W. Burke

Processing of form and motion in area 21a of cat visual cortex

Extracellular recordings from single neurons have been made from presumed area 21a of the cerebral cortex of the cat, anesthetized with N2O/O2/sodium pentobarbitone mixture. Area 21a contains mainly a representation of a central horizontal strip of contralateral visual field about 5 deg above and below the horizontal meridian. Excitatory discharge fields of area 21a neurons were substantially (or slightly but significantly) larger than those of neurons at corresponding eccentricities in areas 17, 19, or 18, respectively. About 95% of area 21a neurons could be activated through either eye and the input from the ipsilateral eye was commonly dominant. Over 90% and less than 10% of neurons had, respectively, C-type and S-type receptive-field organization. Virtually all neurons were orientation-selective and the mean width at half-height of the orientation tuning curves at 52.9 deg was not significantly different from that of neurons in areas 17 and 18. About 30% of area 21a neurons had preferred orientations within 15 deg of the vertical. The mean direction-selectivity index (32.8%) of area 21a neurons was substantially lower than the indices for neurons in areas 17 or 18. Only a few neurons exhibited moderately strong end-zone inhibition. Area 21a neurons responded poorly to fast-moving stimuli and the mean preferred velocity at about 12.5 deg/s was not significantly different from that for area 17 neurons. Selective pressure block of Y fibers in contralateral optic nerve resulted in a small but significant reduction in the preferred velocities of neurons activated via the Y-blocked eye. By contrast, removal of the Y input did not produce significant changes in the spatial organization of receptive fields (S or C type), the size of the discharge fields, the width of orientation tuning curves, or direction-selectivity indices. Our results are consistent with the idea that area 21a receives its principal excitatory input from area 17 and is involved mainly in form rather than motion analysis.

Excitability of the lateral geniculate nucleus in the alert, non-alert and sleeping cat

Summary1.Cats with chronically implanted electrodes have been used to study the excitability of the lateral geniculate nucleus (LGN) during alertness, non-alertness and sleep. Excitability has been assessed by the amplitude of the field response in the LGN to electrical stimulation of the optic tract and by the amplitude of the antidromic response in the optic tract evoked by stimulating the LGN.2.Changes in responsiveness associated with eye movements in the awake animal have been avoided. The change from the alert state to the non-alert state is accompanied by a decrease in the excitability of both the LGN cells and the optic tract nerve endings.3.Postsynaptic excitability decreases further in slow-wave (SW) sleep but during low-voltage-fast-activity (LVF) sleep it returns to a level that is intermediate between that of alertness and non-alertness. During the ponto-geniculooccipital (PGO) waves of LVF sleep excitability is phasically enhanced to above the alert level.4.Presynaptic excitability during SW sleep and LVF sleep in the intervals between the PGO waves is less than during non-alertness, but during the waves it returns to the alert level.5.Our findings indicate that the optic tract endings are tonically depolarized in the alert animal and that the depolarization is reduced in non-alertness and reduced further in sleep. Depolarization is phasically enhanced during the PGO waves.6.It is suggested that the LGN may be the first site on the visual pathway at which information is modified according to the animals state of consciousness.

Selective degeneration of optic nerve fibres in the cat produced by a pressure block.

Using a technique described previously, we have applied pressure to the optic nerve of a cat sufficient to cause conduction block of the t1 response (the response of the Y optic nerve fibres). A greater pressure, usually sufficient to cause a transient block of the t2 response (the response of the X fibres), leads to degeneration of the Y axons caudal to the block. This is demonstrated by the disappearance of the t1 response in this region after 4‐5 days and by the presence in electron micrographs of degenerating large (Y) fibres. Some small fibres also show degeneration, but the medium (X) fibres are largely spared. The time course of loss of response in the Y fibres is similar whether the loss is due to a pressure block or to enucleation, suggesting that the pressure block as used by us causes a disruption of the axon. If the pressure is great enough to block part of the t2 response (X fibres) there is also a similarity in time course of loss of response to that following enucleation. Both for the enucleated and the pressure‐blocked cat the t2 response fails about 1 day before the t1 response. This is in apparent disagreement with the morphological findings in the literature, confirmed here, indicating an earlier degeneration of the larger fibres. The post‐synaptic response in the lateral geniculate nucleus to the t1 input (the r1 response) also fails about 1 day before the t1 response. In the visual cortex the loss of the r1 response reveals more clearly than is normally possible an r2 response, the response of the X optic radiation fibres. The response in the optic nerve or tract to a bright flash of light is dominated by the response of the Y fibres. When these are blocked the response is greatly reduced.

Long Survival of Retinal Ganglion Cells in the Cat After Selective Crush of the Optic Nerve.

In each of four cats gentle pressure was applied to one optic nerve by means of an inflatable cuff in order to disrupt the largest axons (Y fibres) and so produce a conduction block in them. It has previously been shown that this technique, as used by us, causes Wallerian degeneration in the fibres posterior to the site of application of the pressure (the crush site). The optic nerves and retinas in these cats were examined 2–2.8 years later. The optic nerves were prepared for electron microscopy and the retinas were flat‐mounted. Here we report an average 90% loss of large axons (>5 μm diameter) in the nerve posterior to the crush site. However, in the part of the nerve anterior to the crush site there was only a 33% loss and in the retina only a 57.5% reduction in the number of neurons of soma diameter >25 μm (i.e. alpha cells, the cell bodies of the Y neurons). These last two sets of values were significantly different, suggesting that the retinal ganglion cells had shrunk relatively more than the axons. Thus, the crushing technique has effectively axotomized almost all the Y fibres but, in spite of this, about half of the alpha retinal ganglion cells have survived this particular form of axotomy, with their axons intact at least for some distance into the optic nerve. This long survival raises the possibility that these neurons may have regenerated axons which have found targets and thus ensured their survival.

Laminar differences in plasticity in area 17 following retinal lesions in kittens or adult cats.

Circumscribed retinal lesions in adult cats result in a reorganization of circuitry in area 17 such that neurons in the lesion projection zone (LPZ) can now be activated, not from their original receptive fields (RFs) but from regions of normal retina adjacent to the lesion (‘ectopic’ RFs). We have studied this phenomenon further by making circumscribed monocular retinal lesions in 8‐week‐old kittens and recording responses to visual stimuli of neurons in the LPZ of area 17 when these cats reached adulthood. These responses have been compared with those in adult‐lesioned cats either of relatively short postlesion survival (2–24 weeks) or long postlesion survival (3.5–4.5 years). In both kitten‐lesioned and adult‐lesioned animals most LPZ neurons recorded from the supragranular layers (II and III) not only exhibited new ectopic RFs when stimuli were presented via the lesioned eye but the RF properties (e.g. the sizes of excitatory RFs, orientation and direction selectivities, velocity preferences and upper cut‐off velocities) were often indistinguishable from those seen when stimuli were presented via the nonlesioned eye. Similarly, in both kitten‐lesioned and adult‐lesioned animals, most LPZ neurons recorded from the granular and infragranular layers (IV, V, VI), like those recorded from the supragranular layers, were binocular. However, in adult‐lesioned but not in kitten‐lesioned animals, the responses and the upper cut‐off velocities of LPZ cells recorded from the granular and infragranular layers to stimuli presented via ectopic RFs tended to be, respectively, substantially weaker and lower than those for stimuli presented via the nonlesioned eye. The age‐related laminar differences in reorganizational plasticity of cat striate cortex correlate with the lamino‐temporal pattern of distribution of N‐methyl‐d‐aspartate glutamate receptors in striate cortex.

LIMITS OF PARALLEL PROCESSING: EXCITATORY CONVERGENCE OF DIFFERENT INFORMATION CHANNELS ON SINGLE NEURONS IN STRIATE AND EXTRASTRIATE VISUAL CORTICES

1. It has been postulated that the distinct parallel retinogeniculo‐cortical information channels characterizing visual pathways of virtually all mammals are selectively linked to parallel motion, colour and/or form information processing ‘streams’ distinguishable within the primary visual cortices, extrastriate cortical areas of occipital lobes and the temporal and parietal visual cortices.

Feedback signals from cat's area 21a enhance orientation selectivity of area 17 neurons

We have studied the contribution of feedback signals originating from one of the “form-processing” extrastriate cortical areas, area 21a (A21a), to orientation selectivity of single neurons in the ipsilateral area 17 (A17). Consistent with previous findings, reversible inactivation (cooling to 5–10°C) of area 21a resulted in a substantial reduction in the magnitude of the maximum response (Rmax) of A17 cells accompanied by some changes in the half-width at half-height of the Rmax (HWHH). By fitting model functions to the neurons’ response profiles we found that in the vast majority of orientation-tuned A17 cells tested (30/39, 77%), inactivation of A21a resulted in a “flattening” of their orientation-tuning curves. It is characterised by a substantial reduction in the Rmax associated with either a broadening of the orientation-tuning curves (17 cells) or a relatively small reduction (12 cells) or no change (1 cell) in the HWHH. When the “flattening” effect was quantified using a simple ratio index or R/W, defined as Rmax/HWHH, we found that R/W was significantly reduced during inactivation of A21a. The change in R/W is strongly correlated with the change in the maximum slope of the orientation-tuning curves. Furthermore, analysis of response variability indicates that “signal-to-noise” ratio of the responses of A17 neurons decreases during inactivation of A21a. Our results suggest that the predominately excitatory feedback signals originating from A21a play a role in enhancing orientation selectivity of A17 neurons and hence are likely to improve overall orientation discriminability.

Convergence of Y and non-Y channels onto single neurons in the superior colliculi of the cat

Receptive field properties of single neurons in the cat superior colliculus were examined following selective conduction-block of Y-type fibers in contralateral optic nerve. Although the responses evoked by photic stimuli presented via the Y-blocked eye were significantly weaker than those evoked by stimuli presented via the normal eye, > 85% of collicular cells were binocular. Furthermore, when binocular cells were stimulated via the Y-blocked eye their median upper cut-off velocity (100°/s) was significantly lower than that (400°/s) for stimuli presented via the normal eye. Thus, there is a substantial degree of excitatory convergence of Y- and non-Y- information channels on single collicular neurons and the responses to high velocity of motion appear to depend on the integrity of Y-type input.

Non-dominant suppression in the dorsal lateral geniculate nucleus of the cat: laminar differences and class specificity

Binocular non-dominant suppression (NDS) in the dorsal lateral geniculate nucleus (LGNd) of the cat was studied by recording from single neurons in the LGNd of anaesthetized, paralysed cats while stimulating the non-dominant eye with a moving light bar. The maintained discharge rate of LGNd neurons was varied by stimulating the dominant eye in various ways: by varying the size or contrast of a flashed spot, by varying the inner diameter of a flashed annulus of large outer diameter, by varying the velocity of a moving light bar, and by covering the eye. Non-dominant suppression was quantified either as the decrease in the maintained discharge rate (the “dip”), expressed as spikes per second, or as the ratio of the dip to the maintained discharge rate (the “dip ratio”). At low maintained discharge rates the dip, although low in value, frequently approached the maintained rate, i.e. the dip ratio approached unity. As the maintained discharge rate increased the dip value also increased, but more slowly than the maintained discharge rate, i.e. the dip ratio decreased. At maintained discharge rates above about 30 spikes/s, in many neurons the dip appeared to be approaching a constant value. This strong dependence of NDS on the maintained discharge rate of the LGNd neuron suggests that the inhibitory input to the cell arises from a region of the brain that receives an input both from the non-dominant eye and from the LGNd cell. Reasons are given for thinking that this region is the perigeniculate nucleus. Because of the strong dependence of dip and dip ratio on the maintained discharge rate, it was necessary to adopt stringent criteria when comparing NDS in two different sets of neurons or of the same set of neurons in different conditions. We recognized a significant difference in NDS between two classes of neurons or between two states only if: (1) there was no significant difference between the maintained discharge rates, and (2) there was a significant difference for both dip and dip ratio between the two classes or states. Using these criteria we found: (1) no difference between non-lagged X (XNL) and non-lagged Y (YNL) cells, (2) no difference between on-centre and off-centre cells for either XNL or YNL cells, (3) no difference between XNL cells and lagged X (XL) cells. However, there was a significant difference between cells in lamina A and those in lamina A1 for both XNL and YNL cells, dip and dip ratio values being about twice as great in lamina A. In cats in which one optic nerve had been pressure-blocked so as to prevent conduction in the largest axons (Y fibres), loss of conduction in Y fibres crossing the chiasm and projecting to the contralateral LGNd did not affect NDS. Loss of conduction in Y fibres projecting to the ipsilateral LGNd caused a complete loss of NDS in the non-lagged Y cells of lamina A and a substantial decrease in the NDS of the nonlagged X cells of lamina A. The latter cells must, therefore, be partly suppressed by non-Y fibres, presumably X fibres. It also follows that all the NDS of cells in lamina A1 is mediated by non-Y fibres, probably X fibres. Thus, NDS in the cat is partly class-specific and partly not. The discharge of retinal ganglion cells also protects the LGNd cells against NDS. The contribution of Y fibres to this anti-suppressive action was also examined. Contralaterally projecting Y fibres make no contribution. Ipsilaterally projecting Y fibres exert an anti-suppressive action on non-lagged X cells in lamina A1. It follows also that the anti-suppressive action on cells in lamina A mediated by contralaterally projecting fibres is due to non-Y fibres, presumably X fibres. Thus, both the suppressive and the anti-suppressive actions of Y fibres are mediated only by the uncrossed pathway.

Area 21a of cat visual cortex strongly modulates neuronal activities in the superior colliculus

We have examined the influence of cortico‐tectal projections from one of the pattern‐processing extrastriate visual cortical areas, area 21a, on the responses to visual stimuli of single neurones in the superior colliculi of adult cats. For this purpose area 21a was briefly inactivated by cooling to 10 °C using a Peltier device. Responses to visual stimuli before and during cooling as well as after rewarming ipsilateral area 21a were compared. In addition, in a subpopulation of collicular neurones we have studied the effects of reversible inactivation of ipsilateral striate cortex (area 17, area V1). When area 21a was cooled, the temperature of area 17 was kept at 36 °C and vice versa. In the majority of cases (41/65; 63 %), irrespective of the velocity response profiles of collicular neurones, inactivation of area 21a resulted in a significant decrease in magnitude of responses of neurones in the ipsilateral colliculus and only in a small proportion of cells (2/65; 3.1 %) was there a significant increase in the magnitude of responses. Inactivation of area 21a resulted in significant changes in the magnitude of responses of collicular cells located not only in the retino‐recipient layers but also in the stratum griseum intermediale. In most cases, reversible inactivation of area 17 resulted in a greater reduction in the magnitude of responses of collicular cells than inactivation of area 21a. Reversible inactivation of area 21a also affected the direction selectivity indices and length tuning of most collicular cells tested.