Context-sensitive binding by the laminar circuits of V1 and V2: A unified model of perceptual grouping, attention, and orientation contrast

Abstract

A detailed neural model is presented of how the laminar circuits of visual cortical areas V1 and V2 implement context-sensitive binding processes such as perceptual grouping and attention. The model proposes how specific laminar circuits allow the responses of visual cortical neurons to be determined not only by the stimuli within their classical receptive fields, but also to be strongly influenced by stimuli in the extra-classical surround. This context-sensitive visual processing can greatly enhance the analysis of visual scenes, especially those containing targets that are low contrast, partially occluded, or crowded by distractors. We show how interactions of feedforward, feedback and horizontal circuitry can implement several types of contextual processing simultaneously, using shared laminar circuits. In particular, we present computer simulations which suggest how top-down attention and preattentive perceptual grouping, two processes that are fundamental for visual binding, can interact, with attentional enhancement selectively propagating along groupings of both real and illusory contours, thereby showing how attention can selectively enhance object representations. These simulations also illustrate how attention may have a stronger facilitatory effect on low contrast than on high contrast stimuli, and how pop-out from orientation contrast may occur. The specific functional roles which the model proposes for the cortical layers allow several testable neurophysiological predictions to be made. The results presented here simulate only the boundary grouping system of adult cortical architecture. Howeve1~ we also discuss how this model contributes to a larger neural theory of vision which suggests how intracortical and intercortical feedback help to stabilize development and learning within these cortical circuits. Although feedback plays a key role, fast feedforward processing is possible in response to unambiguous information. Model circuits are capable of synchronizing quickly, but context-sensitive persistence of previous events can influence how synchrony develops. Although these results focus on how the interblob cortical processing stream controls boundary grouping and attention, related modeling of the blob cortical processing stream suggests how visible surfaces arc formed, and modeling of the motion stream suggests how transient responses to scenic changes can control long-range apparent motion and also attract spatial attention.

Running as fast as it can: How spiking dynamics form object groupings in the laminar circuits of visual cortex

“…] Laminar cortical models enable a finer analysis of intracortical processing. The circuitry of the LAMINART model (Figure 2; Grossberg, 1999Grossberg, , 2003Grossberg, , 2007Grossberg, Mingolla & Ross 1997;Grossberg & Raizada 2000;Raizada & Grossberg 2001, 2003Yazdanbakhsh & Grossberg 2004) shows how a fast feedforward sweep of activation throughout a cortical hierarchy could occur in response to unambiguous information, consistent with the results of Thorpe et al (2001); see Figure 2e wherein layers 4-to-2/3 in one cortical area project to layers 4-to-2/3 in the next, and so on. However, in response to ambiguous information, self-normalizing competition among alternative cortical interpretations of the data may weaken the activation amplitude and coherence of each alternative, thereby slowing down its processing, and enabling interlaminar, but intracortical, feedback (Figure 2c) to contrast-enhance and thereby choose the alternatives that are supported by the most evidence, thereupon automatically speeding up processing of those choices.…”

“…For example, activation of smaller-scale bipole cells in V1 is modulated by feedback from larger-scale bipole cells in V2, such that contextual elements at a larger spatial scale may sharpen the activity at the level of V1 bipole cells (Anzai et al 2007;Grossberg and Raizada 2000;Grossberg and Swaminathan 2004;Kisvárday et al 1997;Raizada and Grossberg 2001;Shmuel et al 2005). Although this article models neurophysiological data about V1 horizontal interactions, the similarity of V1 and V2 circuits with respect to the presence of anisotropic horizontal collaterals suggests that variants of our results may also apply to V2.…”

“…Each layer 2/3 inhibitory interneuron inhibits the layer 2/3 pyramidal cell and the antagonist interneuron in the same hypercolumn. This inhibitory scheme is designed to realize the bipole property within a laminar cortical circuit (Grossberg and Raizada 2000;Raizada and Grossberg 2001).…”

Linking Attention to Learning, Expectation, Competition, and Consciousness

“…LAMINART also simulates the spread of attention along an illusory contour (Raizada and Grossberg, 2001), consistent with experimental data of Moore, Yantis, and Vaughan (1998), thereby illustrating how the cortex can attend incomplete object data.…”

Attention in hierarchical models of object recognition

“…However, there are other ideas that are not captured by the UNI framework, either because they describe the neurobiological processes at a more detailed level than is provided by the UNI framework, or because they make use of information from sources that fall outside the framework, such as visual context. Grossberg and Raizada (2000) and Raizada and Grossberg (2001), for instance, proposed a model of attention and visual grouping based on biologically realistic models of neurons and neural networks. Their model relies on grouping edges within a laminar cortical structure by synchronous firing, allowing it to extract real as well as illusory contours.…”