Journal of Neuroscience Methods, 44 (1992) 233-240 © 1992 Elsevier Science Publishers B.V. All rights reserved 0165-0270/92/$05.00
233
NSM 01417
The effects of low-pass filtering on the flash visual evoked potential of the albino rat N.A. Shaw
Department of Physiology, School of Medicine, Unicersityof Auckland, Auckland l (New Zealand) (Received 6 April 1992) (Revised version 17 July 1992) (Accepted 23 July 1992)
Key words: Flash visual evoked potential; Low-pass filtering; Visual cortex; (Rat) Flash visual evoked potentials (FVEPs) were recorded from the rat in order to determine the effects of low-pass filtering on the wave form. The low-frequency (high pass) filter remained fixed at 3.2 Hz while the setting of the high-frequency (low-pass) filter was progressively raised from 32 Hz to 3.2 kHz. The amplitude of the primary cortical potential (P30) steadily increased while its latency decreased until asymptotic values were recorded with a low-pass cut-off of 320 Hz. Thereafter, there was little additional change in wave form..It is concluded that a bandpass of 3.2-320 Hz is optimal to record the primary cortical response of the FVEP, and this is consistent with the theory that the P30 potential is generated by comparatively slow post-synaptic activity. In a second experiment the effects of low-pass filtering were examined on the later and more labile secondary components of the FVEP wave form. These were found to be less responsive to low-pass filtering than the early components and assumed a near optimal configuration when the high-frequency cut-off was raised to 80 Hz. The high-frequency filter setting which is most appropriate to record the primary component of the FVEP therefore appears to be more than adequate also to record the secondary responses. It is also shown that the same principles of low-pass filtering on the FVEP will apply irrespective of whether the subject is awake or anaesthetised.
Experiment 1 Introduction N u m e r o u s reports have now described the flash visual evoked p o t e n t i a l ( F V E P ) in the rat. T h e r e a s o n s for the p o p u l a r i t y of this p o t e n t i a l in neurophysiological r e s e a r c h are p r o b a b l y not hard to discern. I n particular, since the striate area occupies a comparatively large p r o p o r t i o n of the rat c e r e b r a l cortex ( A d a m s a n d F o r r e s t e r , 1968; M o n t e r o , 1973; L e V e r e , 1978), the r e c o r d i n g electrode does not n e e d to b e quite so carefully p o s i t i o n e d as it does to successfully record either
Correspondence: N.A. Shaw, Department of Physiology,School of Medicine, University of Auckland, Auckland 1, New Zealand. Tel. 373-7999; FAX: 373-7481.
auditory evoked p o t e n t i a l s or s o m a t o s e n s o r y evoked p o t e n t i a l s (Shaw a n d Cant, 1981; Shaw, 1990). D e s p i t e its w i d e s p r e a d use, the ideal recording c o n d i t i o n s u n d e r which the F V E P should be o b t a i n e d still r e m a i n to be defined. T h e r e is, for example, a lack of c o n s e n s u s o n exactly what is the correct cut-off of the low-pass (or high-frequency) filter of the recording amplifiers. As a c o n s e q u e n c e , F V E P s have b e e n r e c o r d e d using a wide r a n g e of settings of the low-pass filter. If the high-frequency cut-off is set at too low a value, latencies, a m p l i t u d e s a n d overall wave form m o r p h o l o g y will be distorted. Conversely, if the low-pass cut-off is raised too high, there is a d a n g e r that the wave form will be c o n t a m i n a t e d by e x t r a n e o u s high-frequency noise ( L u d e r s et al., 1985). Identifying the most appropriate cut-off setting of the low-pass filter for
234
recording the FVEP is not merely of academic interest. For instance, the FVEP of the rat has been used to measure the impact of a variety of neurotoxins (e.g., Xintaras et al., 1966; Petajan et al., 1976; Dyer and Annau, 1977; Fox et al., 1977; Dyer et al., 1978; Boyes and Dyer, 1984), as well as to determine the mechanisms of action of a range of sedative and anaesthetic agents (e.g., Begleiter et al., 1972; Dafny and Rigor, 1978; Rabe et al., 1980; Hetzler et al., 1981; Hetzler and Oakley, 1981; Hetzler and Dyer, 1984). In addition, the rat FVEP has been employed to simulate abnormalities found in evoked potential recordings from patients with neurological disease (e.g., Dyer et al., 1981; Onofrj and BodisWollner, 1982). Correct interpretation and comparison of such data depends upon the wave BANDPASS
P30 -
32
B. P30 (35.9) ~'~ ~ ~
(37.9) 3.2 - 32
7-35.2
A 3.2 - 80
Methods Subjects were 10 adult male albino rats (300400 g). On the day of the experiment, each animal was initially anaesthetised with a surgical dose of pentobarbital (60 mg/kg). Three extradural stainless steel electrodes were then inserted over both the left and right visual cortices. The position of the cortical electrode was 6-7 mm posterior to bregma and 3 - 4 mm lateral to the sagittal suture. The screw from which the best defined FVEP was recorded served as the active electrode while the other served as the ground. A BANDPASS (HZ)
A.
(HZ)
3.2
forms having been recorded with the optimum bandpass. In the present communication, the effects of low-pass filtering on the FVEP of the rat are described.
~
_+]20 pv lO MS 3.2 - 80
___~...~ t
//\
._...._~
32.8
~ ~.~._...%_
~.,..~.
3.2
160
32.3 3.2 - 320
•
~
31,8 3.2 - 320
32.1 3.2 - 800
A
~...~'~ 33.1
3.2 - 160
+J20 pV l0 MS
34.7
~_.~.,~'~%~,
31.7 3.2 - 800
~~/~,~/.~=-m,..,%
32.2
31.5
3.2 - 1.6K
3.2 - 1.6K
;
.
~
~
%
~
32.5 3.2 - 3.2K
3.2 - 3.2K r "
'~'~
Fig. 1. Two illustrations of the effects of low-pass filtering on the FVEP. In the top trace, of each example, the P30 component is
identified with its actual latency (ms) in parentheses. For the remaining traces, just the peak latencies are indicated.
235
3rd electrode was inserted in the nasal bone to serve as the reference. Other positions for the ground electrode were tested. These included inserting it into the nasal bone close to the reference electrode. This change in montage produced no detectable change in the wave form recorded. Animals were stimulated with light pulses delivered by a Grass model PS 22 photic stimulator. Duration of each flash was 10 /zs and stimulus rate was 2 / s . Flash intensity was set to step 8 on the 1-16 scale. The lamp was positioned approximately 15 cm from the animal's eye. In order to muffle the click which accompanied each light flash, the lamp was enclosed in a foam rubber container with a sheet of perspex cemented over the opening. FVEPs were recorded following monocular stimulation of. the eye contralateral to the cortical electrode. The non-stimulated eye was covered with a plasticine patch. Recordings were made with the animal's eye open. FVEPs were recorded with a Medelec MS6 while the animals were in a state of moderate anaesthesia. Sixty-four responses were averaged to obtain each potential. Sweep length of the trace was 100 ms and sampling interval was 100 ~s. FVEPs were recorded with the cut-off frequency of the lowpass filter set at 32 Hz, 80 Hz, 160 Hz, 320 Hz, 800 Hz, 1.6 kHz, and 3.2 kHz. The cut-off frequency of the high-pass filter remained fixed at 3.2 Hz. The analog filter had a 12 d B / o c t a v e roll-off. On analysis, the filter was found to have a Laplacian transfer function of ~'-~S%1 1• Any effects of phase distortion were considered to be minimal. The 7 F V E P recordings obtained from each animal were made in random order. Throughout the experiment, a steady rectal temperature of 37-38°C was maintained with a heat lamp (Hetzler et al., 1988). All recordings were made in an otherwise totally darkened room.
Results and discussion Two examples of how the F V E P wave form is modified as the high-frequency cut-off is raised are illustrated in Fig. 1. The principal component of each trace was a positive potential with a peak latency which occurred between 30 and 40 ms following the light flash. This response was conse-
Amplitude
1~
of P30 (%)
60
3.2-32
3.2-80
3.2-160
3.2-320
3.2-800 3.2-1.6K 3.2-3.2K
Fig. 2. Mean latencies and amplitudes ( + 1 S.D.) of the P30 component of the FVEP as a function of raising the cut-off frequency of the low-pass filter. Amplitudes are expressed as a percentage of those recorded using a bandpass of 3.2-3.2 kHz because of the inter-animal variability in absolute voltages.
quently labelled P30. In both examples, the latency of P30 decreased while its amplitude increased until the high-frequency filter was set at 320 Hz. Thereafter, there was little detectable change in wave form, even when the bandpass was opened as wide as 3.2-3.2 kHz. Group data summarizing the effects of low-pass filtering on the amplitude and latency of P30 is presented in Fig. 2. Amplitude of P30 was calculated as the difference between P30 and the preceding negative baseline. Fig. 2 confirms the individual trends shown in Fig. 1 that asymptotic values in both amplitude and latency are acquired when the high-frequency filter is raised to 320 Hz. A statistical analysis was made of the data to determine the degree of significance of the decreases in latency with progressive widening of the bandpass. A N O V A revealed significant differences among the 7 bandpass groups, F (6, 63) = 16.50, P < 0.005. Duncan's multiple range test was used to make pairwise comparisons among the groups at the 0.05 level of significance. The mean latency of the 80 Hz group was significantly shorter than that of the 32 Hz group, and both groups had significantly longer latencies than the 160 Hz, 320 Hz, 800 Hz, 1.6 kHz and 3.2 kHz groups. The latter 5 groups did not differ among themselves. A similar analysis was conducted on the amplitude data using the original voltages of P30, not the percentage values shown in Fig. 2.
236
A N O V A revealed significant differences among the 7 groups, F (6, 63) = 6.44, P < 0.01. Duncan's test demonstrated that the 32 Hz group had a significantly lower amplitude than that of any of the other 6 groups. The 80 Hz group did not differ from the 160 Hz or 320 Hz groups but had a significantly smaller amplitude than the 800 Hz, 1.6 kHz and 3.2 kHz groups. No other differences in amplitude among the 7 groups were found. Unlike the latency data, the statistical analysis of the P30 voltages was confounded to some extent by the wide inter-animal variability in amplitudes. In general, the statistical analysis confirmed the visual impression conveyed in Fig. 2 that while a high-frequency cut-off of 160 Hz may be adequate, it seems more appropriate to set the lowpass filter at 320 Hz. Judging by its latency, polarity, relative insensitivity to anaesthesia, and confinement to the visual receiving area, the P30 component of the FVEP wave form most likely represents the primary cortical response. This potential is generated by the afferent volley propagated within the specific fast conducting retino-geniculo-striate pathway (Creel et al., 1974). There is also a close temporal coincidence between the onset of unitary activity in the rat visual cortex and the latency of P30 (Montero, 1973), as well as between P30 and evoked potentials recorded directly from the visual cortex (O'Steen and Anderson, 1971). Little information seems to exist on the effects of low-pass filtering on FVEPs of any type. In a single example, Celesia (1982) has shown that lowering the high-frequency cut-off to 100 Hz will increase the latency and decrease the amplitude of the pattern VEP in man. In the rat, FVEPs have been recorded with the low-pass filter set as low as 50 Hz (Yellin and Jerison, 1973) and as high as 40 kHz (Fox et al., 1977). The present findings suggest that it is prudent to set the low-pass filter to at least 300 Hz in order to avoid distortion of the wave form. Conversely, little appears to be gained by setting the low-pass filter much higher than 300 Hz. This conclusion would also be consistent with the theory that primary cortical responses are generated by relatively slow post-synaptic activity with a frequency composition of less than 300 Hz (Liiders et ai., 1985).
The present study was conducted using albino rats as subjects. There is some evidence that the FVEP recorded from albino animals differs slightly from that recorded from hooded rats (Creel et al., 1970). While the basic wave form appears to be very similar in both strains of rats, the latencies recorded from hooded animals are reported to be marginally shorter than those from albinos. Differences in retinal responsiveness and the number of uncrossed fibres which remain functional in the optic pathways have been proposed to explain the earlier latencies in the hooded animals (Creel et al., 1970). Apart from disparities in physiology and anatomy, albinos may also differ from other strains of rats in their b e h a v i o u r a l r e s p o n s i v e n e s s a n d in t h e i r metabolism of drugs such as the barbiturates (Shearer and Creel, 1978). In the past, albino and hooded animals have been used about equally in studies of the FVEP. However, during the last decade there appears to have been a decline in the use of albino animals. This is presumably because of the belief that hooded rats possess superior eye sight. While there seems no reason to suspect that the results of low-pass filtering on the FVEP found in the present study cannot be generalised to different strains of rat, this should perhaps be done with caution. When recording FVEPs from human patients, the low-pass filter is now usually set at no more than 100 Hz (e.g., Sohmer et al., 1986; Stanley et al., 1987; Taylor et al., 1987; Cedzich et al., 1988). The present results suggest that it may be more appropriate to raise the cut-off frequency to 200300 Hz. This could enhance the clinically significant early cortical potentials which are often hard to obtain.
Experiment 2 Introduction The later or secondary components of the FVEP wave form lack the robustness and stability of P30 and show greater inter-animal variability. They are also markedly sensitive to the animal's state of arousal (Schwartzbaum et al., 1971; Creel et al., 1974). In experiment 1, the F V E P was
237 recorded under a moderate level of pentobarbital anaesthesia which tends to enhance the amplitude o f the primary component, while largely suppressing or abolishing the later responses (Hetzler and Oakley, 1981). As a consequence, experiment 1 was restricted to studying the effects of low-pass filtering on the early or primary components of the FVEP. However, studies of the FVEP are frequently concerned with both the early and late responses and often compare their respective wave forms under various pharmacological and b e h a v i o u r a l c o n d i t i o n s (e.g., Schwartzbaum et al., 1971; Creel et al., 1974; Fox et al., 1977; Dyer et al., 1981; Hetzler et al., 1981, 1988; Hetzler and Oakley, 1981; Boyes and Dyer, 1984). The question therefore arises as to the extent to which the principles of low-pass filtering derived from just the primary response of the FVEP can be generalised to the wave form as a whole. To resolve this matter was the purpose of experiment 2. In order to augment and facilitate the appearance of the secondary components, recordings were made while the animals were in a state of complete wakefulness or very light anaesthesia. Methods Subjects were 3 adult male albino rats (300-400 g). Recording and stimulating conditions were identical to those used in experiment 1 but with the following changes. The analysis time was lengthened to 200 ms in order to capture the later components of the FVEP wave form. Sampling interval was now 200 /xs and each average was derived from 128 samples. In the 1st animal, electrodes were implanted under pentobarbital anaesthesia but recordings of the FVEP were delayed until the animal was in a state of very light anaesthesia or sedation. This was characterised by an increase in the fast activity of the E E G plus the return of blink and withdrawal reflexes. The 2nd animal was treated identically to the first, except that the FVEP recordings were obtained only when the subject had sufficiently recovered from the anaesthesia that not only reflex activity, but also spontaneous movements had resumed. In order to make the FVEP recordings,
the animal was gently restrained in front of the photic stimulator lamp. In the 3rd animal, FVEP recordings were made while it was totally awake but immobilised. T h i s was achieved by chronically implanting the electrodes 1 week before the experiment. On the day of the experiment, the subject was paralysed with an i.p. injection of D-tubocurarine chloride (4 m g / k g ) and artificially ventilated using a rodent respirator connected to the animal via a balloon mask fitted over the snout. Stroke rate and volume of the ventilator were adjusted to maintain a heart rate within normal limits (400-500 b.p.m.). This is a modification of the technique developed by Miller and his colleagues for their studies of biofeedback in the rat (Miller, 1969) and has been used previously to record the FVEP from the awake rat (Schwartzbaum et al., 1971). There is no reason to believe that such a procedure will cause the animal any distress or discomfort, providing it is not subject to any additional painful or stressful stimuli. In deference to any concern about the use of this technique, however, it was restricted to just a single animal and the subject was killed (with an overdose of pentobarbital) immediately after completion of the 7 FVEP recordings. A more comprehensive discussion ot the justification for the use of neuromuscular blockade in the absence of anaesthesia or sedation can be found in Foutz et al. (1983). Results and discussion An example of the effects.of low-pass filtering on both the early and late components of the FVEP recorded from each of 3 subjects is shown in Fig. 3. The animal shown in the 1st illustration (Fig. 3A) was only lightly anaesthetised, the animal in the 2nd illustration (Fig. 3B) had resumed spontaneous movements and the animal in the 3rd illustration (Fig. 3C) was completely awake. Considering the examples as a whole, it is clear that the FVEP in the rat consists of an initial positive (P30)-negative primary component, followed by a much slower positive-negative-positive complex. The last of those components appears to be missing from the traces shown in Fig. 3C. The later a component is being generated, the greater the
238
inter-animal variability in its latency. Superimposed on this basic wave form are a variety of subcomponents which often make peak latencies difficult to identify. An inspection of the data in Fig. 3 confirms the findings of experiment 1 that the latency and amplitude of the primary response (P30) stabilize when the low-pass cut-off is raised to 320 Hz. In contrast, the secondary components appeared to be less responsive to low-pass filtering than the primary component. Even with the high-frequency filter set at just 32 Hz, the later components had
BANDPASS
A.
73
(HZ)
a near optimum amplitude. Allowing for the intrinsic variability in the latency of the secondary responses, the illustrations in Fig. 3 suggest that stable amplitude and latency values are acquired when the high-frequency cut-off is raised to no more than 80 Hz. Subsequent increases in the setting of the high-frequency filter resulted in little additional change to the wave form of the secondary components, apart from a sharpening of their contour. Judging by the 3 examples in Fig. 3, a rule governing the effects of low-pass filtering on the secondary components of the
BANDPASS
B.
(HZ) 3.2
-
117 55 3.2
-
(35)/-,,
64
32
...I 25
pV
3.2
-
32
,~ !
42
~
2O M S
-
80
115
97
52 ~ -
56
130
160
30 3.2-
//'~
39
52 -
320
;
94 61
.,1 V
:
3.2
-
320
113 38
51 3.2
-
119
160
114
3.2
_+.]5o ~v
80 3.2
3.2
99
59
96 61
800 3.2
-
800
113
95 51
~ 3.2-
60
132
1.6K
112
3.2-
1.6K
61
51 3.2
-
135
3.2K
3.2 112
3.2K ,
a7
9-8
~-
Fig. 3. Three examples of the effects of low-pass filtering on both the early and late components of the FVEP. A: the FVEP was recorded from a lightly anaesthetised animal in whom reflex activity but not spontaneous movement had returned. B: the subject had resumed spontaneous movement and the recordings were obtained by gently restraining the animal in front of the photic stimulator lamp. C: the subject was totally awake when the recordings were made. In the top trace of each example, the P30 component is identified with its actual latency (ms) in parentheses. The remaining principal components of the wave form are identified by just their peak latency. W h e n identifying the peak latency, usually the most positive or negative aspect of the relevant wave form was chosen. Note that, in contrast to A and B, only 4 principal components of the wave form can be consistently identified in C.
239 BANDPASS
C.
69
(HZ) 3.2
-
32
63 3.2
-
28
80
._+[25 ~ V
61 3.2
-
165
153
160
i
140 3.2
-
320
60
3.2
-
800
58 3.2-
144
1.6K
57
3.2
148
-
152
3.2K
155
Fig. 3 (continued).
FVEP can therefore be formulated. Providing the optimum high-frequency cut-off to record the primary response has been determined, this should be more than adequate to appropriately record any of the later components. A second potential difficulty with the data of experiment 1 concerns the validity of extrapolating the findings from anaesthetised to awake animals. To try to resolve this issue was a principal reason for recording the FVEP from the paralysed but awake animal. In experiment 1, there was a mean decrease in the latency of P30 of 5 ms as the bandpass was widened (Fig. 2). In the awake animal, by contrast, the decrease in latency of P30 occurred within a shorter latency range (3 ms). This presumably reflected the absence of a pentobarbital-induced prolongation in
the latency of P30. Otherwise, exactly the same principles of low-pass filtering on the FVEP wave form seemed to apply irrespective of whether the animal was anaesthetised or awake. This is clear from Fig. 3C in which the P30 latency stabilised at 26 ms after the bandpass was widened to 3.2-320 Hz. Similar findings are apparent for the amplitude and general configuration of the P30 wave form. The effects of low-pass filtering on the FVEP remain untested in at least one other condition of the animal, w h e n the subject is not only fully awake but also freely moving. Many studies have recorded FVEPs from freely behaving rats and the latencies and basic wave forms apparently remain nearly identical to those of restrained animals (Creel et al., 1970). In view of the very similar effects of low-pass filtering on the FVEP during all other states of arousal, it would seem unlikely that they would differ in the awake and unrestrained animal. Quantifying the effects of different filter settings on the FVEP might also prove to be an unreliable and difficult task considering the amount of movement and E M G artifact which may be generated by freely moving animals. In addition, unrestrained rats have been reported to display a marked intrasubject variability in the amplitudes of their FVEPs (Creel et al., 1970).
References Adams, A.D. and Forrester, J.M. (1968) The projection of the rat's visual field on the cerebral cortex. Quart. J. Exp. Physiol., 53: 327-336. Begleiter, H., Branchey, M.H. and Kissin, B. (1972) Effects of ethanol on evoked potentials in the rat. Behav. Biol., 7: 137-142. Boyes, W.K. and Dyer, R.S. (1984) Chlordimeform produces profound, selective, and transient changes in visual evoked potentials of hooded rats. Exp. Neurol., 86: 434-447. Cedzich, C., Schramm, J., Mengedoht, C.F. and Fahlbusch, R. (1988) Factors that limit the use of flash visual evoked potentials for surgical monitoring. Electroenceph. Clin. Neurophysiol., 71: 142-145. Celesia, G.G. (1982) Clinical applications of evoked potentials. In: E. Niedermeyer and F. Lopes da Silva (Eds.), Electroencephalography, Urban Schwarzenberg, Baltimore, MA, pp. 665-684.
240 Creel, D.J., Dustman, R.E. and Beck, E.C. (1970) Differences in visually evoked responses in albino versus hooded rats. Exp. Neurol., 29: 298-309. Creel, D., Dustman, R.E. and Beck, E.C. (1974) Intensity of flash illumination and the visually evoked potential of rats, guinea pigs and cats. Vis. Res., 14: 725-729. Dafny, N. and Rigor, B.M. (1978) Dose effects of ketamine on photic and acoustic field potentials. Neuropharmacology, 17: 851-862. Dyer, R.S. and Annau, Z. (1977) Carbon monoxide and flash evoked potentials from rat cortex and superior colliculus. Pharmacol. Biochem. Behav., 6: 461-465. Dyer, R.S., Eccles, C.U. and Annau, Z. (1978) Evoked potential alterations following prenatal methyl mercury exposure, Pharmacol. Biochem. Behav., 8: 137-141. Dyer, R.S., Howell, W.E. and MacPhail, R.C. (198l) Dopamine depletion slows retinal transmission. Exp Neurol., 71: 326-340. Foutz, A.S., Dauthier, C. and Kerdelhue, B. (1983) /3-Endorphin plasma levels during neuromuscular blockade in unanesthetised cat.. Brain Res., 263:119-123. Fox, D.A., Lewkowski, J.P. and Cooper, G.P. (1977) Acute and chronic effects of neonatal lead exposure on development of the visual evoked response in rats. Toxicol. Appl. Pharmacol., 40: 449-461. Hetzler, B.E. and Dyer, R.S. (1984) Contribution of hypothermia to effects of chloral hydrate on flash evoked potentials of hooded rats. Pharmacol. Biochem. Behav., 21: 599-607. Hetzler, B.E. and Oakley, K.E. (1981) Dose effects of pentobarbital on evoked potentials in visual cortex and superior colliculus of the albino rat. Neuropharmacology, 20: 969978. Hetzler, B.E., Heilbronner, R.L., Griffin, J. and Griffin, G. (1981) Acute effects of alcohol on evoked potentials in visual cortex and superior colliculus of the rat. Electroenceph. Clin Neurophysiol., 51: 69-79. Hetzler, B.E., Boyes, W.K., Creason, J.P. and Dyer, R.S. (1988) Temperature-dependent ch~inges in visual evoked potentials of rats. Electroenceph. Clin. Neurophysiol., 70: 137-154. LeVere, T.E. (1978) The primary visual system of the rat: a primer of its anatomy. Physiol. Psychol., 6: 142-169. Liiders, H., Lesser, R.P., Dinner, D.S. and Morris, H.H. (1985) Optimizing stimulating and recording parameters in somatosensory evoked potential studies. J. Clin. Neurophysiol., 2: 383-396. Miller, N.E. (1969) Learning of visceral and glandular responses. Science, 163: 434-445.
Montero, V.M. (1973) Evoked responses in the rat's visual cortex to contralateral, ipsilateral and restricted photic stimulation. Brain Res., 53: 192-196. Onofrj, M. and Bodis-Wollner, I. (1982) Dopaminergic deficiency causes delayed visual evoked potentials in rats. Ann. Neurol., 11: 484-490. O'Steen, W.K. and Anderson, K.V. (1971) Photically evoked responses in the visual system of rats exposed to continuous light. Exp. Neurol., 30: 525-534. Petajan, J.H., Packham, S.C., Frens, D.B. and Dinger, B.G. (1976) Sequelae of carbon monoxide-induced hypoxia in the rat. Arch. Neurol., 33: 152-157. Rabe, L.S., Moreno, L., Rigor, B.M. and Dafny, N. (1980) Effects of halothane on evoked field potentials recorded from cortical and subcortical nuclei. Neuropharmacology, 19: 813-825. Schwartzbaum, J.S., Kreinick, C.J. and Gustafson, J.W. (1971) Cortical evoked potentials and behavioral reactivity to photic stimuli in freely moving rats. Brain Res., 27: 295307. Shaw, N.A. (1990) Central auditory conduction time in the rat. Exp. Brain Res., 79: 217-220. Shaw, N.A. and Cant, B.R. (1981) The effect of pentobarbital on central somatosensory conduction time in the rat. Electroenceph. Clin. Neurophysiol., 51:674-677. Shearer, D.E. and Creel, D. (1978) The photically evoked after discharge: current concepts and potential applications. Physiol. Psychol., 6: 369-376. Sohmer, H., Freeman, S. and Malachi, S. (1986) Multi-modality evoked potentials in hypoxaemia. Electroenceph. Clin. Neurophysiol., 64: 328-333. Stanley, O.H., Fleming, P.J. and Morgan, M.H. (1987) Developmental wave form analysis of the neonatal flash evoked potential. Electroenceph. Clin. Neurophysiol., 68: 149-152. Taylor, M.J., Menzies, R., MacMillan, L.J. and Whyte, H.E. (1987) VEPs in normal full-term and premature neonates:longitudinal versus cross-sectional data. Electroenceph. Clin. Neurophysiol., 68: 20-27. Xintaras, C., Johnson, B.K., Ulrich, C.E., Terrill, R.E. and Sobecki, M.F. (1966) Application of the evoked response technique in air pollution toxicology. Toxicol. Appl. Pharmacol., 8: 77-87. Yellin, A.M. and Jerison, H.J. (1973) Visual evoked potentials and inter-stimulus intervals in the rat. Electroenceph. Clin. Neurophysiol., 34: 429-432.
233
NSM 01417
The effects of low-pass filtering on the flash visual evoked potential of the albino rat N.A. Shaw
Department of Physiology, School of Medicine, Unicersityof Auckland, Auckland l (New Zealand) (Received 6 April 1992) (Revised version 17 July 1992) (Accepted 23 July 1992)
Key words: Flash visual evoked potential; Low-pass filtering; Visual cortex; (Rat) Flash visual evoked potentials (FVEPs) were recorded from the rat in order to determine the effects of low-pass filtering on the wave form. The low-frequency (high pass) filter remained fixed at 3.2 Hz while the setting of the high-frequency (low-pass) filter was progressively raised from 32 Hz to 3.2 kHz. The amplitude of the primary cortical potential (P30) steadily increased while its latency decreased until asymptotic values were recorded with a low-pass cut-off of 320 Hz. Thereafter, there was little additional change in wave form..It is concluded that a bandpass of 3.2-320 Hz is optimal to record the primary cortical response of the FVEP, and this is consistent with the theory that the P30 potential is generated by comparatively slow post-synaptic activity. In a second experiment the effects of low-pass filtering were examined on the later and more labile secondary components of the FVEP wave form. These were found to be less responsive to low-pass filtering than the early components and assumed a near optimal configuration when the high-frequency cut-off was raised to 80 Hz. The high-frequency filter setting which is most appropriate to record the primary component of the FVEP therefore appears to be more than adequate also to record the secondary responses. It is also shown that the same principles of low-pass filtering on the FVEP will apply irrespective of whether the subject is awake or anaesthetised.
Experiment 1 Introduction N u m e r o u s reports have now described the flash visual evoked p o t e n t i a l ( F V E P ) in the rat. T h e r e a s o n s for the p o p u l a r i t y of this p o t e n t i a l in neurophysiological r e s e a r c h are p r o b a b l y not hard to discern. I n particular, since the striate area occupies a comparatively large p r o p o r t i o n of the rat c e r e b r a l cortex ( A d a m s a n d F o r r e s t e r , 1968; M o n t e r o , 1973; L e V e r e , 1978), the r e c o r d i n g electrode does not n e e d to b e quite so carefully p o s i t i o n e d as it does to successfully record either
Correspondence: N.A. Shaw, Department of Physiology,School of Medicine, University of Auckland, Auckland 1, New Zealand. Tel. 373-7999; FAX: 373-7481.
auditory evoked p o t e n t i a l s or s o m a t o s e n s o r y evoked p o t e n t i a l s (Shaw a n d Cant, 1981; Shaw, 1990). D e s p i t e its w i d e s p r e a d use, the ideal recording c o n d i t i o n s u n d e r which the F V E P should be o b t a i n e d still r e m a i n to be defined. T h e r e is, for example, a lack of c o n s e n s u s o n exactly what is the correct cut-off of the low-pass (or high-frequency) filter of the recording amplifiers. As a c o n s e q u e n c e , F V E P s have b e e n r e c o r d e d using a wide r a n g e of settings of the low-pass filter. If the high-frequency cut-off is set at too low a value, latencies, a m p l i t u d e s a n d overall wave form m o r p h o l o g y will be distorted. Conversely, if the low-pass cut-off is raised too high, there is a d a n g e r that the wave form will be c o n t a m i n a t e d by e x t r a n e o u s high-frequency noise ( L u d e r s et al., 1985). Identifying the most appropriate cut-off setting of the low-pass filter for
234
recording the FVEP is not merely of academic interest. For instance, the FVEP of the rat has been used to measure the impact of a variety of neurotoxins (e.g., Xintaras et al., 1966; Petajan et al., 1976; Dyer and Annau, 1977; Fox et al., 1977; Dyer et al., 1978; Boyes and Dyer, 1984), as well as to determine the mechanisms of action of a range of sedative and anaesthetic agents (e.g., Begleiter et al., 1972; Dafny and Rigor, 1978; Rabe et al., 1980; Hetzler et al., 1981; Hetzler and Oakley, 1981; Hetzler and Dyer, 1984). In addition, the rat FVEP has been employed to simulate abnormalities found in evoked potential recordings from patients with neurological disease (e.g., Dyer et al., 1981; Onofrj and BodisWollner, 1982). Correct interpretation and comparison of such data depends upon the wave BANDPASS
P30 -
32
B. P30 (35.9) ~'~ ~ ~
(37.9) 3.2 - 32
7-35.2
A 3.2 - 80
Methods Subjects were 10 adult male albino rats (300400 g). On the day of the experiment, each animal was initially anaesthetised with a surgical dose of pentobarbital (60 mg/kg). Three extradural stainless steel electrodes were then inserted over both the left and right visual cortices. The position of the cortical electrode was 6-7 mm posterior to bregma and 3 - 4 mm lateral to the sagittal suture. The screw from which the best defined FVEP was recorded served as the active electrode while the other served as the ground. A BANDPASS (HZ)
A.
(HZ)
3.2
forms having been recorded with the optimum bandpass. In the present communication, the effects of low-pass filtering on the FVEP of the rat are described.
~
_+]20 pv lO MS 3.2 - 80
___~...~ t
//\
._...._~
32.8
~ ~.~._...%_
~.,..~.
3.2
160
32.3 3.2 - 320
•
~
31,8 3.2 - 320
32.1 3.2 - 800
A
~...~'~ 33.1
3.2 - 160
+J20 pV l0 MS
34.7
~_.~.,~'~%~,
31.7 3.2 - 800
~~/~,~/.~=-m,..,%
32.2
31.5
3.2 - 1.6K
3.2 - 1.6K
;
.
~
~
%
~
32.5 3.2 - 3.2K
3.2 - 3.2K r "
'~'~
Fig. 1. Two illustrations of the effects of low-pass filtering on the FVEP. In the top trace, of each example, the P30 component is
identified with its actual latency (ms) in parentheses. For the remaining traces, just the peak latencies are indicated.
235
3rd electrode was inserted in the nasal bone to serve as the reference. Other positions for the ground electrode were tested. These included inserting it into the nasal bone close to the reference electrode. This change in montage produced no detectable change in the wave form recorded. Animals were stimulated with light pulses delivered by a Grass model PS 22 photic stimulator. Duration of each flash was 10 /zs and stimulus rate was 2 / s . Flash intensity was set to step 8 on the 1-16 scale. The lamp was positioned approximately 15 cm from the animal's eye. In order to muffle the click which accompanied each light flash, the lamp was enclosed in a foam rubber container with a sheet of perspex cemented over the opening. FVEPs were recorded following monocular stimulation of. the eye contralateral to the cortical electrode. The non-stimulated eye was covered with a plasticine patch. Recordings were made with the animal's eye open. FVEPs were recorded with a Medelec MS6 while the animals were in a state of moderate anaesthesia. Sixty-four responses were averaged to obtain each potential. Sweep length of the trace was 100 ms and sampling interval was 100 ~s. FVEPs were recorded with the cut-off frequency of the lowpass filter set at 32 Hz, 80 Hz, 160 Hz, 320 Hz, 800 Hz, 1.6 kHz, and 3.2 kHz. The cut-off frequency of the high-pass filter remained fixed at 3.2 Hz. The analog filter had a 12 d B / o c t a v e roll-off. On analysis, the filter was found to have a Laplacian transfer function of ~'-~S%1 1• Any effects of phase distortion were considered to be minimal. The 7 F V E P recordings obtained from each animal were made in random order. Throughout the experiment, a steady rectal temperature of 37-38°C was maintained with a heat lamp (Hetzler et al., 1988). All recordings were made in an otherwise totally darkened room.
Results and discussion Two examples of how the F V E P wave form is modified as the high-frequency cut-off is raised are illustrated in Fig. 1. The principal component of each trace was a positive potential with a peak latency which occurred between 30 and 40 ms following the light flash. This response was conse-
Amplitude
1~
of P30 (%)
60
3.2-32
3.2-80
3.2-160
3.2-320
3.2-800 3.2-1.6K 3.2-3.2K
Fig. 2. Mean latencies and amplitudes ( + 1 S.D.) of the P30 component of the FVEP as a function of raising the cut-off frequency of the low-pass filter. Amplitudes are expressed as a percentage of those recorded using a bandpass of 3.2-3.2 kHz because of the inter-animal variability in absolute voltages.
quently labelled P30. In both examples, the latency of P30 decreased while its amplitude increased until the high-frequency filter was set at 320 Hz. Thereafter, there was little detectable change in wave form, even when the bandpass was opened as wide as 3.2-3.2 kHz. Group data summarizing the effects of low-pass filtering on the amplitude and latency of P30 is presented in Fig. 2. Amplitude of P30 was calculated as the difference between P30 and the preceding negative baseline. Fig. 2 confirms the individual trends shown in Fig. 1 that asymptotic values in both amplitude and latency are acquired when the high-frequency filter is raised to 320 Hz. A statistical analysis was made of the data to determine the degree of significance of the decreases in latency with progressive widening of the bandpass. A N O V A revealed significant differences among the 7 bandpass groups, F (6, 63) = 16.50, P < 0.005. Duncan's multiple range test was used to make pairwise comparisons among the groups at the 0.05 level of significance. The mean latency of the 80 Hz group was significantly shorter than that of the 32 Hz group, and both groups had significantly longer latencies than the 160 Hz, 320 Hz, 800 Hz, 1.6 kHz and 3.2 kHz groups. The latter 5 groups did not differ among themselves. A similar analysis was conducted on the amplitude data using the original voltages of P30, not the percentage values shown in Fig. 2.
236
A N O V A revealed significant differences among the 7 groups, F (6, 63) = 6.44, P < 0.01. Duncan's test demonstrated that the 32 Hz group had a significantly lower amplitude than that of any of the other 6 groups. The 80 Hz group did not differ from the 160 Hz or 320 Hz groups but had a significantly smaller amplitude than the 800 Hz, 1.6 kHz and 3.2 kHz groups. No other differences in amplitude among the 7 groups were found. Unlike the latency data, the statistical analysis of the P30 voltages was confounded to some extent by the wide inter-animal variability in amplitudes. In general, the statistical analysis confirmed the visual impression conveyed in Fig. 2 that while a high-frequency cut-off of 160 Hz may be adequate, it seems more appropriate to set the lowpass filter at 320 Hz. Judging by its latency, polarity, relative insensitivity to anaesthesia, and confinement to the visual receiving area, the P30 component of the FVEP wave form most likely represents the primary cortical response. This potential is generated by the afferent volley propagated within the specific fast conducting retino-geniculo-striate pathway (Creel et al., 1974). There is also a close temporal coincidence between the onset of unitary activity in the rat visual cortex and the latency of P30 (Montero, 1973), as well as between P30 and evoked potentials recorded directly from the visual cortex (O'Steen and Anderson, 1971). Little information seems to exist on the effects of low-pass filtering on FVEPs of any type. In a single example, Celesia (1982) has shown that lowering the high-frequency cut-off to 100 Hz will increase the latency and decrease the amplitude of the pattern VEP in man. In the rat, FVEPs have been recorded with the low-pass filter set as low as 50 Hz (Yellin and Jerison, 1973) and as high as 40 kHz (Fox et al., 1977). The present findings suggest that it is prudent to set the low-pass filter to at least 300 Hz in order to avoid distortion of the wave form. Conversely, little appears to be gained by setting the low-pass filter much higher than 300 Hz. This conclusion would also be consistent with the theory that primary cortical responses are generated by relatively slow post-synaptic activity with a frequency composition of less than 300 Hz (Liiders et ai., 1985).
The present study was conducted using albino rats as subjects. There is some evidence that the FVEP recorded from albino animals differs slightly from that recorded from hooded rats (Creel et al., 1970). While the basic wave form appears to be very similar in both strains of rats, the latencies recorded from hooded animals are reported to be marginally shorter than those from albinos. Differences in retinal responsiveness and the number of uncrossed fibres which remain functional in the optic pathways have been proposed to explain the earlier latencies in the hooded animals (Creel et al., 1970). Apart from disparities in physiology and anatomy, albinos may also differ from other strains of rats in their b e h a v i o u r a l r e s p o n s i v e n e s s a n d in t h e i r metabolism of drugs such as the barbiturates (Shearer and Creel, 1978). In the past, albino and hooded animals have been used about equally in studies of the FVEP. However, during the last decade there appears to have been a decline in the use of albino animals. This is presumably because of the belief that hooded rats possess superior eye sight. While there seems no reason to suspect that the results of low-pass filtering on the FVEP found in the present study cannot be generalised to different strains of rat, this should perhaps be done with caution. When recording FVEPs from human patients, the low-pass filter is now usually set at no more than 100 Hz (e.g., Sohmer et al., 1986; Stanley et al., 1987; Taylor et al., 1987; Cedzich et al., 1988). The present results suggest that it may be more appropriate to raise the cut-off frequency to 200300 Hz. This could enhance the clinically significant early cortical potentials which are often hard to obtain.
Experiment 2 Introduction The later or secondary components of the FVEP wave form lack the robustness and stability of P30 and show greater inter-animal variability. They are also markedly sensitive to the animal's state of arousal (Schwartzbaum et al., 1971; Creel et al., 1974). In experiment 1, the F V E P was
237 recorded under a moderate level of pentobarbital anaesthesia which tends to enhance the amplitude o f the primary component, while largely suppressing or abolishing the later responses (Hetzler and Oakley, 1981). As a consequence, experiment 1 was restricted to studying the effects of low-pass filtering on the early or primary components of the FVEP. However, studies of the FVEP are frequently concerned with both the early and late responses and often compare their respective wave forms under various pharmacological and b e h a v i o u r a l c o n d i t i o n s (e.g., Schwartzbaum et al., 1971; Creel et al., 1974; Fox et al., 1977; Dyer et al., 1981; Hetzler et al., 1981, 1988; Hetzler and Oakley, 1981; Boyes and Dyer, 1984). The question therefore arises as to the extent to which the principles of low-pass filtering derived from just the primary response of the FVEP can be generalised to the wave form as a whole. To resolve this matter was the purpose of experiment 2. In order to augment and facilitate the appearance of the secondary components, recordings were made while the animals were in a state of complete wakefulness or very light anaesthesia. Methods Subjects were 3 adult male albino rats (300-400 g). Recording and stimulating conditions were identical to those used in experiment 1 but with the following changes. The analysis time was lengthened to 200 ms in order to capture the later components of the FVEP wave form. Sampling interval was now 200 /xs and each average was derived from 128 samples. In the 1st animal, electrodes were implanted under pentobarbital anaesthesia but recordings of the FVEP were delayed until the animal was in a state of very light anaesthesia or sedation. This was characterised by an increase in the fast activity of the E E G plus the return of blink and withdrawal reflexes. The 2nd animal was treated identically to the first, except that the FVEP recordings were obtained only when the subject had sufficiently recovered from the anaesthesia that not only reflex activity, but also spontaneous movements had resumed. In order to make the FVEP recordings,
the animal was gently restrained in front of the photic stimulator lamp. In the 3rd animal, FVEP recordings were made while it was totally awake but immobilised. T h i s was achieved by chronically implanting the electrodes 1 week before the experiment. On the day of the experiment, the subject was paralysed with an i.p. injection of D-tubocurarine chloride (4 m g / k g ) and artificially ventilated using a rodent respirator connected to the animal via a balloon mask fitted over the snout. Stroke rate and volume of the ventilator were adjusted to maintain a heart rate within normal limits (400-500 b.p.m.). This is a modification of the technique developed by Miller and his colleagues for their studies of biofeedback in the rat (Miller, 1969) and has been used previously to record the FVEP from the awake rat (Schwartzbaum et al., 1971). There is no reason to believe that such a procedure will cause the animal any distress or discomfort, providing it is not subject to any additional painful or stressful stimuli. In deference to any concern about the use of this technique, however, it was restricted to just a single animal and the subject was killed (with an overdose of pentobarbital) immediately after completion of the 7 FVEP recordings. A more comprehensive discussion ot the justification for the use of neuromuscular blockade in the absence of anaesthesia or sedation can be found in Foutz et al. (1983). Results and discussion An example of the effects.of low-pass filtering on both the early and late components of the FVEP recorded from each of 3 subjects is shown in Fig. 3. The animal shown in the 1st illustration (Fig. 3A) was only lightly anaesthetised, the animal in the 2nd illustration (Fig. 3B) had resumed spontaneous movements and the animal in the 3rd illustration (Fig. 3C) was completely awake. Considering the examples as a whole, it is clear that the FVEP in the rat consists of an initial positive (P30)-negative primary component, followed by a much slower positive-negative-positive complex. The last of those components appears to be missing from the traces shown in Fig. 3C. The later a component is being generated, the greater the
238
inter-animal variability in its latency. Superimposed on this basic wave form are a variety of subcomponents which often make peak latencies difficult to identify. An inspection of the data in Fig. 3 confirms the findings of experiment 1 that the latency and amplitude of the primary response (P30) stabilize when the low-pass cut-off is raised to 320 Hz. In contrast, the secondary components appeared to be less responsive to low-pass filtering than the primary component. Even with the high-frequency filter set at just 32 Hz, the later components had
BANDPASS
A.
73
(HZ)
a near optimum amplitude. Allowing for the intrinsic variability in the latency of the secondary responses, the illustrations in Fig. 3 suggest that stable amplitude and latency values are acquired when the high-frequency cut-off is raised to no more than 80 Hz. Subsequent increases in the setting of the high-frequency filter resulted in little additional change to the wave form of the secondary components, apart from a sharpening of their contour. Judging by the 3 examples in Fig. 3, a rule governing the effects of low-pass filtering on the secondary components of the
BANDPASS
B.
(HZ) 3.2
-
117 55 3.2
-
(35)/-,,
64
32
...I 25
pV
3.2
-
32
,~ !
42
~
2O M S
-
80
115
97
52 ~ -
56
130
160
30 3.2-
//'~
39
52 -
320
;
94 61
.,1 V
:
3.2
-
320
113 38
51 3.2
-
119
160
114
3.2
_+.]5o ~v
80 3.2
3.2
99
59
96 61
800 3.2
-
800
113
95 51
~ 3.2-
60
132
1.6K
112
3.2-
1.6K
61
51 3.2
-
135
3.2K
3.2 112
3.2K ,
a7
9-8
~-
Fig. 3. Three examples of the effects of low-pass filtering on both the early and late components of the FVEP. A: the FVEP was recorded from a lightly anaesthetised animal in whom reflex activity but not spontaneous movement had returned. B: the subject had resumed spontaneous movement and the recordings were obtained by gently restraining the animal in front of the photic stimulator lamp. C: the subject was totally awake when the recordings were made. In the top trace of each example, the P30 component is identified with its actual latency (ms) in parentheses. The remaining principal components of the wave form are identified by just their peak latency. W h e n identifying the peak latency, usually the most positive or negative aspect of the relevant wave form was chosen. Note that, in contrast to A and B, only 4 principal components of the wave form can be consistently identified in C.
239 BANDPASS
C.
69
(HZ) 3.2
-
32
63 3.2
-
28
80
._+[25 ~ V
61 3.2
-
165
153
160
i
140 3.2
-
320
60
3.2
-
800
58 3.2-
144
1.6K
57
3.2
148
-
152
3.2K
155
Fig. 3 (continued).
FVEP can therefore be formulated. Providing the optimum high-frequency cut-off to record the primary response has been determined, this should be more than adequate to appropriately record any of the later components. A second potential difficulty with the data of experiment 1 concerns the validity of extrapolating the findings from anaesthetised to awake animals. To try to resolve this issue was a principal reason for recording the FVEP from the paralysed but awake animal. In experiment 1, there was a mean decrease in the latency of P30 of 5 ms as the bandpass was widened (Fig. 2). In the awake animal, by contrast, the decrease in latency of P30 occurred within a shorter latency range (3 ms). This presumably reflected the absence of a pentobarbital-induced prolongation in
the latency of P30. Otherwise, exactly the same principles of low-pass filtering on the FVEP wave form seemed to apply irrespective of whether the animal was anaesthetised or awake. This is clear from Fig. 3C in which the P30 latency stabilised at 26 ms after the bandpass was widened to 3.2-320 Hz. Similar findings are apparent for the amplitude and general configuration of the P30 wave form. The effects of low-pass filtering on the FVEP remain untested in at least one other condition of the animal, w h e n the subject is not only fully awake but also freely moving. Many studies have recorded FVEPs from freely behaving rats and the latencies and basic wave forms apparently remain nearly identical to those of restrained animals (Creel et al., 1970). In view of the very similar effects of low-pass filtering on the FVEP during all other states of arousal, it would seem unlikely that they would differ in the awake and unrestrained animal. Quantifying the effects of different filter settings on the FVEP might also prove to be an unreliable and difficult task considering the amount of movement and E M G artifact which may be generated by freely moving animals. In addition, unrestrained rats have been reported to display a marked intrasubject variability in the amplitudes of their FVEPs (Creel et al., 1970).
References Adams, A.D. and Forrester, J.M. (1968) The projection of the rat's visual field on the cerebral cortex. Quart. J. Exp. Physiol., 53: 327-336. Begleiter, H., Branchey, M.H. and Kissin, B. (1972) Effects of ethanol on evoked potentials in the rat. Behav. Biol., 7: 137-142. Boyes, W.K. and Dyer, R.S. (1984) Chlordimeform produces profound, selective, and transient changes in visual evoked potentials of hooded rats. Exp. Neurol., 86: 434-447. Cedzich, C., Schramm, J., Mengedoht, C.F. and Fahlbusch, R. (1988) Factors that limit the use of flash visual evoked potentials for surgical monitoring. Electroenceph. Clin. Neurophysiol., 71: 142-145. Celesia, G.G. (1982) Clinical applications of evoked potentials. In: E. Niedermeyer and F. Lopes da Silva (Eds.), Electroencephalography, Urban Schwarzenberg, Baltimore, MA, pp. 665-684.
240 Creel, D.J., Dustman, R.E. and Beck, E.C. (1970) Differences in visually evoked responses in albino versus hooded rats. Exp. Neurol., 29: 298-309. Creel, D., Dustman, R.E. and Beck, E.C. (1974) Intensity of flash illumination and the visually evoked potential of rats, guinea pigs and cats. Vis. Res., 14: 725-729. Dafny, N. and Rigor, B.M. (1978) Dose effects of ketamine on photic and acoustic field potentials. Neuropharmacology, 17: 851-862. Dyer, R.S. and Annau, Z. (1977) Carbon monoxide and flash evoked potentials from rat cortex and superior colliculus. Pharmacol. Biochem. Behav., 6: 461-465. Dyer, R.S., Eccles, C.U. and Annau, Z. (1978) Evoked potential alterations following prenatal methyl mercury exposure, Pharmacol. Biochem. Behav., 8: 137-141. Dyer, R.S., Howell, W.E. and MacPhail, R.C. (198l) Dopamine depletion slows retinal transmission. Exp Neurol., 71: 326-340. Foutz, A.S., Dauthier, C. and Kerdelhue, B. (1983) /3-Endorphin plasma levels during neuromuscular blockade in unanesthetised cat.. Brain Res., 263:119-123. Fox, D.A., Lewkowski, J.P. and Cooper, G.P. (1977) Acute and chronic effects of neonatal lead exposure on development of the visual evoked response in rats. Toxicol. Appl. Pharmacol., 40: 449-461. Hetzler, B.E. and Dyer, R.S. (1984) Contribution of hypothermia to effects of chloral hydrate on flash evoked potentials of hooded rats. Pharmacol. Biochem. Behav., 21: 599-607. Hetzler, B.E. and Oakley, K.E. (1981) Dose effects of pentobarbital on evoked potentials in visual cortex and superior colliculus of the albino rat. Neuropharmacology, 20: 969978. Hetzler, B.E., Heilbronner, R.L., Griffin, J. and Griffin, G. (1981) Acute effects of alcohol on evoked potentials in visual cortex and superior colliculus of the rat. Electroenceph. Clin Neurophysiol., 51: 69-79. Hetzler, B.E., Boyes, W.K., Creason, J.P. and Dyer, R.S. (1988) Temperature-dependent ch~inges in visual evoked potentials of rats. Electroenceph. Clin. Neurophysiol., 70: 137-154. LeVere, T.E. (1978) The primary visual system of the rat: a primer of its anatomy. Physiol. Psychol., 6: 142-169. Liiders, H., Lesser, R.P., Dinner, D.S. and Morris, H.H. (1985) Optimizing stimulating and recording parameters in somatosensory evoked potential studies. J. Clin. Neurophysiol., 2: 383-396. Miller, N.E. (1969) Learning of visceral and glandular responses. Science, 163: 434-445.
Montero, V.M. (1973) Evoked responses in the rat's visual cortex to contralateral, ipsilateral and restricted photic stimulation. Brain Res., 53: 192-196. Onofrj, M. and Bodis-Wollner, I. (1982) Dopaminergic deficiency causes delayed visual evoked potentials in rats. Ann. Neurol., 11: 484-490. O'Steen, W.K. and Anderson, K.V. (1971) Photically evoked responses in the visual system of rats exposed to continuous light. Exp. Neurol., 30: 525-534. Petajan, J.H., Packham, S.C., Frens, D.B. and Dinger, B.G. (1976) Sequelae of carbon monoxide-induced hypoxia in the rat. Arch. Neurol., 33: 152-157. Rabe, L.S., Moreno, L., Rigor, B.M. and Dafny, N. (1980) Effects of halothane on evoked field potentials recorded from cortical and subcortical nuclei. Neuropharmacology, 19: 813-825. Schwartzbaum, J.S., Kreinick, C.J. and Gustafson, J.W. (1971) Cortical evoked potentials and behavioral reactivity to photic stimuli in freely moving rats. Brain Res., 27: 295307. Shaw, N.A. (1990) Central auditory conduction time in the rat. Exp. Brain Res., 79: 217-220. Shaw, N.A. and Cant, B.R. (1981) The effect of pentobarbital on central somatosensory conduction time in the rat. Electroenceph. Clin. Neurophysiol., 51:674-677. Shearer, D.E. and Creel, D. (1978) The photically evoked after discharge: current concepts and potential applications. Physiol. Psychol., 6: 369-376. Sohmer, H., Freeman, S. and Malachi, S. (1986) Multi-modality evoked potentials in hypoxaemia. Electroenceph. Clin. Neurophysiol., 64: 328-333. Stanley, O.H., Fleming, P.J. and Morgan, M.H. (1987) Developmental wave form analysis of the neonatal flash evoked potential. Electroenceph. Clin. Neurophysiol., 68: 149-152. Taylor, M.J., Menzies, R., MacMillan, L.J. and Whyte, H.E. (1987) VEPs in normal full-term and premature neonates:longitudinal versus cross-sectional data. Electroenceph. Clin. Neurophysiol., 68: 20-27. Xintaras, C., Johnson, B.K., Ulrich, C.E., Terrill, R.E. and Sobecki, M.F. (1966) Application of the evoked response technique in air pollution toxicology. Toxicol. Appl. Pharmacol., 8: 77-87. Yellin, A.M. and Jerison, H.J. (1973) Visual evoked potentials and inter-stimulus intervals in the rat. Electroenceph. Clin. Neurophysiol., 34: 429-432.
