Informative features of local field potential signals in primary visual cortex during natural image stimulation.

The local field potential (LFP) is of growing importance in neurophysiology as a metric of network activity and as a readout signal for use in brain-machine interfaces. However, there are uncertainties regarding the kind and visual field extent of information carried by LFP signals, as well as the specific features of the LFP signal conveying such information, especially under naturalistic conditions. To address these questions, we recorded LFP responses to natural images in V1 of awake and anesthetized macaques using Utah multielectrode arrays. First, we have shown that it is possible to identify presented natural images from the LFP responses they evoke using trained Gabor wavelet (GW) models. Because GW models were devised to explain the spiking responses of V1 cells, this finding suggests that local spiking activity and LFPs (thought to reflect primarily local synaptic activity) carry similar visual information. Second, models trained on scalar metrics, such as the evoked LFP response range, provide robust image identification, supporting the informative nature of even simple LFP features. Third, image identification is robust only for the first 300 ms following image presentation, and image information is not restricted to any of the spectral bands. This suggests that the short-latency broadband LFP response carries most information during natural scene viewing. Finally, best image identification was achieved by GW models incorporating information at the scale of ∼ 0.5° in size and trained using four different orientations. This suggests that during natural image viewing, LFPs carry stimulus-specific information at spatial scales corresponding to few orientation columns in macaque V1.

Imaging activity in neurons and glia with a Polr2a-based and cre-dependent GCaMP5G-IRES-tdTomato reporter mouse.

New strategies for introducing genetically encoded activity indicators into animal models facilitate the investigation of nervous system function. We have developed the PC::G5-tdT mouse line that expresses the GCaMP5G calcium indicator in a Cre-dependent fashion. Instead of targeting the ROSA26 locus, we inserted the reporter cassette nearby the ubiquitously expressed Polr2a gene without disrupting locus integrity. The indicator was tagged with IRES-tdTomato to aid detection of positive cells. This reporter system is effective in a wide range of developmental and cellular contexts. We recorded spontaneous cortical calcium waves in intact awake newborns and evaluated concentration-dependent responses to odorants in the adult olfactory bulb. Moreover, PC::G5-tdT effectively reports intracellular calcium dynamics in somas and fine processes of astrocytes and microglial cells. Through electrophysiological and behavioral analyses, we determined that GCaMP5G expression had no major impact on nervous system performance. PC::G5-tdT will be instrumental for a variety of brain mapping experiments.

Distinct patterns of corticogeniculate feedback to different layers of the lateral geniculate nucleus.

In primates, feedforward visual pathways from retina to lateral geniculate nucleus (LGN) are segregated to different layers. These layers also receive strong reciprocal feedback pathways from cortex. The degree to which feedforward streams in primates are segregated from feedback streams remains unclear. Here, we asked whether corticogeniculate cells that innervate the magnocellular (M), parvocellular (P), and koniocellular (K) layers of the LGN in the prosimian primate bush baby (Otolemur garnettii) can be distinguished based on either the laminar distribution or morphological characteristics of their axons and synaptic contacts in LGN, or on their cell body position, size, and dendritic distribution in cortex. Corticogeniculate axons and synapses were labeled anterogradely with biotinylated dextran injections in layer 6 of cortex. Corticogeniculate cell bodies were first labeled with fluorescent dextran injections limited to individual M, P, or K LGN layers and then filled with biotinylated Lucifer yellow. Results showed that feedback to the M or P LGN layers arises from cells with dendrites primarily confined to cortical layer 6 and axons restricted to either M or P LGN layers, but not both. Feedback to K LGN layers arises from cells: 1) whose dendrites distribute rather evenly across cortical layers 5 and 6; 2) whose dendrites always extend into layer 4; and 3) whose axons are never confined to K layers but always overlap with either P or M layers. Corticogeniculate axons also showed distributions that were retinotopically precise based on known receptive field sizes of layer 6 cells, and these axons mainly made synapses with glutamatergic projection neurons in the LGN in all layers. Taken together with prior physiological results, we argue that the morphological differences between the three corticogeniculate pathways show that the M and P feedback pathways could rapidly and specifically enhance local LGN activity, while we speculate that the K feedback pathway is organized to temporally synchronize activity between LGN and cortex.

Two projection streams from macaque V1 to the pale cytochrome oxidase stripes of V2.

In the primate visual cortex, areas V1 and V2 distribute information they receive from the retina to virtually all extrastriate cortex, parsing this information into dorsal and ventral streams. Therefore, understanding the connectivity between V1 and V2 is crucial to understand visual cortical processing. Cytochrome oxidase staining in V2 reveals a repeating pattern of pale-thick-pale-thin stripes. V1 sends parallel output pathways to distinct V2 stripes. Previous models proposed either three or two parallel V1-to-V2 pathways in macaque, but both models viewed the two pale stripes within a single stripe cycle as a single compartment. However, recent studies have suggested that the two pale stripes may be functionally distinct, and in marmosets they also differ anatomically in the laminar origin of projections they receive from V1. Here we have asked whether the two pale stripes are also anatomically distinct in macaque. We made small retrograde tracer injections in different pale stripe types. We found that while both pale stripes receive a predominant V1 input from layers 2/3, only one set of pale stripes (pale lateral) receives significant projections from layer 4B, while the other set (pale medial) receives few or no layer 4B projections. Moreover, different tracer injections in nearby pale stripe types revealed that 97-99% of layer 2/3 cells only project to a single pale stripe type. These results demonstrate that in macaque, the two pale stripes are anatomically distinct compartments, and support the notion of two distinct projection streams from V1 to the two pale stripes of V2.

Different orientation tuning of near- and far-surround suppression in macaque primary visual cortex mirrors their tuning in human perception.

In primary visual cortex (V1), neuronal responses to stimuli inside the receptive field (RF) are usually suppressed by stimuli in the RF surround. This suppression is orientation specific. Similarly, in human vision surround stimuli can suppress perceived contrast of a central stimulus in an orientation-dependent manner. The surround consists of two regions likely generated by different circuits: a near-surround generated predominantly by geniculocortical and intra-V1 horizontal connections, and a far-surround generated exclusively by interareal feedback. Using stimuli confined to the near- or far-surround of V1 neurons, and similar stimuli in human psychophysics, we find that near-surround suppression is more sharply orientation tuned than far-surround suppression in both macaque V1 and human perception. These results point to a similarity between surround suppression in macaque V1 and human vision, and suggest that feedback circuits are less orientation biased than horizontal circuits. We find the sharpest tuning of near-surround suppression in V1 layers (3, 4B, 4Cα) with patterned and orientation-specific horizontal connections. Sharpest tuning of far-surround suppression occurs in layer 4B, suggesting greater orientation specificity of feedback to this layer. Different orientation tuning of near- and far-surround suppression may reflect a statistical bias in natural images, whereby nearby edges have higher probability than distant edges of being co-oriented and belonging to the same contour. Surround suppression would, thus, increase the coding efficiency of frequently co-occurring contours and the saliency of less frequent ones. Such saliency increase can help detect small orientation differences in nearby edges (for contour completion), but large orientation differences in distant edges (for directing saccades/attention).

High-resolution mapping of anatomical connections in marmoset extrastriate cortex reveals a complete representation of the visual field bordering dorsal V2.

The primate visual cortex consists of many areas. The posterior areas (V1, V2, V3, and middle temporal) are thought to be common to all primate species. However, the organization of cortex immediately anterior to area V2 (the "third tier" cortex) remains controversial, particularly in New World primates. The main point of contention has been whether the third tier cortex consists of a single area V3, representing lower and upper visual quadrants in dorsal and ventral cortex, respectively, or of 2 distinct areas (the dorsomedial [DM] area and a V3-like area). Resolving this controversy is crucial to understand the function and evolution of the third tier cortex. We have addressed this issue in marmosets, by performing high-precision mapping of corticocortical connections in cortex bordering dorsal V2. Multiple closely spaced neuroanatomical tracer injections were placed across the full width of dorsal V2 or adjacent anterior cortex, and the location of resulting labeled cells mapped throughout whole flattened visual cortex. The resulting topographic patterns of labeled connections allowed us to define areas and their boundaries. We found that a complete representation of the visual field borders dorsal V2 and that the third tier cortex consists of 2 distinct areas. These results unequivocally support the DM model.

Strong recurrent networks compute the orientation tuning of surround modulation in the primate primary visual cortex.

In macaque primary visual cortex (V1), neuronal responses to stimuli inside the receptive field (RF) are modulated by stimuli in the RF surround. This modulation is orientation specific. Previous studies suggested that, for some cells, this specificity may not be fixed but changes with the stimulus orientation presented to the RF. We demonstrate, in recording studies, that this tuning behavior is instead highly prevalent in V1 and, in theoretical work, that it arises only if V1 operates in a regime of strong local recurrence. Strongest surround suppression occurs when the stimuli in the RF and the surround are iso-oriented, and strongest facilitation when the stimuli are cross-oriented. This is the case even when the RF is suboptimally activated by a stimulus of nonpreferred orientation but only if this stimulus can activate the cell when presented alone. This tuning behavior emerges from the interaction of lateral inhibition (via the surround pathways), which is tuned to the preferred orientation of the RF, with weakly tuned, but strong, local recurrent connections, causing maximal withdrawal of recurrent excitation at the feedforward input orientation. Thus, horizontal and feedback modulation of strong recurrent circuits allows the tuning of contextual effects to change with changing feedforward inputs.

Contrast-dependence of surround suppression in Macaque V1: experimental testing of a recurrent network model.

Neuronal responses in primary visual cortex (V1) to optimally oriented high-contrast stimuli in the receptive field (RF) center are suppressed by stimuli in the RF surround, but can be facilitated when the RF center is stimulated at low contrast. The neural circuits and mechanisms for surround modulation are still unknown. We previously proposed that topdown feedback connections mediate suppression from the "far" surround, while "near' surround suppression is mediated primarily by horizontal connections. We implemented this idea in a recurrent network model of V1. A model assumption needed to account for the contrast-dependent sign of surround modulation is a response asymmetry between excitation and inhibition; accordingly, inhibition, but not excitation, is silent for weak visual inputs to the RF center, and surround stimulation can evoke facilitation. A prediction stemming from this same assumption is that surround suppression is weaker for low than for high contrast stimuli in the RF center. Previous studies are inconsistent with this prediction. Using single unit recordings in macaque V1, we confirm this model's prediction. Model simulations demonstrate that our results can be reconciled with those from previous studies. We also performed a systematic comparison of the experimentally measured surround suppression strength with predictions of the model operated in different parameter regimes. We find that the original model, with strong horizontal and no feedback excitation of local inhibitory neurons, can only partially account quantitatively for the experimentally measured suppression. Strong direct feedback excitation of V1 inhibitory neurons is necessary to account for the experimentally measured surround suppression strength.

Four projection streams from primate V1 to the cytochrome oxidase stripes of V2.

In the primate visual system, areas V1 and V2 distribute information they receive from the retina to all higher cortical areas, sorting this information into dorsal and ventral streams. Therefore, knowledge of the organization of projections between V1 and V2 is crucial to understand how the cortex processes visual information. In primates, parallel output pathways from V1 project to distinct V2 stripes. The traditional tripartite division of V1-to-V2 projections was recently replaced by a bipartite scheme, in which thin stripes receive V1 inputs from blob columns, and thick and pale stripes receive common input from interblob columns. Here, we demonstrate that thick and pale stripes, instead, receive spatially segregated V1 inputs and that the interblob is partitioned into two compartments: the middle of the interblob projecting to pale stripes and the blob/interblob border region projecting to thick stripes. Double-labeling experiments further demonstrate that V1 cells project to either thick or pale stripes, but rarely to both. We also find laminar specialization of V1 outputs, with layer 4B contributing projections mainly to thick stripes, and no projections to one set of pale stripes. These laminar differences suggest different contribution of magno, parvo, and konio inputs to each V1 output pathway. These results provide a new foundation for parallel processing models of the visual system by demonstrating four V1-to-V2 pathways: blob columns-to-thin stripes, blob/interblob border columns-to-thick stripes, interblob columns-to-pale(lateral) stripes, layer 2/3-4A interblobs-to-pale(medial) stripes.

Comparison of spatial summation properties of neurons in macaque V1 and V2.

In visual cortex, responses to stimulation of the receptive field (RF) are modulated by simultaneous stimulation of the RF surround. The mechanisms for surround modulation remain unidentified. We previously proposed that in the primary visual cortex (V1), near surround modulation is mediated by geniculocortical and horizontal connections and far surround modulation by interareal feedback connections. To understand spatial integration in the secondary visual cortex (V2) and its underlying circuitry, we have characterized spatial summation in different V2 layers and stripe compartments and compared it to that in V1. We used grating stimuli in circular and annular apertures of different sizes to estimate the extent and sensitivity of RF and surround components in V1 and V2. V2 RFs and surrounds were twice as large as those in V1. As in V1, V2 RFs doubled in size when measured at low contrast. In both V1 and V2, surrounds were about fivefold the size of the RF and the far surround could exceed 12.5 degrees in radius, averaging 5.5 degrees in V1 and 9.2 degrees in V2. The strength of surround suppression was similar in both areas. Thus although differing in spatial scale, the interactions among RF components are similar in V1 and V2, suggesting similar underlying mechanisms. As in V1, the extent of V2 horizontal connections matches that of the RF center, but is much smaller than the largest far surrounds, which likely derive from interareal feedback. In V2, we found no laminar or stripe differences in size and magnitude of surround suppression, suggesting conservation across stripes of the basic circuit for surround modulation.

Anatomical evidence for classical and extra-classical receptive field completion across the discontinuous horizontal meridian representation of primate area V2.

In primates, a split of the horizontal meridian (HM) representation at the V2 rostral border divides this area into dorsal (V2d) and ventral (V2v) halves (representing lower and upper visual quadrants, respectively), causing retinotopically neighboring loci across the HM to be distant within V2. How is perceptual continuity maintained across this discontinuous HM representation? Injections of neuroanatomical tracers in marmoset V2d demonstrated that cells near the V2d rostral border can maintain retinotopic continuity within their classical and extra-classical receptive field (RF), by making both local and long-range intra- and interareal connections with ventral cortex representing the upper visual quadrant. V2d neurons located <0.9-1.3 mm from the V2d rostral border, whose RFs presumably do not cross the HM, make nonretinotopic horizontal connections with V2v neurons in the supra- and infragranular layers. V2d neurons located <0.6-0.9 mm from the border, whose RFs presumably cross the HM, in addition make retinotopic local connections with V2v neurons in layer 4. V2d neurons also make interareal connections with upper visual field regions of extrastriate cortex, but not of MT or MTc outside the foveal representation. Labeled connections in ventral cortex appear to represent the "missing" portion of the connectional fields in V2d across the HM. We conclude that connections between dorsal and ventral cortex can create visual field continuity within a second-order discontinuous visual topography.

Response facilitation from the "suppressive" receptive field surround of macaque V1 neurons.

In primary visual cortex (V1), neuronal responses to optimally oriented stimuli in the receptive field (RF) center are usually suppressed by iso-oriented stimuli in the RF surround. The mechanisms and pathways giving rise to surround modulation, a possible neural correlate of perceptual figure-ground segregation, are not yet identified. We previously proposed that highly divergent and fast-conducting top-down feedback connections are the substrate for fast modulation arising from the more distant regions of the surround. We have recently implemented this idea into a recurrent network model (Schwabe et al. 2006). The purpose of this study was to test a crucial prediction of this feedback model, namely that the suppressive "far" surround of V1 neurons can be facilitatory under conditions that weakly activate neurons in the RF center. Using single-unit recordings in macaque V1, we found iso-orientation far-surround facilitation when the RF center was driven by a low-contrast stimulus and the far surround by a small annular stimulus. Suppression occurred when the center stimulus contrast or the size of the surround stimulus was increased. This suggests that center-surround interactions result from excitatory and inhibitory mechanisms of similar spatial extent, and that changes in the balance of local excitation and inhibition, induced by surround stimulation, determine whether facilitation or suppression occurs. In layer 4C, the main target of geniculocortical afferents, lacking long-range intra-cortical connections, far-surround facilitation was rare and large surround fields were absent. This strongly suggests that feedforward connections do not contribute to far-surround modulation and that the latter is generated by intra-cortical mechanisms, likely involving top-down feedback.

Organization of the feedback pathway from striate cortex (V1) to the lateral geniculate nucleus (LGN) in the owl monkey (Aotus trivirgatus).

The function of the corticogeniculate feedback pathway from the striate cortex (V1) to the lateral geniculate nucleus (LGN) in primates is not well understood. Insight into possible function can be gained by studying the morphology and projection patterns of corticogeniculate axons in the LGN. The goal of this research was to examine how corticogeniculate axons innervate the functionally specific (e.g., parvocellular [P], magnocellular [M], and koniocellular [K]) and eye-specific layers of the LGN. Pressure injections of biotinylated dextran were made into owl monkey V1, and the resulting labeled axons were reconstructed through serial sections of the LGN. All of the corticogeniculate axons, regardless of termination pattern, were thin with boutons en passant or at the ends of small stalks, as described in cats. Axons were found in all layers of the LGN, and two main patterns of innervation were observed. In the first pattern, axons terminated in individual M or P LGN layers. In the second pattern of innervation, axons terminated in pairs of functionally matched layers. Examples of this type were seen within pairs of M, P, or K layers. In most cases, both classes of axons contain arbors focused within the P or M layers but also had collateral side branches in neighboring K layers. Unlike corticogeniculate axons seen in the cat, corticogeniculate axons in the owl monkey maintained topographic innervation in the LGN layers that was consistent with receptive field sizes represented in V1. The patterns of layer projections along with the retinotopic match of corticogeniculate axons within the LGN suggest that in primates V1 can modulate activity in the LGN through functionally specific projections in a more tightly tuned retinotopic fashion than previously believed.