Journal of Neurophysiology

Kirby, M. A.; Wilson, P. D.

doi: N/Apmid: 3097274

Abstract Relay cells in the lateral geniculate nucleus (LGN) of the North American opossum were classified as types 1, 2, and 3 on the basis of their receptive field properties and afferent latencies to optic nerve (ON) and optic chiasm (OX) stimulation and antidromic latencies to stimulation of cortex. Type 1 and type 3 cells gave transient responses and type 2 cells gave sustained responses to appropriate standing contrast in the receptive field center. The differences in response pattern were quantified with a phasic-tonic index (PTI); the PTI values for type 2 cells (PTI less than 63) did not overlap those for type 1 cells (PTI greater than 68) or type 3 cells (PTI greater than 80). With a homogeneous field (1.3 cd/m2), the maintained discharge rates (spikes/s) of type 3 cells (less than 1-11) were significantly lower than those of type 1 cells (3-23) and of type 2 cells (1-22). For all type 2 cells tested with a counterphased sine-wave grating, a null position of the grating was found and the cells were classified as linear. The type 1 and type 3 cells tested were nonlinear (i.e., exhibited excitatory doubling and did not have a null grating position). The maximum velocity of movement that reliably elicited responses (cut-off velocity) was low for type 3 cells (mean = 29.1 degrees/s) and relatively high for type 1 cells (mean = 70.3 degrees/s). Cut-off velocities for type 2 cells (mean = 61.2 degrees/s) were slightly lower than for type 1 cells. Type 1 cells had relatively short afferent (ON and OX) latencies, fast afferent conduction velocities, and short antidromic (cortex) latencies; type 2 cells had intermediate afferent and antidromic latencies and intermediate afferent conduction velocities; and type 3 cells had relatively long afferent and antidromic latencies and slow afferent conduction velocities. The receptive field center diameters were similar for type 1 (4.2-25.5 degrees; mean = 11.4 degrees) and type 3 cells (2.8-23.7 degrees; mean = 11.0 degrees), whereas the receptive field centers for type 2 cells (2.9-15.1 degrees; mean = 6.9 degrees) were significantly smaller. The majority of type 1 (66.7%) and type 3 cells (63.3%) had on-center receptive fields, whereas the proportion of on-center fields was even greater for type 2 cells (83.0%). Only a few of the cells encountered in the opossum LGN (6%) had on-off receptive fields, and a portion of these could not be shown to be relay cells.(ABSTRACT TRUNCATED AT 400 WORDS) Copyright © 1986 the American Physiological Society

Malpeli, J. G.; Lee, C.; Schwark, H. D.; Weyand, T. G.

doi: N/Apmid: 3783229

Abstract Reversible inactivation of individual layers of the cat lateral geniculate and medial interlaminar nuclei was used to investigate the necessary and sufficient inputs for maintaining visually driven activity and receptive field properties in area 17. Neither orientation selectivity nor direction selectivity depends on any individual geniculate layer. We identified two groups of cortical layers on the basis of the pattern of thalamic inputs providing visual driving through the contralateral eye. One group, consisting of layers 4 and 6, has geniculate layer A as its only necessary and sufficient input. The other, consisting of supragranular layers, integrates at least two sufficient thalamic inputs, one of which is layer A. Several major receptive field properties are independently generated in these two groups of layers. Copyright © 1986 the American Physiological Society

Mays, L. E.; Porter, J. D.; Gamlin, P. D.; Tello, C. A.

doi: N/Apmid: 3783225

Abstract Single-unit recordings were made from midbrain areas in monkeys trained to make both conjugate and disjunctive (vergence) eye movements. Previous work had identified cells with a firing rate proportional to the vergence angle, without regard to the direction of conjugate gaze. The present study describes the activity of neurons that burst for disjunctive eye movements. Convergence burst cells display a discrete burst of activity just before and during convergence eye movements. For most of these cells, the profile of the burst is correlated with instantaneous vergence velocity and the number of spikes in the burst is correlated with the size of the vergence movement. Some of these cells also have a tonic firing rate that is positively correlated with vergence angle (convergence burst-tonic cells). Divergence burst cells have similar properties, except that they fire for divergent and not convergent movements. Divergence burst cells are encountered far less often than convergence burst cells. Both convergence and divergence burst cells were found in an area of the mesencephalic reticular formation just dorsal and lateral to the oculomotor nucleus. Convergence burst cells were also recorded in another more dorsal mesencephalic region, rostral to the superior colliculus. Both of the areas also contain cells that encode vergence angle. Models of the vergence system derived from psychophysical data imply the existence of a vergence integrator, the output of which is vergence angle. Some models also suggest the presence of a parallel element that improves the frequency response of the vergence system, but has no effect on the steady-state behavior of the system. Vergence burst cells would be suitable inputs to a vergence integrator. By providing a vergence velocity signal to motoneurons, they may improve the dynamic response of the vergence system. The behavior of vergence burst cells during vergence movements is similar to that of the medium-lead burst cells during saccades. The proposed roles for vergence velocity cells are analogous to those of the saccadic burst cells. In this respect, the neural organization of the vergence system resembles that of the saccadic system, despite the distinct difference in the kinematics of these two types of eye movements. Copyright © 1986 the American Physiological Society

Rudomin, P.; Solodkin, M.; Jimenez, I.

doi: N/Apmid: 3783240

Abstract The characteristics of the primary afferent depolarization (PAD) of Ia- and Ib-fibers generated by segmental and descending inputs have been analyzed in the spinal cord of anesthetized cats. The PAD was inferred from the changes produced by conditioning inputs on the intraspinal stimulus current required to produce a constant antidromic firing of single group I afferent fibers from the gastrocnemius (GS) or posterior biceps and semitendinosus (PBSt) nerves. Group I GS and PBSt fibers ending in the intermediate nucleus could be classified in three different types according to their PAD patterns in response to stimulation of cutaneous nerves and of descending fibers. In one set of group I fibers stimulation of cutaneous nerves and of the ipsilateral brain stem reticular formation, or the contralateral red nucleus, produced no PAD, but was able to inhibit the PAD generated by stimulation of group I fibers from flexors (type A PAD pattern). PBSt nerve fibers with this PAD pattern had peripheral thresholds and conduction velocities between 1.01 and 1.56 times threshold and 76.3 to 118 m/s, respectively. A second set of group I fibers was found to be depolarized by cutaneous nerves as well as by stimulation of rubrospinal and reticulospinal fibers (type B PAD pattern). The peripheral thresholds and conduction velocities of PBSt afferent fibers with a type B PAD pattern were of 1.66-2.03 times threshold and 71-83 m/s, respectively. We found a third set of group I fibers that were also depolarized by reticulospinal and rubrospinal inputs, but not by cutaneous nerves that instead inhibited the PAD elicited by group I volleys in flexor nerves (type C PAD pattern). All PBSt afferent fibers with a type C PAD pattern, with the exception of two, had peripheral thresholds and velocities between 1.46 and 2.16 times threshold and between 72 and 89 m/s, respectively. Stimulation of the Deiter's nucleus was found to depolarize the intraspinal terminals of a small fraction of group I GS fibers with a type A PAD pattern and of all group I GS and PBSt fibers with type B and C PAD patterns. The PAD produced by vestibulospinal stimulation in fibers with type A and C PAD patterns could be inhibited by conditioning volleys applied to cutaneous nerves. It is suggested that group I afferent fibers from flexors and extensors with a type A PAD pattern are group Ia, and that most fibers with type B and type C PAD patterns are group Ib.(ABSTRACT TRUNCATED AT 400 WORDS) Copyright © 1986 the American Physiological Society

Meyers, D. E.; Snow, P. J.

doi: N/Apmid: 3783226

Abstract Single-unit spike-triggered averaging has been used to study the distribution of activity in the central processes of 11 single hair follicle afferent (HFA) fibers in relation to the somatotopic organization of dorsal horn neurons (DHNs). Central responses could be recorded from all but one HFA, and the waveforms of these responses were similar to the biphasic, monophasic positive and compound terminal potentials (TPs), the triphasic, positive-negative-positive axonal (preterminal) potentials (APs), and the focal synaptic potentials (FSPs) described by other workers in different preparations. No central responses could be recorded from one HFA even though this axon was shown to be intact throughout the experiment and noise levels in the averaged records were below 3.1 microV. The spatial organization of TPs and APs mirrored the anatomical organization determined using intraaxonal staining only in that the region containing these potentials was a longitudinally orientated narrow (mean width = 405 microns) strip of dorsal horn. Within this strip large TPs, APs, and FSPs were usually found only in those regions in which the receptive field (RF) of the HFA was relatively central to the RFs of DHNs. The region in which the RFs of DHNs encompassed the RF of the HFA, the somatotopically appropriate region, was also organized into a longitudinally orientated strip of dorsal horn of approximately the same width as the strip containing TPs and APs. In any single experiment the strips formed by the somatotopically appropriate region and the TP-AP region occupied the same mediolateral position, but in contrast, the rostral and/or caudal boundaries of these strips often occurred at different levels along the dorsal horn. In some cases the TP-AP strip extended rostrally and/or caudally beyond the somatotopically appropriate region or ended at the same rostrocaudal level. In other cases the somatotopically appropriate region extended rostrally and/or caudally beyond the TP-AP strip. These results are discussed in relation to the rostrocaudal spread of the dendritic trees and the RF organization of DHNs. The results show clearly that in intact cats, anesthetized with alpha-chloralose, some HFAs give rise to collaterals in somatotopically inappropriate regions of the dorsal horn and that at least some parts of these collaterals are invaded by incoming action potentials. The question of whether some HFAs give rise to collaterals that are either infrequently invaded or not invaded at all is discussed. Copyright © 1986 the American Physiological Society

Vaadia, E.; Benson, D. A.; Hienz, R. D.; Goldstein, M. H.

doi: N/Apmid: 3783237

Abstract The influence of sound localization behavior on unit activity in the frontal cortex of awake rhesus monkeys was examined by comparing responses under three behavioral conditions: auditory localization, during which a response was required to the location of a sound (broad-band noise) source; auditory detect, during which a response was required to indicate the occurrence of the sound regardless of location; visual localization, during which no sounds were presented and a response was required to the location of a visual stimulus; and nonperform, presentation of auditory stimuli as in the first two conditions, but with the animal sitting passively. Extracellular microelectrode recordings were made in the periarcuate region and dorsal and ventral prefrontal areas near the principal sulcus. Four monkeys were used with a total of 498 cells studied. Of the total population, only five cells were found to have characteristics similar to those of auditory units in the primary auditory cortex and the surrounding belt area. More typically, units were found that had strong short-latency responses specific to the auditory and/or visual localization tasks. These units had no or weak responses when the same sound stimuli were presented in the auditory detect task or when a monkey received the sound stimuli in a nonperforming condition. Two regions were identified, one medial and/or posterior to the arcuate sulcus, in Brodmann's area 6; the second included parts of areas 8 and 9 within the genu of the arcuate sulcus. Units from these regions are referred to, respectively, as the postarcuate and the prearcuate populations. Both populations responded predominantly during active localization behavior. Sixty-two percent of the postarcuate population responded during auditory localization, 32% responded during auditory detect, and only 18% responded to acoustic stimuli presented in the nonperforming condition. In the prearcuate population percentages in these three conditions were 35, 25, and 12%, respectively. For visual localization, 54% in the postarcuate population responded, whereas 42% in the prearcuate responded. Spatial tuning of units during auditory localization was similar to that seen in units of the primary auditory cortex, with the greatest percentages of units responding to stimuli contralateral to the recording site. Similar tuning was observed for the visual localization task as well. Similarities in spatial tuning between the auditory and visual localization conditions were examined to assess the "bimodal" nature of the units.(ABSTRACT TRUNCATED AT 400 WORDS) Copyright © 1986 the American Physiological Society

Sieving, P. A.; Frishman, L. J.; Steinberg, R. H.

doi: N/Apmid: 3783228

Abstract We describe a new response in proximal retina of cat that is present under scotopic conditions, clearly differs from PII (b-wave and DC component) and contributes a negative potential at threshold to the diffuse electroretinogram (ERG). We have termed this response the scotopic threshold response (STR). Extracellular potentials evoked in response to circular spots of light at dark-adapted threshold, and with dim backgrounds, were recorded with microelectrodes placed intraretinally at different depths. The dark-adapted response of proximal retina (STR) consisted of a graded negative potential to the onset of illumination that maintained amplitude during illumination and decayed back toward the base line at stimulus offset without evidence of a negative-going off response. It thereby differed in form from the photopic M-wave response of proximal retina, which has a negative-going off response. It also did not exhibit spatial tuning, simply increasing in size with stimulus area. In addition, the STR appears to be a rod-driven response whose threshold approximates that of the most sensitive ganglion cells in cat, whereas the M-wave is a much higher threshold cone-driven response. The STR could be clearly distinguished from PII on the basis of its form, depth-distribution, and dynamic range. For example, the STR had its maximum amplitude in the proximal retina at 17% retinal depth, whereas scotopic PII had its maximum in the distal retina at 48% retinal depth. Also, the STR had a lower threshold than PII intraretinally and saturated well below the level of saturation of scotopic PII (rod saturation). By analogy to the PNR and M-wave, the STR is hypothesized to represent either an extracellular voltage arising from proximal retinal neurons or Muller cell responses to K+ released by these neurons. Recording in the vitreous, near the retinal surface, showed that the STR always had a negative polarity. The polarity reversal of the STR at 50-60% retinal depth (from negative, proximal to positive, distal) suggested the presence of a sink proximal to the reversal point and a source distal to it. We also recorded the vitreal ERG in response to diffuse illumination of the dark-adapted retina. The STR could be clearly identified in the scotopic ERG as a threshold negative potential that had been observed previously in the mammalian ERG. The STR differs, therefore, from PII (b-wave and DC component) that is a higher threshold positive component in the diffuse ERG.(ABSTRACT TRUNCATED AT 400 WORDS) Copyright © 1986 the American Physiological Society

Crabtree, J. W.; Spear, P. D.; McCall, M. A.; Jones, K. R.; Kornguth, S. E.

doi: N/Apmid: 3783234

Abstract The cat's superior colliculus (SC) receives direct inputs from retinal W-cells (a W-D input) and Y-cells (Y-D input) and an indirect Y-cell input via the lateral geniculate nucleus and visual cortex (Y-I input). In previous studies we have shown that intraocular injection of antibodies raised against large retinal ganglion cells produces a dose-dependent reduction in the Y retinogeniculate pathway. Furthermore, when a sufficiently high antibody concentration is used, there is a substantial loss of the Y pathway and no apparent loss of the W pathway. In the present study, we used the antibodies to investigate the contributions of the Y and W pathways to functional organization within the SC. Binocular injections of low (330 micrograms/100 microliters) or high (1,000 micrograms/100 microliters) antibody concentrations were made. The antibody-mediated effects on SC cells' response properties were compared directly with effects of early binocular deprivation, which have been attributed to a loss of Y-I input. Extracellular single-cell recordings were made from the SC, and cells were classified as receiving Y-D, Y-I, or W-D inputs on the basis of their response latencies to electrical stimulation of the optic chiasm and optic tract. Injections of the low antibody concentration produced no significant effects on inputs to the SC. However, injections of the high antibody concentration resulted in a 70% reduction in SC cells with a Y-D input and an 82% reduction in SC cells with a Y-I input. There was no effect on the percentage of cells with a W-D input. Binocular deprivation produced a 76% reduction in the percentage of cells with Y-I input. Visual response properties of SC cells also were assessed. Injections of the high antibody concentration produced a 55% reduction in cells that respond with a directional preference and a 51% reduction in cells that respond to high-velocity stimuli. Binocular deprivation produced a 78% reduction in the proportion of directional cells and a 25% reduction in cells that respond to the ipsilateral eye. Taken together, the results of this and previous studies using cortical lesions, visual deprivation, and immunoablation suggest that Y-D input is the primary basis for responses to high stimulus velocity, Y-I input is an important basis for directional responses and response through the ipsilateral eye, and W-D input is important for responses to low stimulus velocity.(ABSTRACT TRUNCATED AT 400 WORDS) Copyright © 1986 the American Physiological Society

Sieving, P. A.; Frishman, L. J.; Steinberg, R. H.

doi: N/Apmid: 3783227

Abstract There has been relatively little known about responses from proximal retina in mammals that could contribute to the electroretinogram (ERG). Recently, there has been evidence that the proximal retina is involved in generating the pattern electroretinogram (PERG). In the present work we investigated proximal retinal activity in the intact cat eye during light adaptation. Extracellular potentials evoked in response to circular spots of light, flashed on steady backgrounds, were recorded with microelectrodes placed intraretinally at different depths. Prominent negative responses were found in proximal retina that could be identified as the M-wave previously observed only in cold-blooded retinas. Like the cold-blooded responses, the cat's M-wave consisted of negative-going potentials at stimulus onset and offset that were maximum in amplitude with small spots. By analogy to the cold-blooded data, the cat M-wave is presumed to be the extracellular voltage arising from Muller cell responses to K+ released by proximal retinal neurons. In addition, the cat M-wave only appeared with backgrounds at and above rod saturation and had short latencies (30 ms) at stimulus onset and offset, indicating that it is a cone-driven response. The M-wave could be clearly distinguished from PII (b-wave and DC component) on the basis of its form, depth distribution, and stimulus-response characteristics. For example, photopic PII had its maximum voltage in the distal retinal at 55% retinal depth, whereas the M-wave was maximal in the proximal retina at 25% retinal depth. Also, PII simply increased in amplitude as stimulus spots were enlarged, whereas the M-wave exhibited spatial tuning. Under light-adapted conditions and with small-spot stimuli the M-wave is the largest extracellular voltage in cat retina. By recording the vitreal ERG near the retinal surface with the microelectrode referenced to a silver wire in the vitreous, we found that the M-wave in response to a small spot always had a negative polarity in the vitreous. Thus, unlike PII, the M-wave does not reverse polarity at the vitreo-retinal border. Because of stray-light effects, however, we were not able to assess the amplitude of the M-wave's contribution to the ERG obtained with diffuse retinal illumination. We conclude that the M-wave is present in the cat as a prominent cone-driven response of proximal retina that is separate from the b-wave, and whose significance for electroretinographic recordings remains to be determined. Copyright © 1986 the American Physiological Society

Gustafsson, B.; Pinter, M. J.; Wigstrom, H.

doi: N/Apmid: 3023562

Abstract Posttetanic potentiation (PTP) of composite Ia excitatory postsynaptic potentials (EPSPs) has been studied in normal cat alpha-motoneurons and in motoneurons axotomized 2-3 wk earlier by ventral root section. The maximal amount of PTP of EPSP amplitude (expressed relative to unpotentiated amplitude) was considerably less in the axotomized population compared with the normal population. The decrease in PTP provoked by axotomy occurs in association with a postaxotomy increase of input resistance, the net effect being that PTP in axotomized cells was much the same as that observed by others in normal motoneurons possessing similarly high input resistance. In agreement with previous results, EPSP peak amplitudes were decreased after axotomy. This decrease seemed to be largely related to an absence of the largest EPSPs, since otherwise the EPSP distributions of normal and axotomized motoneurons showed considerable overlap. It is suggested that the observed decrease in PTP after axotomy is related to a change in synaptic release properties and not secondary to changes in the electrical properties of motoneurons. A previous analysis has suggested that axotomy causes an alteration of the distribution of passive electrical properties among motoneurons such that axotomized cells resemble normal high-resistance motoneurons. The present results suggest that axotomy may affect the distribution of Ia synaptic release properties in a similar manner, since PTP in axotomized motoneurons resembles that observed in normal high-resistance motoneurons. Copyright © 1986 the American Physiological Society