The role of feedback projections in feature tuning and neuronal excitability in the early primate visual system.

A general assumption in visual neuroscience is that basic receptive field properties such as orientation and direction selectivity are constructed within intrinsic neuronal circuits and feedforward projections. In addition, it is assumed that general neuronal excitability and responsiveness in early visual areas is to a great extent independent of feedback input originating in areas higher in the stream. Here, we review the contribution of feedback projections from MT, V4 and pulvinar to the receptive field properties of V2 neurons in the anesthetized and paralyzed monkey. Importantly, our results contradict both of these assumptions. We separately inactivated each of these three brain regions using GABA pressure injections, while simultaneously recording V2 single unit activity before and hours after inactivation. Recordings and GABA injections were carried out in topographically corresponding regions of the visual field. We outline the changes in V2 activity, responsiveness and receptive field properties for early, mid and late post-injection phases. Immediately after injection, V2 activity is globally suppressed. Subsequently, there is an increase in stimulus-driven relative to spontaneous neuronal activity, which improves the signal-to-noise coding for the oriented moving bars. Notably, V2 tuning properties change substantially relative to its pre-injection selectivity profile. The resulting increase or decrease in selectivity could not be readily predicted based on the selectivity profile of the inactivated site. Finally, V2 activity rebounds before returning to it pre-injection profile Our results show that feedback projections profoundly impact neuronal circuits in early visual areas, and may have been heretofore largely underestimated in their physiological role.

Changes of ongoing activity in Cebus monkey perirhinal cortex correlate with behavioral performance.

A Cebus apella monkey weighing 4 kg was trained in a saccadic eye movement task and while the animal performed the task we recorded the extracellular activity of perirhinal cortical neurons. Although the task was very simple and maintained at a constant level of difficulty, we observed considerable changes in the performance of the monkey within each experimental session. The behavioral states responsible for such variation may be related to arousal, motivation or attention of the animal while engaged in the task. In approximately 20% (16/82) of the units recorded, long-term direct or inverse correlations could be demonstrated between the monkey's behavioral state and the cells' ongoing activity (independent of the visual stimulation or of the specific behavior along a trial). The perirhinal cortex and other medial temporal structures have long been associated with normal memory function. The data presented here were interpreted in terms of recent reports focusing on the subcortical afferents to temporal lobe structures and their possible role in controlling arousal, motivation, or attention.

Visual topography of V1 in the Cebus monkey.

The representation of the visual field in the striate cortex (V1) was mapped with multiunit electrodes in the Cebus monkey. Nine Cebus apella, anesthetized with N2O and immobilized with pancuromium bromide were studied in repeated recording sessions. In each hemisphere, V1 contains a continuous representation of the contralateral visual hemifield. The representation of the vertical meridian (VM) forms the external border of V1 except at the anteriormost portion of the calcarine fissure. The representation of the horizontal meridian (HM) divides the area so that the representation of the lower visual field is located dorsally, and that of the upper field ventrally. The convoluted surface of V1 can be only partially unfolded, and no precise "flattened" map can be obtained without introducing surface discontinuities. The visual topography of V1 is presented in a series of coronal sections and in "flattened" maps. The representation of the central visual field is magnified relative to that of the periphery in V1. The evaluation of the cortical magnification factors measured along isoeccentric and isopolar dimensions in the partially unfolded model of V1 revealed anisotropies in the representation of the visual field with larger magnification along isopolar lines than along isoeccentric lines. Receptive field size increases with increasing eccentricity, whereas point image size decreases with increasing eccentricity.

Representation of the visual field in the second visual area in the Cebus monkey.

The representation of the visual field in the second visual area (V2) was reconstructed from multiunit visual responses and anatomical tracers. Receptive field plotting was performed during multiple recording sessions in seven Cebus apella monkeys under N2O/O2 and immobilized with pancuronium bromide. V2 forms a continuous belt of variable width around striate cortex (V1) except at the most anterior portion of the calcarine sulcus. In each hemisphere V2 contains a visuotopic representation of the contralateral visual hemifield. The representation of the vertical meridian is adjacent to that of V1 and forms the posterior border of V2. The representation of the fovea of V2 is adjacent to that of V1. The representation of the horizontal meridian (HM) is continuous with that of V1; then it splits to form the anterior border of V2, both dorsally and ventrally. The lower quadrant of the visual field is represented dorsally and the upper quadrant ventrally. The visual topography of V2 is coarser than that of V1. In V2, receptive fields corresponding to recording sites separated by a cortical distance of up to 4 mm may represent the same portion of the visual field. In three additional animals, combined injections of fluorescent tracers along the HM representation in V1 yielded two projection sites at the anterior border of V2. The split of the HM representation is estimated to occur at an eccentricity below 1 degree. Quantitative analysis showed that in V2 the representation of the central visual field is magnified relative to that of the periphery. The cortical magnification factor is greater along the isopolar dimension than along the isoeccentric one. Receptive field size in V2 increases with increasing eccentricity. In sections stained for myelin by the Heidenhein-Wöelcke method V2 can be distinguished from the surrounding cortex for most of its extent.

Visual topography of V2 in the macaque.

The representation of the visual field in the area adjacent to striate cortex was mapped with multiunit electrodes in the macaque. The animals were immobilized and anesthetized and in each animal 30 to 40 electrode penetrations were typically made over several recording sessions. This area, V2, contains a topographically organized representation of the contralateral visual field up to an eccentricity of at least 80 degrees. The representation of the vertical meridian is adjacent to that in striate cortex (V1) and forms the posterior border of V2. The representation of the horizontal meridian in V2 forms the anterior border of V2 and is split so that the representation of the lower visual field is located dorsally and that of the upper field ventrally. As in V1, the representation of the central visual field is magnified relative to that of the periphery. The area of V2 is slightly smaller than that of V1. At a given eccentricity, receptive field size in V2 is larger than in V1. The myeloarchitecture of V2 is distinguishable from that of the surrounding cortex. The location of V2 corresponds, at least approximately, to that of cytoarchitectonic Area OB. V2 is bordered anteriorly by several other areas containing representations of the visual field.

Topographic organization of cortical input to striate cortex in the Cebus monkey: a fluorescent tracer study.

Cortical afferents to area V1 were studied in seven Cebus monkeys by means of retrograde fluorescent tracers. Injections were placed in V1, under electrophysiological guidance, in the regions of representation of both the upper and lower visual quadrants, at eccentricities that ranged from 0.5 to 64 degrees. In all cases retrogradely filled neurons were found in retinotopically corresponding portions of areas V2 and MT, as defined electrophysiologically (Rosa et al: J. Comp. Neurol. 275:326, 1988; Fiorani et al: J Comp Neurol 287:98, 1989). The results also revealed two other visual zones located anterior to V2 here named third and fourth visual areas. A topographical organization of the connections was observed in these areas, with upper quadrant located ventrally and lower quadrant located dorsally. A clear central-peripheral gradient, from the lateral to the medial cortical surface, was also observed in these areas. Lower field injections revealed crude topographic organization in area DZ and a diffuse projecting zone in the annectent gyrus. Peripheral injections in V1 revealed a clear upper and lower field segregation in areas PO and prostriata as well as a complex topography in MST. In addition, another region of labeling revealed the presence of an area, the temporal ventral posterior region, with an organized topographic representation of the upper field, with a central to peripheral gradient, from the lateral to the medial cortical surface. Three groups of cortical areas were distinguished according to the laminar distribution of neurons labeled from V1. In the first group, which is characterized by dense infra- and supragranular labeling, only V2 was included. The second group consists of areas V3, MT, and PO. These areas show dense labeling in the infragranular layers and occasionally sparse labeling in the supragranular layers. Finally, V4 and the other projecting areas, which are characterized by exclusive labeling of the infragranular layers were included in the third group.

Topographical organization of cortical afferents to extrastriate visual area PO in the macaque: a dual tracer study.

We have examined the origin and topography of cortical projections to area PO, an extrastriate visual area located in the parieto-occipital sulcus of the macaque. Distinguishable retrograde fluorescent tracers were injected into area PO at separate retinotopic loci identified by single-neuron recording. The results indicate that area PO receives retinotopically organized inputs from visual areas V1, V2, V3, V4, and MT. In each of these areas the projection to PO arises from the representation of the periphery of the visual field. This finding is consistent with neurophysiological data indicating that the representation of the periphery is emphasized in PO. Additional projections arise from area MST, the frontal eye fields, and several divisions of parietal cortex, including four zones within the intraparietal sulcus and a region on the medial dorsal surface of the hemisphere (MDP). On the basis of the laminar distribution of labeled cells we conclude that area PO receives an ascending input from V1, V2, and V3 and receives descending or lateral inputs from all other areas. Thus, area PO is at approximately the same level in the hierarchy of visual areas as areas V4 and MT. Area PO is connected both directly and indirectly, via MT and MST, to parietal cortex. Within parietal cortex, area PO is linked to particular regions of the intraparietal sulcus including VIP and LIP and two newly recognized zones termed here MIP and PIP. The wealth of connections with parietal cortex suggests that area PO provides a relatively direct route over which information concerning the visual field periphery can be transmitted from striate and prestriate cortex to parietal cortex. In contrast, area PO has few links with areas projecting to inferior temporal cortex. The pattern of connections revealed in this study is consistent with the view that area PO is primarily involved in visuospatial functioning.

Tangential distribution of cytochrome oxidase-rich blobs in the primary visual cortex of macaque monkeys.

We studied the tangential distribution of cytochrome c oxidase (CytOx)-rich blobs in four striate cortices of three normal monkeys (Macaca mulatta). The spatial density and cross-sectional area of blobs were analyzed in CytOx-reacted tangential sections of flat-mounted preparations of the striate cortex (V1). Well-delimited CytOx-rich blobs were found in the middle portion of cortical layer III of the V1. Throughout the binocular field representation, the spatial density of blobs was nearly constant with a mean value of four to five blobs per mm2. In the monocular portions of V1, however, blob spatial density diminished. In all cases, the mean cross-sectional area of blobs was constant in the V1. The small variation of CytOx blob topography with visual field eccentricity contrasts with the variation described in previously published material.

Visual area MT in the Cebus monkey: location, visuotopic organization, and variability.

The representation of the visual field in the dorsal portion of the superior temporal sulcus (ST) was studied by multiunit recordings in eight Cebus apella, anesthetized with N2O and immobilized with pancuronium bromide, in repeated recording sessions. On the basis of visuotopic organization, myeloarchitecture, and receptive field size, area MT was distinguished from its neighboring areas. MT is an oval area of about 70 mm2 located mainly in the posterior bank of the superior temporal sulcus. It contains a visuotopically organized representation of at least the binocular visual field. The representation of the vertical meridian forms the dorsolateral, lateral, and ventrolateral borders of MT and that of the horizontal meridian runs across the posterior bank of ST. The fovea is represented at the lateralmost portion of MT, while the retinal periphery is represented medially. The representation of the central visual field is magnified relative to that of the periphery in MT. The cortical magnification factor in MT decreases with increasing eccentricity following a negative power function. Receptive field size increases with increasing eccentricity. A method to evaluate the scatter of receptive field position in multiunit recordings based on the inverse of the magnification factor is described. In MT, multiunit receptive field scatter increases with increasing eccentricity. As shown by the Heidenhain-Woelcke method, MT is coextensive with two myeloarchitectonically distinct zones: one heavily myelinated, located in the posterior bank of ST, and another, less myelinated, located at the junction of the posterior bank with the anterior bank of ST. At least three additional visual zones surround MT: DZ, MST, and FST. The areas of the dorsal portion of the superior temporal sulcus in the diurnal New World monkey Cebus are comparable to those described for the diurnal Old World monkey, Macaca. This observation suggests that these areas are ancestral characters of the simian lineage and that the differences observed in the owl monkey may be secondary adaptations to a nocturnal ecological niche.

Identification and visuotopic organization of areas PO and POd in Cebus monkey.

Two visual areas of the anterior bank of the parietooccipital sulcus, areas PO and POd, were identified and their visual field representations were studied in six anesthetized and paralyzed Cebus monkeys. The definition of these areas was based on electrophysiological mapping and myeloarchitecture. PO is located in the ventral aspect of the anterior bank of the parietooccipital sulcus and ventral precuneate gyrus. It borders area V2 posteriorly and ventrally in the depth of the parietooccipital sulcus, area V3d laterally, and another undescribed visual area medially. POd was located dorsal to area PO and ventral to architectonic area PE. The representations of the visual field in areas PO and POd are complex. In each hemisphere, these areas have a virtually complete representation of the contralateral visual hemifield. Different from the previously described visual areas, in PO and POd there is a precise organization of isopolar lines and a complex organization of the isoeccentric ones. In PO, as well as in POd, the representation of the horizontal meridian runs dorsoventrally along the parietooccipital sulcus. The upper visual quadrant is represented medially and the lower visual quadrant laterally. A large and complex representation of the periphery, from 20 degrees to 60 degrees eccentricity is present at the lateral and medial portions of these areas. By contrast, the representation of the central 20 degrees is very small in both PO and POd. The central visual field is represented ventrally in PO and dorsally in area POd. Area POd shows a more stratified myeloarchitectonic pattern than PO and both areas can be distinguished from other surrounding areas by their heavier myelinated pattern.

Visual cortical projections and chemoarchitecture of macaque monkey pulvinar.

We investigated the patterns of projections from the pulvinar to visual areas V1, V2, V4, and MT, and their relationships to pulvinar subdivisions based on patterns of calbindin (CB) immunostaining and estimates of visual field maps (P(1), P(2) and P(3)). Multiple retrograde tracers were placed into V1, V2, V4, and/or MT in 11 adult macaque monkeys. The inferior pulvinar (PI) was subdivided into medial (PI(M)), posterior (PI(P)), central medial (PI(CM)), and central lateral (PI(CL)) regions, confirming earlier CB studies. The P(1) map includes PI(CL) and the ventromedial portion of the lateral pulvinar (PL), P(2) is found in ventrolateral PL, and P(3) includes PI(P), PI(M), and PI(CM). Projections to areas V1 and V2 were found to be overlapping in P(1) and P(2), but those from P(2) to V2 were denser than those to V1. V2 also received light projections from PI(CM) and, less reliably, from PI(M). Neurons projecting to V4 and MT were more abundant than those projecting to V1 and V2. Those projecting to V4 were observed in P(1), densely in P(2), and also in PI(CM) and PI(P) of P(3). Those projecting to MT were found in P(1)- P(3), with the heaviest projection from P(3). Projections from P(3) to MT and V4 were mainly interdigitated, with the densest to MT arising from PI(M) and the densest to V4 arising from PI(P) and PI(CM). Because the calbindin-rich and -poor regions of P(3) corresponded to differential patterns of cortical connectivity, the results suggest that CB may further delineate functional subdivisions in the pulvinar.

Neurofilament protein is differentially distributed in subpopulations of corticocortical projection neurons in the macaque monkey visual pathways.

Previous studies of the primate cerebral cortex have shown that neurofilament protein is present in pyramidal neuron subpopulations displaying specific regional and laminar distribution patterns. In order to characterize further the neurochemical phenotype of the neurons furnishing feedforward and feedback pathways in the visual cortex of the macaque monkey, we performed an analysis of the distribution of neurofilament protein in corticocortical projection neurons in areas V1, V2, V3, V3A, V4, and MT. Injections of the retrogradely transported dyes Fast Blue and Diamidino Yellow were placed within areas V4 and MT, or in areas V1 and V2, in 14 adult rhesus monkeys, and the brains of these animals were processed for immunohistochemistry with an antibody to nonphosphorylated epitopes of the medium and heavy molecular weight subunits of the neurofilament protein. Overall, there was a higher proportion of neurons projecting from areas V1, V2, V3, and V3A to area MT that were neurofilament protein-immunoreactive (57-100%), than to area V4 (25-36%). In contrast, feedback projections from areas MT, V4, and V3 exhibited a more consistent proportion of neurofilament protein-containing neurons (70-80%), regardless of their target areas (V1 or V2). In addition, the vast majority of feedback neurons projecting to areas V1 and V2 were located in layers V and VI in areas V4 and MT, while they were observed in both supragranular and infragranular layers in area V3. The laminar distribution of feedforward projecting neurons was heterogeneous. In area V1, Meynert and layer IVB cells were found to project to area MT, while neurons projecting to area V4 were particularly dense in layer III within the foveal representation. In area V2, almost all neurons projecting to areas MT or V4 were located in layer III, whereas they were found in both layers II-III and V-VI in areas V3 and V3A. These results suggest that neurofilament protein identifies particular subpopulations of corticocortically projecting neurons with distinct regional and laminar distribution in the monkey visual system. It is possible that the preferential distribution of neurofilament protein within feedforward connections to area MT and all feedback projections is related to other distinctive properties of these corticocortical projection neurons.

Distribution of calbindin, parvalbumin and calretinin in the lateral geniculate nucleus and superior colliculus in Cebus apella monkeys.

We studied the distribution of the calcium-binding proteins calbindin, parvalbumin and calretinin, in the superior colliculus and in the lateral geniculate nucleus of Cebus apella, a diurnal New World monkey. In the superior colliculus, these calcium-binding proteins show different distribution patterns throughout the layers. After reaction for calretinin one observes a heavy staining of the neuropil with few labeled cells in superficial layers, a greater number of large and medium-sized cells in the stratum griseum intermediale, and small neurons in deep layers. The reaction for calbindin revealed a strong staining of neuropil with a large number of small and well stained cells, mainly in the upper half of the stratum griseum superficiale. Intermediate layers were more weakly stained and depicted few neurons. There were few immunopositive cells and little neuropil staining in deep layers. The reaction for parvalbumin showed small and medium-sized neurons in the superficial layers, a predominance of large stellate cells in the stratum griseum intermediale, and medium-sized cells in the deep layers. In the lateral geniculate nucleus of Cebus, parvalbumin is found in the cells of both the P and M pathways, whereas calbindin is mainly found in the interlaminar and S layers, which are part of the third visual pathway. Calretinin was only found in cells located in layer S. This pattern is similar to that observed in Macaca, showing that these calcium-binding proteins reveal different components of the parallel visual pathways both in New and Old World monkeys.

GABAergic retinocollicular projection in the New World monkey Cebus apella.

GABA immunoreactivity was examined in the retina of the New World monkey Cebus apella. Labeled cell bodies were identified as horizontal, bipolar, interplexiform, amacrine and a population of putative ganglion cells. To determine whether ganglion cells were immunoreactive to GABA, double-labeling experiments were performed using Fast Blue as retrograde neuronal tracer injected into the superior colliculus. Retinas containing FB-labeled ganglion cells were subsequently incubated with antiserum against GABA. Although retinocollicular ganglion cells were found in three different layers (ganglion cell layer, inner nuclear layer and inner plexiform layer), our experiments revealed GABA-positive ganglion cells only in the outer half of the ganglion cell layer.

Single unit response types in the pulvinar of the Cebus monkey to multisensory stimulation.

Response to sensory stimulation was studied in 162 neurons in the pulvinar of Cebus monkeys in 4 acute and 5 chronic preparations. Two basic response patterns were observed: type I responses, similar to those obtained in primary relay centers, were only observed after visual stimulation. Type II responses were obtained after stimulation of more than one sensory modality. Characteristically these responses presented fatigue and habituation. Temporal relationship between stimulus and response was not as clear as in type I responses, afterdischarge frequently occurred. Taking these response types into consideration two groups of units were identified in the pulvinar. Units of group A (91 neurons) showed type I response to visual stimulation. For these units receptive fields similar to those found in other regions of visual projection could be defined. As a rule units of group A displayed type II responses to other sensory modalities. Units of group B (71) did not display type I responses; they always responded to visual, somatic, auditory and olfactory stimuli with type II responses. They could be activated by a single sensory modality (B, unimodal) or by more than one sensory modality (B, multimodal).

Visual receptive fields of units in the pulvinar of cebus monkey.

Visually driven units, isolated in the ventrolateral group -- Pv1g (109) and in subnucleus Pmu (33) of the pulvinar of the cebus monkey, were studied in acute and chronic preparations under nitrous oxide N2O/O2 anesthesia during periods of EEG arousal. Taking into consideration the response properties to static or moving stimuli as well as the organization of the receptive fields, units isolated in the pulvinar were subdivided into 8 groups. Units displaying dynamic properties predominate over static ones. Static units were classified in 3 groups; of these, one showed uniform receptive fields; the remaining two groups, with non-uniform RFs, were further subdivided in terms of orientation selectivity. By testing for directional sensitivity, organization of the RFs and orientation selectivity, the dynamic units were divided in 5 groups. Among these there was a predominance of directional units, displaying uniform RFs and showing orientation selectivity. Although the receptive fields would extend into the ipsilateral hemifield (up to 10 degrees), their centers were always located in the contralateral visual hemifield. Binocularly driven units predominate in both static and dynamic categories.

Visuotopic organization of the cebus pulvinar: a double representation the contralateral hemifield.

The projection of the visual field in the pulvinar nucleus was studied in 17 Cebus monkeys using electrophysiological techniques. Visual space is represented in two regions of the pulvinar; (1) the ventrolateral group, Pvlg, comprising nuclei P delta, P delta, P gamma, P eta and P mu 1; and (2) P mu. In the first group, which corresponds to the pulvinar inferior and ventral part of the pulvinar lateralis, we observed a greater respresentation of the central part of the visual field. Approximately 58% of the volume of the ventrolateral group is concerned with the visual space within 10 degrees of the fovea. This portion of the visual field is represented at its lateral aspects, mainly close to the level of the caudal pole of the lateral geniculate nucleus (LGN). Projection of the vertical meridian runs along its lateral border while that of the horizontal one is found running from the dorsal third of the LGN's hilus to the medial border of the ventro-lateral group. The lower quadrant is represented at its dorsal portion while the upper quadrant is represented at the ventral one. In Pmu the representation is rotated 90 degrees clockwise around the rostrocaudal axis: the vertical meridian is found at the ventromedial border of this nucleus. Thus, the lower quadrant is represented at the later portion of Pmu and the upper at its medial portion. Both projections are restricted to the contralateral hemifield.

Cortical streams of visual information processing in primates.

1. The topographic organization of the cortical visual areas in the Cebus monkey and their anatomical connections support the subdivision of the visual pathways into ventral and dorsal streams of visual information processing. 2. We propose that the dorsal stream, as defined by Ungerleider and Mishkin (In: Ingle DJ, Goodale MA and Mansfield RJW (Editors), Analysis of Visual Behavior. MIT Press, Boston, 1982), be subdivided into dorsolateral and dorsomedial streams, which are concerned with different aspects of the processing of motion and spatial perception. 3. The data support the hypothesis of concurrent, modular processing of visual attributes in cortical visual areas in the different streams, and highlight some features of the visual field representation in each area which may reflect functional specialization of these streams. 4. The visual topography is locally disrupted in some cortical areas by the existence of functionally different modules. However, a global visuotopic organization is preserved in most areas. 5. The visuotopic organization may provide the address of space coordinates to integrate information concerning the same retinotopic locus across different visual areas.

The brain decade in debate: VI. Sensory and motor maps: dynamics and plasticity.

This article is an edited transcription of a virtual symposium promoted by the Brazilian Society of Neuroscience and Behavior (SBNeC). Although the dynamics of sensory and motor representations have been one of the most studied features of the central nervous system, the actual mechanisms of brain plasticity that underlie the dynamic nature of sensory and motor maps are not entirely unraveled. Our discussion began with the notion that the processing of sensory information depends on many different cortical areas. Some of them are arranged topographically and others have non-topographic (analytical) properties. Besides a sensory component, every cortical area has an efferent output that can be mapped and can influence motor behavior. Although new behaviors might be related to modifications of the sensory or motor representations in a given cortical area, they can also be the result of the acquired ability to make new associations between specific sensory cues and certain movements, a type of learning known as conditioning motor learning. Many types of learning are directly related to the emotional or cognitive context in which a new behavior is acquired. This has been demonstrated by paradigms in which the receptive field properties of cortical neurons are modified when an animal is engaged in a given discrimination task or when a triggering feature is paired with an aversive stimulus. The role of the cholinergic input from the nucleus basalis to the neocortex was also highlighted as one important component of the circuits responsible for the context-dependent changes that can be induced in cortical maps.

GABA-induced inactivation of Cebus apella V2 neurons: effects on orientation tuning and direction selectivity.

We investigated the GABA-induced inactivation of V2 neurons and terminals on the receptive field properties of this area in an anesthetized and paralyzed Cebus apella monkey. Extracellular single-unit activity was recorded using tungsten microelectrodes in a monkey before and after pressure-injection of a 0.25 or 0.5 M GABA solution. The visual stimulus consisted of a bar moving in 8 possible directions. In total, 24 V2 neurons were studied before and after blocker injections in 4 experimental sessions following GABA injection into area V2. A group of 10 neurons were studied over a short period. An additional 6 neurons were investigated over a long period after the GABA injection. A third group of 8 neurons were studied over a very long period. Overall, these 24 neurons displayed an early (1-20 min) significant general decrease in excitability with concomitant changes in orientation or direction selectivity. GABA inactivation in area V2 produced robust inhibition in 80% and a significant change in directional selectivity in 60% of the neurons examined. These GABA projections are capable of modulating not only levels of spontaneous and driven activity of V2 neurons but also receptive field properties such as direction selectivity.