The period of susceptibility for binocular depth vision was studied in kittens by subjecting them to periods of monocular deprivation beginning at different ages. In an initial study, we found that normally reared kittens can learn a depth-discrimination task much more rapidly when tested binocularly than monocularly, even when testing is begun as early at 30 d. In subsequent experiments, kittens were monocularly deprived by eyelid suture, following which their monocular and binocular depth thresholds were measured using the jumping-stand procedure. We obtained the following results: (1) When monocular deprivation is applied before the time of natural eye opening but is discontinued by no later than 30 d, there is very Little effect on binocular depth thresholds. (2) When deprivation is begun at 90 d, binocular depth thresholds are unaffected. (3) When deprivation is begun between these two ages, the magnitude of the deficit varies with the period of deprivation and the age at which it begins. (4) By imposing brief (5 or 10 d) periods of deprivation, beginning at different ages, we were able to demonstrate that the peak of the sensitive period is between the ages of 35 and 45 d, with a fairly rapid decline in susceptibility outside those age limits. (5) Even with as little as 5 d of deprivation, substantial permanent deficits in binocular depth vision can be induced.

The cellular connectivity of the lacertilian parietal eye is not well understood. Because the intercellular connections establish the foundation for information processing, we have investigated cellular connectivity of one cell type in this simple vertebrate retina. We also developed an in vitro preparation to study the anatomy of the parietal eye visual system. Horseradish peroxidase transport in the in vitro preparation revealed a class of displaced ganglion cells occupying positions among the photoreceptors, in a location where the presence of interneurons had been suggested. Three-dimensional reconstruction at the electron-microscopic level showed that the morphology and synaptic input of these displaced ganglion cells is different from that of the previously known ganglion cells. The displaced ganglion cells receive an average of about 13 ribbon synapses from photoreceptors. The ribbon input is equally distributed between the soma and dendritic arbor. Junctional membrane measurement and ethnolic phosphotungstic acid-staining provided evidence for the existence of non-ribbon synaptic contacts (synaptoid junctions). Displaced ganglion cells make about 20 synaptoid junctions, 65% of which are on the dendritic arbor. The morphology of the displaced ganglion cell is such that a significant measure of synaptic input to the dendritic arbor will be transmitted to the soma.

Actions of antagonists for excitatory amino-acid (EAA) receptors on extracellularly and intracellularly recorded responses of layer II/III cells to electrical stimulation of the underlying white matter were studied in a slice preparation of rat's visual cortex. Antagonists used were 2-amino-5-phosphonovalerate (APV) and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), which are selective antagonists for EAA receptors of N-methyl-D-aspartate (NMDA) and quisqualate/kainate (non-NMDA) type, respectively. In extracellular recordings, it was found that responses of almost all of the cells were suppressed by CNQX. In contrast, sensitivity to APV was different between cells with short- and long-latency responses; 81% of the former responses were not suppressed by APV, while about a half of the latter were suppressed. Excitatory postsynaptic potentials (EPSPs) evoked by white-matter stimulation were recorded intracellularly from 42 neurons. Most of polysynaptically elicited EPSPs were sensitive to APV, whereas the majority of monosynaptic EPSPs were not. CNQX almost completely suppressed EPSPs irrespective of monosynaptically or polysynaptically evoked, but in some cases slow EPSPs with low amplitude were spared. These CNQX-resistant EPSPs were elicited polysynaptically and had an anomalous voltage dependence, a characteristic of NMDA receptors. It is suggested that non-NMDA receptors contribute dominantly to first-order synaptic transmission while NMDA receptors participate substantially in second-order transmission so as to serve as a booster of outputs from visual cortex.

Interposed between areas 17 and 18 of cat cerebral cortex is an architectonically distinct zone that represents a substantial portion of the ipsilateral visual field. The extent of this representation was assessed following severance of the corpus callosum by recording the activities of neurons and plotting their receptive fields. The results show that, even after the hemispheres are disconnected, the transition zone still contains a representation of part of the ipsilateral visual hemifield, albeit a reduced one. The extent of this representation is contracted towards the midline so that just one-half to one-third of the azimuths mapped in intact cats can be plotted. As in the intact cat, the width of the region represented is not equal at all elevations, for it extends to only −1.4 deg near the visual axis whereas it extends to −6.6 deg at an elevation of −20 deg. The retention of this representation after the transition zone neurons have been disconnected from the opposite hemisphere indicates that other pathways make a functional contribution to the representation of the ipsilateral field in this region of cortex.

A role for endogenous dopamine in the control of rod and contributions to a second-order retinal neuron, the horizontal cell (HC) was studied in the Xenopus retina. Relative rod and cone contributions were estimated from HC responses to scotopically balanced 491- and 650-nm flashes. In eyecups prepared in light then placed in darkness, cone input to the HC slowed and diminished on a time scale of hours. The decline in cone input was balanced by a slow growth of rod input to the HC. Administration of D-amphetamine, a dopamine releasing agent, restored the light-adapted waveform.

The kinetics of slow light adaptation were examined by recording HC responses from eyecups that had been dark-adapted previously for 11–14 h. When test flashes fell on a dark field, cone input to the HC grew for 2–4 h, reached a plateau, and later declined. If, however, flashes were superimposed on a weak background field, cone input to the HC continued to increase monotonically at about 10%/h. This increase was abolished by superfusion with a nonspecific dopamine receptor blocker, cis-flupenthixol (50 μM), resulting in the complete suppression of cone-to-horizontal cell synaptic transfer and the enhancement of rod-to-horizontal cell communication. Subcutaneous injection of reserpine, a drug that depletes dopamine stores (2 mg/kg on 1–4 successive days), or intraocular injection of the dopamine neurotoxin, 6-hydroxydopamine (10–30 μg) slowed and reduced the amplitude of cone input to the HC, even in completely light-adapted eyes. Subsequent treatment with D-amphetamine (5–50 μM) or dopamine (10 μM) partially restored the normal response.

Our experimental findings are consistent with the following hypothesis. Weak light is sufficient to stimulate dopamine release; dopamine augments cone-to-horizontal cell synaptic transfer and reduces rod-to-horizontal cell communication. The rapid kinetics of the fully light-adapted response depend on the presence of dopamine. Thus, dopamine appears to be an intraretinal signal for slow light adaptation.

To test whether visual experience can affect development of visual behavior in the fruitfly, Drosophila, we measured the visually-guided choice behavior of groups of flies reared in complete darkness, compared with controls reared in a normal light/dark cycle. We used a simple visual preference test, i.e. choice among four different visual targets each consisting of vertical black lines of a particular width on a white background, using a blind testing procedure so that the individual rearing histories were not known by the tester. Both groups of flies were strongly attracted to the vertical lines; however, generally the dark-reared flies were more attracted to the wider stimulus lines than were the control flies. Control experiments in which normally reared adults were kept for several days in darkness showed that the effects of dark-rearing were not simply due to being in darkness, but depended upon the timing of the deprivation. The results indicate that the development of visual behavior can be affected by visual experience in Drosophila and thus open the possibility of using Drosophila for genetic dissection of mechanisms of visual plasticity.

Previous electrophysiological studies have shown that neurons in the nucleus of the basal optic root (nBOR) of the pigeon respond best to wholefield stimuli moving slowly in a particular direction in the contralateral visual field. In this study, we have found that some nBOR neurons respond to wholefield stimulation of both eyes. These binocular neurons have spatially separate receptive fields in both visual fields. Some binocular neurons prefer the same direction of wholefield motion in both eyes, and thus respond best to wholefield visual motion which would result from translation movements of the bird, either ascent, descent, or forward and backward motion. Other neurons prefer opposite directions of wholefield motion in each eye and therefore respond optimally to wholefield visual motion simulating rotational movements of the bird, either roll or yaw. These binocular neurons may play a crucial part in the locomotor behavior of the pigeon by providing visual information distinguishing translational and rotational movements.

The course of optic axons that take different routes at the chiasm have been traced through horizontally sectioned optic nerves in the cat, after unilateral injections of horseradish peroxide into the optic tract. Behind the eye and for most of the course of the nerve, nearly all of the axons that remain uncrossed at the chiasm are located in a retinotopically appropriate position, in the lateral aspect of the nerve. However, in the most caudal segment of the nerve an increasing proportion of these axons are located in regions that are retinotopically inappropriate. Just before the nerve joins the chiasm, uncrossed axons can be found across the full medio-lateral extent of the nerve, although there is still a relative increase in their density laterally.

Labeled axons that cross at the chiasm course in a relatively parallel manner along the greater proportion of the nerve. However, in the caudal segment of the nerve their relative positions change and they appear to course in an irregular manner. This occurs where the uncrossed projection becomes increasingly more widespread.

Axons in the optic nerve are grouped into fascicules. This pattern of organization also changes in the caudal region of the nerve. Although clear fascicular patterns are present along the greater part of the nerve, they become progressively less distinct caudally. The change in the pattern of fasciculation occurs over the same region of the nerve as the relative changes in axon trajectory and distribution.

These results demonstrate that irrespective of chiasmatic route, optic axons in the cat are reorganized in the caudal segment of the nerve. This reorganization is not confined to changes in relative axon position, but is reflected in the structure of the nerve by the change of axon grouping from a fascicular to a non-fascicular arrangement.

The main objective of the present study was to describe the postnatal development of magnocellular and parvocellular LGN axons within the primate striate cortex. For this purpose, we bulk labeled axons in neonatal prosimians (galagos) in vivo or in vitro at regular intervals from birth (PO) to 12 weeks after birth by injecting horseradish peroxidase (HRP) into white matter anterior to the striate cortex. Filled axons within layer IV were reconstructed, quantitatively analyzed, and compared to a population of adult axons described previously (Florence & Casagrande, 1987).

Our results show that although axons are morphologically immature at birth, they are restricted to the upper (IVα) and lower (IVβ) tiers of layer IV of the striate cortex as in adults. In adults, we referred to the presumed magnocellular LGN axons terminating in IVα as type I and the presumed parvocellular axons terminating in IVβ as type II. We used the same convention for developing axons.

From birth to 3 weeks postnatal, type I and II axon classes are more variable in appearance than adult counterparts, and are not morphologically class distinct. As axons mature, parent axon shafts increase in caliber, arbors become smaller and more radial, and other immature features (e.g. spikes, protrusions, growth cones) are less evident. Both arbor classes mature slowly and some still exhibit immature features (e.g. growth cones) as late as 12 weeks postnatally. Although arbors do not show class-distinctive features until late in development, each class does show some unique maturational trends. Type I arbors are only slightly larger than adult counterparts at birth, whereas type II arbors are dramatically larger. Type I arbors increase in branch complexity with age, whereas type II arbors simply show a shift in complexity toward the center of the arbor with decreasing size over time. These growth trends suggest that magnocellular and parvocellular pathways to cortex could be differentially vulnerable to the manipulation of postnatal visual experience.

Current approaches to the problem of equating different colors for luminance (chromatic isoluminance) rely upon human reports of perceptual events that are reduced at some luminance ratio. In this report, a technique is described that evokes a vivid percept of motion of a textured pattern only at isoluminance. Furthermore, in both humans and monkeys, the moving stimulus produces a striking optokinetic response in the same direction as the perceived motion. If used in this manner, the technique can provide an estimate of chromatic isoluminance in a variety of species and be used to corroborate a human subjects's perceptual judgement.

The presence and concentrations of each of the three subtypes of opioid receptors (mu, kappa, and delta) has been studied in the accessory optic nuclei (dorsal, lateral, and medial terminal nuclei and the interstitial nucleus of the superior fasciculus, posterior fibers: DTN, LTN, MTN, and inSFp) in normal young rats with radioligands directed towards each opioid receptor subtype. The changes in mu opioid receptors have also been investigated in monocularly enucleated rats in which one eye was removed and the rats sacrificed at postoperative day (PO) 2, 3, 5, 7, 14, and 30. As the MTN is the only accessory optic nucleus of the rat large enough for semiquantitative evaluation, the mu receptor population of the MTN has been subjected to optical microdensitometric analysis.

All four of the accessory optic nuclei (AOS nuclei) are found to contain exceedingly high levels of mu opioid receptor binding with the selective radioligand [3H]-[D-Ala, MePhe4, Gly-ol5] (DAGO), low levels of kappa opioid receptor binding using the radioligand [3H]-[ethylketocyclazocine] (EKC) together with the competing agents [D-Pro4]-morphiceptin and [D-Ser2, Thr6]-Leu-enkephalin, and an absence of delta opioid receptor binding with the radioligand [3H]-[D-A1a2, D-Leu5]-enkephalin (DADLE) combined with the competing agent [D-Pro4]-morphiceptin. Monocular enucleation, as studied on the mu opioid receptor population with this experimental approach, results in virtually a complete loss of mu opioid receptors throughout all four of the contralaterally located AOS nuclei, including both dorsal and ventral subdivisions of the medial terminal nucleus (MTNd, v). Kappa and delta receptors are very few (kappa receptors) or are lacking (delta receptors) in the AOS nuclei, and for this reason, the effects of monocular enucleation on these two opioid receptor subtypes have not been investigated. Monocular enucleation also produces a significant lowering in mu receptor binding in other primary optic nuclei (the lateral geniculate nuclei, nucleus of the optic tract, and superficial layers of the superior colliculus) and in the pars principalis of the medial geniculate nucleus (description of changes in mu receptors in non-accessory optic primary optic nuclei will be considered elsewhere).

Microdensitometric study of the MTNd, v shows that the decreased binding of mu receptors in this nucleus is barely detectable (about 6%) at PO2 and rises to 6–15% at PO3. At PO5 receptor loss reaches approximately 62%, whereas at PO7 it is about 81% complete. At PO14 and PO30, the mu receptor loss is nearly complete at around 93%. Mu receptor loss involves all of the AOS nuclei contralateral, but none ipsilateral, to ocular enucleation, an observation entirely consistent with the overwhelmingly crossed (about 97%) nature of the retinofugal projection to the rat accessory optic nuclei. These opioid receptors represent a prominent feature in the AOS and other primary optic nuclei of the rat. Their role in visuomotor control remains uncertain but probably involves the fine-tuning of information concerned with compensatory eye movements.

The presence of rod and cone photorceptors has traditionally been linked to well-defined classes of visual capacity by the generalization known as duplicity theory. This paper summarizes results obtained from studies of vision and the visual system in ground squirrels (Spermophilus sp.) that reveal instances where structure/function linkages depart from expectations based in duplicity theory. The details of these exceptions are reviewed and their possible mechanisms discussed.

The first (V-I) and second (V-II) visual areas of primates contain three types of anatomical segregations of neurons as parts of hypothesized “P-B” or “color”, “P-I” or “form,” and “M” or “motion” processing channels. These channels remain distinct in relays of P-B and P-I information to the inferior temporal lobe via V-II and dorsolateral visual cortex for object recognition, and “M” information to posterior parietal cortex via the middle temporal visual area (MT) for visual tracking and attention. The present anatomical experiments demonstrate another channel where “P-B” modules in V-I and “P-B” and “M” modules in V-II merge in the projections to the dorsomedial visual area (DM), which relays to MT and posterior parietal cortex. This integrative area may function in unifying our perception of the visual world, and may allow “color” as well as “motion” to play a role in visual tracking and attention.