Anita M. Schmid

Sparse coding and high-order correlations in fine-scale cortical networks

Connectivity in the cortex is organized at multiple scales, suggesting that scale-dependent correlated activity is particularly important for understanding the behaviour of sensory cortices and their function in stimulus encoding. We analysed the scale-dependent structure of cortical interactions by using maximum entropy models to characterize multiple-tetrode recordings from primary visual cortex of anaesthetized macaque monkeys (Macaca mulatta). We compared the properties of firing patterns among local clusters of neurons (<300 μm apart) with those of neurons separated by larger distances (600–2,500 μm). Here we report that local firing patterns are distinctive: whereas multi-neuronal firing patterns at larger distances can be predicted by pairwise interactions, patterns within local clusters often show evidence of high-order correlations. Surprisingly, these local correlations are flexible and rapidly reorganized by visual input. Although they modestly reduce the amount of information that a cluster conveys, they also modify the format of this information, creating sparser codes by increasing the periods of total quiescence, and concentrating information into briefer periods of common activity. These results imply a hierarchical organization of neuronal correlations: simple pairwise correlations link neurons over scales of tens to hundreds of minicolumns, but on the scale of a few minicolumns, ensembles of neurons form complex subnetworks whose moment-to-moment effective connectivity is dynamically reorganized by the stimulus.

Subpopulations of neurons in visual area v2 perform differentiation and integration operations in space and time.

The interconnected areas of the visual system work together to find object boundaries in visual scenes. Primary visual cortex (V1) mainly extracts oriented luminance boundaries, while secondary visual cortex (V2) also detects boundaries defined by differences in texture. How the outputs of V1 neurons are combined to allow for the extraction of these more complex boundaries in V2 is as of yet unclear. To address this question, we probed the processing of orientation signals in single neurons in V1 and V2, focusing on response dynamics of neurons to patches of oriented gratings and to combinations of gratings in neighboring patches and sequential time frames. We found two kinds of response dynamics in V2, both of which were different from those of V1 neurons. While V1 neurons in general preferred one orientation, one subpopulation of V2 neurons (“transient”) showed a temporally dynamic preference, resulting in a preference for changes in orientation. The second subpopulation of V2 neurons (“sustained”) responded similarly to V1 neurons, but with a delay. The dynamics of nonlinear responses to combinations of gratings reinforced these distinctions: the dynamics enhanced the preference of V1 neurons for continuous orientations and the preference of V2 transient neurons for discontinuous ones. We propose that transient neurons in V2 perform a differentiation operation on the V1 input, both spatially and temporally, while the sustained neurons perform an integration operation. We show that a simple feedforward network with delayed inhibition can account for the temporal but not for the spatial differentiation operation.

Visual processing of informative multipoint correlations arises primarily in V2

Using the visual system as a model, we recently showed that the efficient coding principle accounted for the allocation of computational resources in central sensory processing: when sampling an image is the main limitation, resources are devoted to compute the statistical features that are the most variable, and therefore the most informative (eLife 2014;3:e03722. DOI: 10.7554/eLife.03722 Hermundstad et al., 2014). Building on these results, we use single-unit recordings in the macaque monkey to determine where these computations—sensitivity to specific multipoint correlations—occur. We find that these computations take place in visual area V2, primarily in its supragranular layers. The demonstration that V2 neurons are sensitive to the multipoint correlations that are informative about natural images provides a common computational underpinning for diverse but well-recognized aspects of neural processing in V2, including its sensitivity to corners, junctions, illusory contours, figure/ground, and ‘naturalness.’ DOI: http://dx.doi.org/10.7554/eLife.06604.001

Responses to Orientation Discontinuities in V1 and V2: Physiological Dissociations and Functional Implications

Segmenting the visual image into objects is a crucial stage of visual processing. Object boundaries are typically associated with differences in luminance, but discontinuities in texture also play an important role. We showed previously that a subpopulation of neurons in V2 in anesthetized macaques responds to orientation discontinuities parallel to their receptive field orientation. Such single-cell responses could be a neurophysiological correlate of texture boundary detection. Neurons in V1, on the other hand, are known to have contextual response modulations such as iso-orientation surround suppression, which also produce responses to orientation discontinuities. Here, we use pseudorandom multiregion grating stimuli of two frame durations (20 and 40 ms) to probe and compare texture boundary responses in V1 and V2 in anesthetized macaque monkeys. In V1, responses to texture boundaries were observed for only the 40 ms frame duration and were independent of the orientation of the texture boundary. However, in transient V2 neurons, responses to such texture boundaries were robust for both frame durations and were stronger for boundaries parallel to the neurons preferred orientation. The dependence of these processes on stimulus duration and orientation indicates that responses to texture boundaries in V2 arise independently of contextual modulations in V1. In addition, because the responses in transient V2 neurons are sensitive to the orientation of the texture boundary but those of V1 neurons are not, we suggest that V2 responses are the correlate of texture boundary detection, whereas contextual modulation in V1 serves other purposes, possibly related to orientation “pop-out.”

Laminar and Orientation-Dependent Characteristics of Spatial Nonlinearities: Implications for the Computational Architecture of Visual Cortex

A full understanding of the computations performed in primary visual cortex is an important yet elusive goal. Receptive field models consisting of cascades of linear filters and static nonlinearities may be adequate to account for responses to simple stimuli such as gratings and random checkerboards, but their predictions of responses to complex stimuli such as natural scenes are only approximately correct. It is unclear whether these discrepancies are limited to quantitative inaccuracies that reflect well-recognized mechanisms such as response normalization, gain controls, and cross-orientation suppression or, alternatively, imply additional qualitative features of the underlying computations. To address this question, we examined responses of V1 and V2 neurons in the monkey and area 17 neurons in the cat to two-dimensional Hermite functions (TDHs). TDHs are intermediate in complexity between traditional analytic stimuli and natural scenes and have mathematical properties that facilitate their use to test candidate models. By exploiting these properties, along with the laminar organization of V1, we identify qualitative aspects of neural computations beyond those anticipated from the above-cited model framework. Specifically, we find that V1 neurons receive signals from orientation-selective mechanisms that are highly nonlinear: they are sensitive to phase correlations, not just spatial frequency content. That is, the behavior of V1 neurons departs from that of linear-nonlinear cascades with standard modulatory mechanisms in a qualitative manner: even relatively simple stimuli evoke responses that imply complex spatial nonlinearities. The presence of these findings in the input layers suggests that these nonlinearities act in a feedback fashion.

Cannabinoid Neuromodulation in the Adult Early Visual Cortex

Sensory processing is an active process involving the interaction of ongoing cortical activity with incoming stimulus information. However, the modulators and circuits involved in this interaction are incompletely understood. One potential candidate is the cannabinoid-signaling system, which is known to modulate the dynamics of cortical networks. Here, we show that in the primate primary and secondary visual cortices, the cannabinoid CP55940 modulates not only population dynamics but also influences the dynamics of the stimulus-response relationship of individual neurons. At the population level, CP55940 decreases EEG power, LFP power, and LFP coherence. At the single-neuron level, intrinsic spike train dynamics appear relatively unchanged, but visual receptive fields are altered: CP55940 induced an overall delay and broadening of the temporal component of V1 and V2 spatiotemporal receptive fields. Our findings provide neurophysiologic evidence for a link between cannabinoid-signaling, network dynamics and the function of a canonical cortical circuit.

Modulation of early cortical processing during divided attention to non-contiguous locations

We often face the challenge of simultaneously attending to multiple non‐contiguous regions of space. There is ongoing debate as to how spatial attention is divided under these situations. Whereas, for several years, the predominant view was that humans could divide the attentional spotlight, several recent studies argue in favor of a unitary spotlight that rhythmically samples relevant locations. Here, this issue was addressed by the use of high‐density electrophysiology in concert with the multifocal m‐sequence technique to examine visual evoked responses to multiple simultaneous streams of stimulation. Concurrently, we assayed the topographic distribution of alpha‐band oscillatory mechanisms, a measure of attentional suppression. Participants performed a difficult detection task that required simultaneous attention to two stimuli in contiguous (undivided) or non‐contiguous parts of space. In the undivided condition, the classic pattern of attentional modulation was observed, with increased amplitude of the early visual evoked response and increased alpha amplitude ipsilateral to the attended hemifield. For the divided condition, early visual responses to attended stimuli were also enhanced, and the observed multifocal topographic distribution of alpha suppression was in line with the divided attention hypothesis. These results support the existence of divided attentional spotlights, providing evidence that the corresponding modulation occurs during initial sensory processing time‐frames in hierarchically early visual regions, and that suppressive mechanisms of visual attention selectively target distracter locations during divided spatial attention.

Orientation change detection and orientation pooling in space and time performed by two subpopulations of neurons in V2

In natural images, some boundaries are de ned by luminance di erences; others are de ned by texture di erences. Most neurons in primary visual cortex (V1) are well-driven by luminance boundaries at the appropriate orientation1-2. Boundaries de ned by di erences in texture, however, are more e ective stimuli for neurons in the secondary visual cortex (V2)3-9. Since the larger receptive elds of V2 are produced by combining the output of V1 neurons10-12, the extraction of texture boundaries by V2 receptive elds must involve computations on its V1 inputs across space. These computations must accomplish a speci c goal – extraction of texture boundaries – while preserving the luminance-boundary information already extracted by V1.