Binocular Units Research Articles

Indirect, across-the-midline retinotectal projections and representation of ipsilateral visual field in superior colliculus of the cat.

1. In agreement with previous work, we have found that the ipsilateral visual field is represented in an extensive rostral portion--from one-third to one-half--of the superior colliculus (SC) of the cat. This representation is binocular. The SC representation of the ipsilateral visual field can be mediated both directly, by crossed retinotectal connections originating from temporal hemiretina, and indirectly, by across-the-midline connections relaying visual information from one-half of the brain to contralateral SC. 2. In order to study the indirect, across-the-midline visual input to the SC, we have recorded responses of SC neurons to visual stimuli presented to either the ipsilateral or the contralateral eye of cats with a midsagittal splitting of the optic chiasm. Units driven by the ipsilateral eye, presumably through the direct retinotectal input and/or corticotectal connections from ipsilateral visual cortex, were found throughout the SC, except at its caudal pole, which normally receives fibers from the extreme periphery of the contralateral nasal hemiretina. Units driven by the contralateral eye, undoubtedly through an indirect across-the-midline connection, were found only in the anterior portion of the SC, in which is normally represented the ipsilateral visual field. Receptive fields in both ipsilateral and contralateral eye had properties typical of SC receptive fields in cats with intact optic pathways. 3. All units having a receptive field in the contralateral eye had also a receptive field in the ipsilateral eye; for each of these units, the receptive fields in both eyes invariably abutted the vertical meridian of the visual field. The receptive field in one eye had about the same elevation relative to the horizontal meridian and the same vertical extension as the receptive field in the other eye; the two receptive fields of each binocular unit matched each other at the vertical meridian and formed a combined receptive field straddling the vertical midline of the horopter...

Discrimination of orientation and position disparities by binocularly activated neurons in cat straite cortex.

1. We have examined and compared the ability of binocularly activated striate neurons to make both position disparity and orientation disparity discrimination in the anesthetized (N2O/O2) and paralyzed cat preparation. 2. Accurate knowledge of eye position is essential for disparity studies. Using a retinal projection technique able to detect eye drifts of less than 3' arc per retinal landmark and less than 18' arc cyclorotation disparity, we determined eye drift during the course of 2- to 4-day experiments. After the initial eye rotation due to the anesthesia and the onset of paralysis (see below), rotational drift thereafter was mainly excyclorotatory and, from all causes, rarely totaled more than 4 degrees disparity. All our data have been corrected for this residual cyclorotatory drift. 3. Optimal stimulus orientation disparities were determined from quantitative monocular orientation tuning curves for 74 binocularly activated striate cells (37 simple, 3 hypercomplex I, 31 complex, 3 hypercomplex II) from nine cats. Without exception, the mean optimal stimulus orientation disparity in each of our animals showed a departure from zero disparity equivalent to an incyclorotation of the eyes (mean, 9.2 degrees; range, 2.7 degrees-15.9 degrees). 4. We attribute this mean optimal stimulus orientation disparity shift to ocular cyclorotation as a result of the initial anesthesia and paralysis. Assuming equal intortion, incyclorotation for each eye averages 4.6 degrees. On the assumption that the mean optimal stimulus orientation disparity is zero in normal life, we pooled results from the nine animals about their individual means. For the 74 cells the resulting distribution of the optimal stimulus orientation disparities had a range of about +/-15 degrees (simple cells: SD 4.9 degrees; complex cells: SD 7.4 degrees). 5. We examined the relationship of the sharpness of the orientation tuning curves to ocular dominance, to absolute orientation preference, and to other unit properties. The striking observation was the high correlation between the sharpness of orientation tuning curves for the two eyes of a binocular neuron. For simple cells the mean difference for the half-widths of half-height was only 2.54 degrees, with sharpness showing a high correlation between the two eyes (r=0.915) over half-width at half-heights ranging from 8.5 degrees to 41.8 degrees. Complex cells showed a similar, albeit weaker, correlation. 6. Having shown that, assessed monocularly binocular units show different orientation tunings in the two eyes, we undertook binocular experiments to ascertain if these differences were the optimal disparities of sharply tuned stimulus orientation disparity channels. Using a matrix stimulation paradigm to minimize the effects of spontaneous changes in responsiveness, we have simultaneously extracted bionocular stimulus orientation disparity and position disparity tuning curves from single striate neurons...

The conditions required for the maintenance of binocularity in the kitten's visual cortex.

1. In young kittens, cortical neurones, which are usually binocularly driven, have their binocularity reduced if one eye is covered, or if the eyes are made strabismic or alternately occluded. Some of the factors causing these changes were analysed. 2. If the contrast of one retinal image is abolished with no difference in mean illumination, the input from that eye is virtually lost. 3. If one eye merely has its mean retinal illumination attenuated, that eye does not specifically lose its influence in the cortex, although there is a reduction in the proportion of binocular units. This change might partly be due to a difference in the timing of signals from the two eyes but is more likely to be caused by a difference in the strength of the discharges. 4. There is little change in binocularity if one image is dimmed but contrast is absent from both. 5. If contours of very different orientation fall simultaneously on corresponding retinal regions, binocularity breaks down, as in the case of strabismus or when different patterns are presented to the two eyes. But as long as the patterns on corresponding retinal points have similar orientation, even if the visual axes are misaligned, binocularity can be maintained. 6. If the eyes are not stimulated simultaneously, binocularity is reduced, even if the contours falling on the two retinae (at different times) are identical. 7. Roughly simultaneous stimulation, with roughly congruent patterns on the two receptive fields, are needed for the upkeep of binocular connexions on to cortical cells.

Reversal of the physiological effects of monocular deprivation in the kitten's visual cortex.

1. Twenty-three kittens were monocularly deprived of vision until the age of 4, 5, 6 or 7 weeks. Their deprived eyes were then opened, and their experienced eyes shut for a further 3-63 days. After this time physiological recordings were made in the visual cortex, area 17. Three control kittens, monocularly deprived for various periods, showed that at the time of reverse-suturing, few neurones could be influenced at all from the deprived eye. 2. Following reverse-suturing, the initially deprived eye regained control of cortical neurones. This switch of cortical ocular dominance was most rapid following reverse-suturing at the age of 4 weeks. Delaying the age of reverse-suturing reduced the rate and then the extent of the cortical ocular dominance changes. 3. The cortex of reverse-sutured kittens is divided into regions of cells dominated by one eye or the other. The relative sizes of these ocular dominance columns changed during reversed deprivation. The columns devoted to the initially deprived eye were very small in animals reverse-sutured for brief periods, but in animals that underwent longer periods of reversed deprivation, the columns driven by that eye were larger, while those devoted to the initially open eye were smaller. 4. Clear progressions of orientation columns across the cortex were apparent in many of the kittens, but, in contrast to the situation in normal or strabismic kittens, these sequences were disrupted at the borders of eye dominance columns: the cortical representations of orientation and ocular dominance were not independent. 5. Binocular units in these kittens were rather rare, but those that could be found often had dissimilar receptive field properties in the two eyes. Commonly, a cell would have a normal orientation selective receptive field in one eye, and an immature, unselective receptive field in the other. Cells that had orientation selective receptive fields in both eyes often had greatly differing orientation preferences in the two eyes, occasionally by nearly 90 degrees. 6. During the reversal of deprivation effects, the proportion of receptive fields exhibiting mature properties declined in the initially experienced eye, while the proportion increased in the initially deprived eye. Similarly, the average band width of orientation tuning of receptive fields in the initially deprived eye decreased, while that of receptive fields in the initially experienced eye increased. 7. One kitten was reverse-sutured twice, to demonstrate that cortical ocular dominance may be reversed a second time, even after one reversal of ocular dominance. 8. It is suggested that the sensitive period for cortical binocular development consists of two phases. In the first phase, all cortical neurones may be modified by experience, but the rate at which they may be modified decreases with age. In the second phase, an increasing number of cortical neurones becomes fixed in their properties, while those that remain modifiable are as modifiable as they were at the end of the first phase. 9...

The organization of visual processing in the pigeon cerebellum.

1. Visual responses were sought in the cerebella of decerebrate pigeons using extracellular micro-electrodes, and were found in folia VIc-IXb, especially folia VII and VIII. The responses were mainly, but not exclusively, from the ipsilateral eye. Four binocular units were recorded.2. In the anterior and posterior walls of folium VII the organization was clearly, though rather crudely, retinotopic. The temporal field was represented laterally, on the ipsilateral side, and the nasal field medially; the superior field was represented superficially, and the inferior field towards the base of the folium. In the lateral wall of folium VII there was a small anomalous region innervated by the contralateral eye.3. The visual input arrived via the mossy fibre system.4. Units exhibited a strong preference for moving targets, 20-60 degrees /sec being the range of optimal speeds. About three quarters of the units responded most strongly to a particular direction of target motion. The preferred direction was most frequently upwards or backwards.5. Units in the granule layer were to some extent clustered according to their direction-preference, which tended to change gradually as the electrode advanced along the granule layer.6. Units were classified as Gr-units (granule cells or mossy fibre rosettes) or P-units (Purkinje cells) or were left unclassified. Receptive fields of Gr-units were usually 5-30 degrees across; they were much larger for P-units in folium VIII, but not obviously so for those in folium VII. Gr-units were more frequently direction-sensitive than were P-units.

Input characteristics of the intrinsic cells of the optic tectum of teleost fish

Cells of the optic tectum were studied by electrical recording methods in the goldfish and some other species. New findings were: 1. 1. Binocular units. 2. 2. Wide field units responding to movement over 180° of arc. 3. 3. Strongly inhibitory and excitatory effects that could be driven from the hindbrain. 4. 4. A marked focus of spontaneous cells in the goldfish in the area associated with forward vision. 5. 5. It was suggested that Sutteriin and Prosser's Type I and Type II units. may be part of a continuous range of spontaneous units.

Binocular units in the lateral geniculate nucleus of chronic cats

Summary The effect of monocular and binocular stimulation on cells in the lateral geniculate nucleus was investigated in awake cats with painless head fixation. The cells were stimulated by moving visual patterns presented on a tangent screen. The features of the behavior of the cells were a relatively high level of spontaneous activity and a vigorous response to the moving patterns without any evidence of directional selectivity. Typically, the cells showed either on-, off- or on-off-responses to stimulation with diffuse light. When both eyes were stimulated by the moving patterns, the response varied from cell to cell. According to the form of the response, the cells were classified into 8 types. Depending on receptive field organization and balance of the center-surround influences, responses were either purely excitatory or inhibitory, or combinations of excitation and inhibition. Binocular response was compared with each of two monocular responses in 406 cells. The response was obtained exclusively by stimulation of one eye in 222 cells (54.7%). Binocular interaction occurred in the remaining 184 cells (45.3%). Of these, 85 cells (20.9%) were excited binocularly and various types of binocular interaction, either facilitation, summation or occlusion , occurred. In 70 cells (17.2%), excitation occurred upon stimulation of one eye and inhibition occurred upon stimulation of the other eye, and the response from the dominant eye was markedly suppressed when both eyes were stimulated simultaneously. In 16 cells (3.9%) no response occurred with monocular stimulation of each of the two eyes, but a clear response occurred upon binocular stimulation. Nine cells (2.3%) were activated monocularl, but the input from the non-dominant eye gave a facilitatory influence upon the response and the binocular response was markedly larger than the monocular response. The non-dominant eye gave an inhibitory influence in 4 cells (0.9%). The binocular response was smaller than the dominant eye response. Binocularly activated cells tended to be located in two areas. One area is in the dorsal part of layer A and the other area is below layer A 1 . On the other hand, the cells that were excited by stimulation of one eye and inhibited by stimulation of the other were located predominantly in the two main dorsal layers, A and A 1 .

The representation of three-dimensional visual space in the cat's striate cortex.

1. Binocularly driven single units were recorded in the cat's striate cortex. For each neurone the two monocular receptive fields were stimulated simultaneously in order to assess the optimal positioning of the image in both eyes to give the best binocular response.2. The electrode was driven perpendicular to the surface of the brain to explore cortical columns, all the cells of which are known to have the same preferred target orientation.3. All orientation columns were found to fit into one of two classes according to their binocular organization.4. In a constant depth column the receptive fields of binocular neurones cover a small retinal area and they are laid out in almost identical arrays in the two eyes. Consequently, the horizontal disparity is practically the same for all the units. The depth column as a whole is viewing a thin sheet of visual space, a few degrees wide, floating at some distance from the cat. There may be about 0.6 degrees disparity difference between neighbouring depth columns.5. In a constant direction column the binocular units' fields are all super-imposed on the retina contralateral to the hemisphere containing the column. In the ipsilateral eye they are more scattered horizontally. Therefore the horizontal disparity varies enormously from cell to cell and the column as a whole is viewing a cylinder of visual space directed towards the contralateral eye. Neighbouring direction columns may vary by about 4 degrees in their oculocentric visual direction.6. This columnar arrangement is probably important for space perception in the cat. Activity in only one depth and one direction column would specify the orientation and the three-dimensional locus of an object in space.7. The two types of column may be involved in the control of disjunctive and conjugate eye movements.

Electrophysiological correlate of binocular depth perception in man.

IT is known that within the visual pathway of the brains of anaesthetized animals single units can be found which respond optimally when both eyes are simultaneously stimulated with light1–3. The characteristics of such binocular units are consistent with psychophysically observed features of binocular depth perception4–6. At present, human subjects have important advantages over animals in studies of stereopsis because with human subjects the critical binocular fusion and subjective reports of depth perception can easily be obtained. With human subjects, however, any electrophysiological information must be obtained by the analysis of scalp evoked potentials (EPs). This communication reports the finding of an EP correlate of binocular depth perception in man.

Single-unit analysis of binocular neurons in the frog optic tectum

Responses of single, postsynaptic neurons in the deeper cellular layers of the frog optic tectum were recorded with glass-coated metal microelectrodes. Particular emphasis was placed upon the analysis of input and response properties of binocularly driven neurons. Section of the posterior and tectal commissures did not affect binocular response activity. However, removal of the contralateral tectal lobe or section of the deep tegmental commissures eliminated the ipsilateral component of binocular unit activity. The pathway of visual information from the ipsilateral retina is probably via the contralateral tectum, through the tegmental commissures to the ipsilateral tectum. The majority of tectal neurons observed could be activated by visual, tactile, and occasionally, auditory stimuli. Such multisensory units showed a spontaneous level of activity, whereas purely visual neurons did not. Visual receptive fields of binocular neurons were larger than those of monocular neurons in the intact preparation. Tectal neurons responded only to moving stimulus targets and often showed rapid adaptation to a repeated stimulus. However, some units showed little stimulus adaptation and the majority of units observed fell between those two extremes. Tectal neurons showed a greater response to vertical than to horizontal stimulus motion, which may be related to the predominance of vertical eye movements in the frog.

Micro-electrode studies of the medial protocerebrum of some Lepidoptera—II. Responses to visual flicker

The compound eyes of Sphinx and of saturniid moths were stimulated with rectangular light pulses whose duration and repetition rate could be varied independently. The activity of single neurons in the medial protocerebrum and the third synaptic region of the optic lobes was monitored with steel micro-electrodes. The neurons included three categories of L-unit, and fast dark-units. The following results were obtained: (1) L-units may be subdivided into those whose spikes follow flicker by regrouping, while the mean frequency remains constant at all stimulus repetition rates (s.r.r.) (Type 2 L-units), and those whose mean frequency increases with the s.r.r. up to a maximum at or approaching the flicker fusion frequency (f.f.f.) and thereafter declines (Type 1 L-units and rhythmic units). Fast dark-units have a maximum mean spike frequency at the lower s.r.r.'s. (2) For all classes of unit, increase in s.r.r. results in a decline in the mean number of spikes/response. (3) L-units suffer a progressive latency increase with rise in s.r.r. above a variable threshold rate, and this shift is linear or nearly so when plotted against interflash dark interval. The latencies of fast dark-units may be either fixed or labile in relation to s.r.r. (4) The latencies of L-units relate inversely to stimulus brightness; those of fast dark-units also relate inversely to flash length. (5) A variety of techniques, including double-flash experiments, and observations on Type A binocular units have shown that the latency shifts and the reduction in mean spike frequency can be accounted for in terms of two processes: (a) The active inhibition of each ‘on’-response by the ‘off’-component of the preceding stimulus and (b) adaptation; the possible role of a third process (c) facilitation, is less certain. (6) The ‘off’-response of a fast dark-unit is partially inhibited by a proximate ‘on’-component. (7) During trains of flashes, the number of spikes response given by L-units rapidly diminishes, while that given by fast dark units sometimes increases. The functional arrangement of ‘on’- and ‘off’-sensitive systems is compared with some of the properties of cat retinal ganglion cells, whose responses to visual flicker are closely similar.

Micro-electrode studies of the medial protocerebrum of some Lepidoptera—I. Responses to simple, binocular visual stimulation

The responses of single neurons in the medial protocerebral lobes of Sphinx and of several species of saturniid moths have been studied with extracellular, metal micro-electrodes, using binocular, rectangular light pulses as stimuli. Most categories of neuron respond to illumination of the eyes, but one type is strongly inhibited by light and responds to ‘off’ with a strong discharge. These are termed fast dark units. The main categories of binocular unit give: (A) Positive responses to the ‘on’-components of the stimuli to both eyes; (B) Positive responses to the ‘on’-components to both eyes, but with a strong inhibitory action of the ‘off’-component of one of them; (C) Positive responses to the ‘on’-component to one eye and strong inhibition to the ‘on’-component to the contralateral eye. Thus, flashes of equal duration and approximately matched intensity, exactly superimposed, may cancel the response; (D) Positive responses to the ‘on’-component to one eye and weak inhibition to the contralateral ‘on’-component. Thus, in this case the resting discharge alone may be blocked. There are also other patterns of interaction, less frequently encountered, which have not been analysed. The responses of 67 individual binocular neurons have been classified. The inhibition of a neuron may last for the duration of a brief stimulus, or it may persist beyond it. In Sphinx the effects on a Type B neuron of a 2 msec light flash with an inhibitory ‘off’-component can be detected 2–3 sec after the stimulus, and immediate blockage of the response to a 1 sec ipsilateral flash may last for 300–400 msec. Although sampling error cannot be excluded as a source of bias, there seem to be differences in the distribution of types of neuron between species. Type A and B neurons predominate in Sphinx, whereas Type C and D units predominate in the Saturniidae. Evidence is discussed which suggests that visual responses are relayed to the ventral nerve cord via both ipsilateral and contralateral pathways which include synapses in the medial regions of the protocerebrum.