surrey TW20 9TY, U.K. 0166-4328/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)
118 achieved either by variable but controlled damping, or the eye muscles must be capable of exerting a flexible variety of force patterns upon a permanently overdamped system. We decided to investigate the damping characteristics of the pigeon eye using a technique previously employed by Zuber 19, and Reinhart and Zuber 17 on cats and Montgomery 12 on dogfish, in which direct stimulation of oculomotor nerves or motomeurons combined with systems analysis has provided a mathematical description of the dynamic characteristics of the oculomotor plant. The above experiments confirmed that in the cat and dogfish the dynamics approximate to that of a first order, low pass filter model incorporating a fixed time delay. Considerable damping is present. That is not to say that the components of the cat and dogfish plant were similar in detail. Thus eye rotation in the dogfish was linearly related to stimulation frequency up to 20 Hz, whereas in the cat linearity holds up to 150 Hz. The gain of the system is much higher in the cat and the phase lag less than in the dogfish (see Fig. 3). The better performance of the cat oculomotor plant, driven in both species by the lateral rectus muscle, is reflected by a characteristic or comer frequency of 3.5 H z t7 compared with 0.23 Hz in the dogfish. We have examined the pigeon oculomotor system using the same method employed by Reinhart and Zuber 17 to provide a direct comparison with the mechanics of the cat and dogfish eye. MATERIALS AND METHODS
Five adult pigeons (Columbia livia) of both sexes, 300-500g, were initially anaesthetized with Saffan (30 mg/kg, Glaxovet). Anaesthesia was maintained with Urethane (1 g/kg) at a level where eyelid and corneal reflexes were absent. Both anaesthetics were administered intramuscularly. The upper and lower eyelids were sewn open and the nictitating membrane removed. Xylocaine (0.5 ~ solution) was applied to the surface of the eye, injected beneath the skin around the eyelids and around the ears. The pigeon was then placed in a stereotaxic head holder with the beak secured to a bar passing between the
mandibles. The calvarium was exposed by a longitudinal cut and xylocaine applied to the exposed surfaces. A 4-mm diameter hole was drilled through the skull directly above the abducens nucleus 1°, the dura mater removed from the region underlying the hole and any accompanying bleeding controlled by cautery. The temperature of the bird was maintained at 41 ° C. A concentric bipolar stainless steel electrode (Rhodes SNE- 100, electrode separation 0.5-1 mm) was lowered through the brain until a trial stimulus train elicited eye movements in a posterior (temporal) direction. Electrical stimuli were in the form of trains of square wave pulses (1 ms, 30-100/aA) delivered at a constant frequency in the range 0-300 Hz or sinusoidally modulated between 40-110 Hz at modulation frequencies in the range 0.01-6 Hz. After each experiment the position of the electrode in the nucleus abducens was confirmed histologically. Horizontal eye movements were recorded by an opto-electronic movement detector9a which provides a voltage whose magnitude is linearly related to the position of an illuminated patch of 'Scotchlite' reflective tape, 2 x 2 mm, attached to the cornea. Care was taken when fLxingthe reflective patch to the eye, so that it did not impede eye movements. The weight of the patch was less than 0.05 ~o of the weight of the eyeball. The movement detector was calibrated at the beginning of each experiment. The linear movement of the patch was converted into eye angular rotation using 0 = sin/(A/r), r being the distance between the cornea and the centre of rotation, estimated at 4.6 mm. Movements of less than 0.1 ° could be discerned. Pulse frequency and eye movements were monitored on a digital storage oscilloscope (Gould OS 1420) and stored on a Racal instrument recorder. Data obtained from 3 preparations was plotted from the recorder onto an xy plotter (Gould Bryans 60000). RESULTS
The relation between stimulation frequency and eye deviation away from the rest position was determined by pulse trains delivered at fixed frequencies (Fig. 1). To determine the current ampli-
119 10 Fm=O
01~4z
& -- 2'0
Zo L
J nII
.
'~15
2o~/."
.~/
. . . . . . . . . . 20 4.0 60 80
/
LIJ
/ ~
",, -~ 100
120
10
12
. . . . 140 160
-,o.
o Frequency
2;0
3;0
Hz)
Fig. 1. Frequency response curve. Maximum stable eye rotation induced by constant frequency pulse trains in the range 0-300 Hz. The response is linear in the 40-110 Hz range saturating at higher frequencies.
-
2.5.
>-
.~
2
/,
0-2
04
6
g
14
16
'f'2 ' t 4
16
•
a.°"'3'0'.
tude used in the frequency response measurements, eye deviation was plotted against current intensity. The current intensity adopted was that which caused maximum eye deviation where the relationship between eye deviation and current intensity remained linear. Each pulse train of a given frequency was maintained until the eye had reached a stable position. At 110 Hz this time was 0.3-0.8 s. Between 40 and 110 Hz the relationship between stimulus frequency and eye deviation from the rest position was approximately linear; above 110 Hz the response increment declined, saturating at frequencies above 300 Hz. Within the linear region, low frequency gain was 0.06°/pulse/s. There was a fixed time delay of 9.8 ms between onset of a new level of stimulation and the beginning of eye movement. With sinusoidally modulated pulse trains (40-110 Hz) the eye deviated from its primary position to oscillate around a position determined by the mean frequency. At low modulating frequencies (0.01 Hz) the sinusoidal stimulus trains elicited sinusoidal eye movements whose gain was assigned an arbitrary value of 1. As modulation frequency increased (Fig. 2) the gain declined and the phase lag between stimulation waveform and eye position, already evident at 0.01 Hz, increased. The Bode plot for gain and phase characteristics of the system are shown in Fig. 3. In Fig. 3A the solid line represents the best-fit firstorder model for gain with a characteristic or
2.5
2-o~ 06 08 1"0 TINE ~sec0nds)
Fig. 2. Records of eye movement in response to sinusoidally modulated pulse trains at 3 different frequencies. In each case pulse frequencies are from 64-104 Hz. Note the low frequency lag at 0.01 Hz.
corner frequency of 0.45 Hz. The data for dogfish ]4 and cat 17 are also given in Fig. 3 for comparison with the pigeon. The pigeon eye did not follow modulation frequencies above 6 Hz. From the above results the transfer function of the system 17 can be given as
G(S) =
0.06 (deg/pps)e- 0.0098 s 1 + 0.35 s
which will be valid for frequencies between 40 and 110 Hz and modulation frequencies in the range 0.01-6.0 Hz. DISCUSSION
The results and model presented in this study are essentially the same as those for the cat 17 and the dogfish ]2. In all 3 preparations the eye was
120 1"0:
• I
,., ; :
........
-.
0-5
o fi, h - .
"..
0,1 I
. . . . . .
'I
phase contributionof delay, pigeon . . . . . . . . : : :.....]................../ "~ .cat, to '["phase Lag
o _
I
30
oo,,o.
[
%."
",,
pig+on,
~t~? phase[ a/'g " " " . . . ' ~ ~ t h e ° r e f i c a l
1
pigeen f0fat
ase ',
150 t
•
\\\
",
1801 B
001
'
'
'I
01
'
'
' '
'"I
'
10 Frequency ( Hz}
'
"'F"I
10
'
'
20
Fig. 3. Bode plots of gain and phase response for frequencies in the range 0.01-6.0 Hz. Each dot represents the mean of 10 measurements from one bird at a particular modulation frequency. Ten min were allowed to elapse between trials. A: log gain vs log modulation frequency. The solid line represents the best fit first order model. B: phase lag vs log modulation frequency. The upper solid line represents the best fit theoretical first order model while the lower solid line includes the phase lag caused by the fixed time delay. Gain and phase data from Reinhart and Zuber ~7 for the cat and from Montgomery ~2 for the dogfish are also plotted. See text for details.
driven by contraction of the lateral rectus muscle, though Montgomery stimulated the abducens nerve directly with suction electrodes applied to the cut end of the abducens nerve, while the pigeon and cat lateral rectus muscle was driven by brainstem stimulation. The transfer characteristics in each case approximate to a first order low pass filter system with a fixed time delay, though all 3 species fail to conform well to the model at low modulation frequencies, a disparity attributed to the presence of slow muscle fibres by Montgomery 12. In all 3 species there is a range on the stimulus response curve in which eyeball rotation is approximately linear with respect to stimulation frequency. This region occurs at a much lower frequency range in the dogfish (0-20 Hz) than in the
cat (50-150 Hz) or the pigeon (40-110 Hz). Saturation frequencies show a similar trend. Eye rotation saturates above 40 Hz in the dogfish; in the cat, eye rotation continues up to frequencies of 200 Hz and probably beyond 19 while the pigeon eye reaches maximum rotation at a stimulation frequency in excess of 300 Hz. All 3 species exhibit significant fixed delays between stimulation and the onset of eye movement. The delay in the pigeon (9.8 ms) is slightly less than in the cat (12 ms); both are significantly quicker to respond than the dogfish where the delay is 50 ms. In the dogfish this long time delay probably has little influence on a system which may not operate at modulation frequencies above 1 Hz 12. The latency between stimulation and contraction, frequency range over which rotation is linear and saturating frequencies in the pigeon are similar to those of the cat. When the gain and phase delay of the pigeon eye at different modulation frequencies is compared with that of the dogfish and cat, the pigeon is found to have an intermediate performance. The pigeon's eye ceases to follow modulation frequencies greater than 6 Hz; below this the phase lag for the pigeon is intermediate between the dogfish and the cat while the characteristic frequency (0.45 Hz) is much closer to the dogfish (0.23 Hz) than the cat (3.5 Hz). Pigeons, like other small to intermediate sized birds, move their heads in a manner that is reminiscent of mammalian saccades and have been shown to follow striped drums with smooth pursuit movements 7. The need for rapid, nontremor eye movements might be doubted, but Bloch et al. 2 and Martinoya et al. 11 have clearly demonstrated the presence of organised eye movements in pigeons which appear to be coordinated with redirected visual attention. The results from this study suggest that eye movements in pigeons are mediated through an ocular motor plant that is essentially similar to that found in the dogfish and cat and which can be represented by a first order low pass filter model containing a significant amount of damping. The underdamping which might underlie the rapid eye oscillations measured by N y e 16 w a s not revealed in these experiments.
121 REFERENCES 1 Bloch, S., Rivaud, S. and Martinoya, C., Comparing frontal and lateral viewing in the pigeon. III. Different patterns of eye movements for binocular and monocular fixation, Behav. Brain Res., 13 (1984) 173-182. 2 Bloch, S., Martinoya, C. and Rivaud, S., Eye movements in pigeons: participation in binocular fixation and visual pursuit, J. Physiol. (London), 320 (1981) 20-21. 3 Carpenter, R.H.S., Movements of the Eyes, Pion Press, London, 1977. 4 Easter, S.S., Spontaneous eye movements in restrained goldfish, Vision Res., II (1971) 333-342. 5 Easter, S.S., Pursuit eye movements in goldfish (Carrassius auratus), Vision Res., 12 (1972) 673-688. 6 Fender, D.H. and Nye, P.W., An investigation of the mechanisms of eye movement control, Kybernetik, 1 (1961) 81-88. 7 Fite, K.V., Optokinetic nystagmus and the pigeon visual system. In A.M. Granada and J.H. Maxwell (Eds.), Neural Mechanisms of Behaviour in the Pigeon, Plenum Press, New York, 1978, pp. 395-407. 8 Gestrin, P. and Sterling, P., Anatomy and physiology of goldfish oculomotor system. II. Firing patterns of neurons in abducens nucleus and surrounding medulla and their relation to eye movements, J. Neurophysiol., 40 (1977) 573-588. 9 Harris, A.J., Eye movements of the dogfish Squalus acanthias L, J. Exp. Biol., 43 (1965) 107-130. 9a Helversen, O.N. and Eisner, N., The stridulatory movements of acridid grasshoppers recorded with an optoelectronic device, J. Comp. Physiol., 122 (1977) 53-64.
10 Karten, H.J. and Hodos, W., A Sterotaxic Atlas of the Brain of the Pigeon, (Columba livia), John Hopkins Press, Baltimore, 1967. 11 Martinoya, C., Houezec, and Bloch, S., Pigeon's eyes converge during feeding: evidence for frontal binocular fixation in a lateral-eyed bird, Neurosci. Lett., 45 (1984) 335-339. 12 Montgomery, J.C., Eye movement dynamics in the dogfish, J. Exp. BioL, 105 (1983) 297-303. 13 Montgomery, J.C. and Macdonald, J.A., Performance of motor systems in antarctic fishes, J. Comp. Physiol., A. 154 (1984) 241-248. 14 Montgomery, J.C., McVean, A.R. and McCarthy, D., The effects of lowered temperature on spontaneous eye movements in a teleost fish, Comp. Biochem. Physiol., 75 A (1983) 363-368. 15 Northmore, D.P.M., Williams, B. and Vanegas, H., The teleostean torus longitudinalis: responses related to eye movements visuotopic mapping, and functional relations with the optic tectum, J. Comp. Physiol., 150 (1983) 39-50. 16 Nye, P.W., The monocular eye movements of the pigeon, Vision Res., 9 (1969) 133-144. 17 Reinhart, R.J. and Zuber, B.L., Horizontal eye movements from abducens nerve stimulation in the cat, IEEE Trans. Biomed. Eng., BME-17 (I) (1970) 11-14. 18 Robinson, D.A., The use of control systems analysis in the neurophysiology of eye movements, Ann Rev. Neurosci., 4 (1981) 463-503. 19 Zuber, B.L., Sinusoidal eye movements from brain stimulation in the cat, 8 (1968) 1073-1079.