Ko Sakai

Surrounding suppression and facilitation in the determination of border ownership

Contextual modulation reported in early- to intermediate-level visual areas could be an essential component to signal border ownership (BO) that specifies the direction of figure along a contour. The surrounding regions that evoke significant suppression or facilitation are highly localized and asymmetric with respect to the center of the classical receptive field (CRF). We propose a hypothesis that such surrounding modulation is a basis for BO-selectivity. Although this idea has been discussed for several years, it is uncertain how many of a vast variety of surrounding organizations could signal correctly the direction of ownership, and how many could signal consistently for various stimuli. We carried out computationally a population study of the surrounding effects to investigate how many cells exhibit effective and consistent BO signals. We tested hundreds of various organizations, and found that most of the asymmetric, iso-orientation suppressive regions, regardless of position or size, lead to surprisingly high consistency in the direction of ownership for various stimuli. The combinations of iso-orientation suppression and cross-orientation facilitation indicate both high robustness and consistency in the ownership determination. We constructed a model for BO-selective neurons based on the surrounding effects, and investigated whether the model reproduces major characteristics of the neuronal responses, including a variety in the BO selectivity among neurons, consistency with respect to various stimuli, invariance to stimulus size, and co-selectivity to BO and contrast. The model reproduced successfully the major characteristics of BO-selective neurons.

Characterization of the spatial-frequency spectrum in the perception of shape from texture

The major cue to shape from texture is the compression of texture as a function of surface curvature [J. Exp. Psychol. 13, 242 (1987); Vision Res. 33, 827 (1993)]. A number of computational models have been proposed in which compression is measured by detection of changes in the spatial-frequency spectrum [Comput. Graphics Image Process. 5, (1976)]. We propose that the visual system uses a strategy of characterizing the frequency spectrum by a simple set of measures and of tracking the changes in this characterization rather than determining changes in the shape of the actual spectra. Our evidence is based on a number of psychophysical demonstrations that use stimuli with specifically tailored frequency spectra, constructed from white noise filtered in the frequency domain. Our evidence suggests that the visual system determines the average peak frequency of the spectrum and uses this measure as its characterization. Changes in fp are strongly correlated with the degree of surface curvature, and, over a range of stimuli, fp takes account of the variance in local estimates of the frequency spectrum. One computes fp by determining the peak frequency at each spatial location and then averaging these frequency values over a local spatial region. We show that fp is related to the second-order moment but is more biologically plausible and shows superior ability to function in the presence of noise. As a test of this model, we have constructed a neural network architecture for computing shape from texture. Our model is limited to orthographically projected, homogeneous textures without in-surface rotation. The early stages of the model consist of multiple simple-cell units tuned to different orientations and spatial frequencies. We show that these simple cells are inadequate for the determination of compression but that the outputs of complex-cell-like units after normalization generate estimates of surface slant and tilt. The network shows qualitative agreement with human perception of shape from texture over a wide range of real and artificial stimuli.

Determination of border ownership based on the surround context of contrast

Abstract We investigate the neural mechanisms for border-ownership (BO) determination, specifically whether the determination of BO is plausible from the contrast configuration within a certain range that extends beyond the classical receptive fields. The relevance of the contrast is suggested since the majority of BO-selective neurons in V2 and V4 show co-selectivity to the contrast. We hypothesize that the spatial structure of surrounding inhibition/excitation recently revealed in V1, or the similar structure in V2, is a key to integrate surrounding contrast to determine BO. The model reproduces a range of the neuronal activities responding to complex figures including occlusion.

Consistent and robust determination of border ownership based on asymmetric surrounding contrast

Determination of the figure region in an image is a fundamental step toward surface construction, shape coding, and object representation. Localized, asymmetric surround modulation, reported neurophysiologically in early-to-intermediate-level visual areas, has been proposed as a mechanism for figure-ground segregation. We investigated, computationally, whether such surround modulation is capable of yielding consistent and robust determination of figure side for various stimuli. Our surround modulation model showed a surprisingly high consistency among pseudorandom block stimuli, with greater consistency for stimuli that yielded higher accuracy of, and shorter reaction times in, human perception. Our analyses revealed that the localized, asymmetric organization of surrounds is crucial in the detection of the contrast imbalance that leads to the determination of the direction of figure with respect to the border. The model also exhibited robustness for gray-scaled natural images, with a mean correct rate of 67%, which was similar to that of figure-side determination in human perception through a small window and of machine-vision algorithms based on local processing. These results suggest a crucial role of surround modulation in the local processing of figure-ground segregation.

Spatial attention in early vision for the perception of border ownership

Spatial attention alters contrast gain in early visual areas, which might affect the determination of border ownership (BO) that indicates the direction of figure with respect to the border. We investigated the role of spatial attention applied to early vision in the determination of BO with a computational model that consists of V1, V2, and posterior parietal (PP) modules. Attention alters contrast gain in the V1 module so that it enhances local contrast. The V2 module determines BO based on the surrounding contrast extracted by the V1 module. The simulation results showed that the attention significantly modulates BO; BO is even flipped in figures with ambiguous BO while BO is stable for unambiguous figures such as a simple square. To evaluate the model quantitatively, we carried out psychophysical experiments to measure the effects of attention in the perception of BO and compared the results with those from corresponding simulations. The model showed good agreement with human perception including the determination of BO for ambiguous random-block stimuli. These results indicate that the activity of BO-selective neurons could be modulated significantly by spatial attention that alters local contrast gain in V1, which may account in part for automatic, bi-stable perception in ambiguous figures.

Neural grouping and geometric effect in the determination of apparent orientation

We propose that neural grouping of retinotopically distributed responses in the primary visual cortex (V1) is essential for the determination of apparent tilt, including the tilt illusion. Our psychophysical study shows that apparent tilt is independent of stereo disparity, hue, or contrast of bars, which determine the ownership of their intersection. This leads us to suspect that the neuronal responses within the intersection are excluded from the computation of apparent tilt. To investigate the underlying cortical mechanisms, we developed and examined a V1 network model including the collinear connections observed in the superficial layers. The model shows good agreement with the results of psychophysical experiments, including segmentation independence, contrast dependence, and apparent tilt for various stimuli. The results suggest that collinear connections underlie the neural grouping that excludes the intersection region and establishes the independence of segmentation.

The computational model for border-ownership determination consisting of surrounding suppression and facilitation in early vision

We have proposed the computational model for border-ownership (BO) determination based on the contrast configurations within a certain range that extends beyond the classical receptive field (CRF). In this study, we adopt two crucial functions of the surrounding modulation reported by the recent physiological studies; (1) changes in functional connection depending on the amplitude of contrast, and (2) a variety of surrounding suppression/facilitation depending on the orientation and retinotopic position of surrounding stimuli relative to the CRF. Simulation results show that the model reproduces the major characteristics of BO selective neurons.

Early representation of shape by onset synchronization of border-ownership-selective cells in the V1-V2 network

Construction of surface is a crucial step toward the representation of shape through the integration of local information. Physiological studies have reported that the primary visual cortex (V1) codes the medial axis (MA) that is a skeletal structure equidistant from nearby contours, suggesting the early representation of surface in V1. Although the neural basis of surface construction has not been clarified, the onset synchronization of border ownership (BO)-selective cells is a plausible candidate for the generation of surface. We investigated computationally the representation of surface in a biophysically detailed model of primate V1-V2 networks. The simulation results showed that the simultaneous arrival of signals from BO-selective cells evoked strong responses of V1 cells located around the MA. The simulation results lead to a prediction that the perception of the direction of figure (DOF) depends on the degree of synchronous presentation of contour. We conducted a psychophysical experiment and showed that the perception of the DOF is biased toward a highly synchronized contour. These results suggest a crucial role of the onset synchronization of BO-selective cells for the construction of early representation of shape.

A shape-from-texture algorithm based on human visual psychophysics

We propose an algorithm for the determination of three dimensional shape and perspective based on the response of the human visual system to changes in visual textures. Current computer vision algorithms are computationally intensive and show inherent difficulties in integrating additional cues for the determination of shape, such as shading, contour, or motion. In order to develop a fast and simple mechanism less constrained for integrating other cues, we incorporated aspects of the physiological properties of cortical cells in VI into a network model. We provide psychophysical evidence that the local spatial frequency spectrum is represented by the spatially averaged peak frequency (APF). After normalization, this APF measures texture compression and leads to estimates of 3D shape and depth. Simulations of the model show good agreement with human responses to a range of textured images.<<ETX>>

Border-ownership-dependent tilt aftereffect in incomplete figures

A recent physiological finding of neural coding for border ownership (BO) that defines the direction of a figure with respect to the border has provided a possible basis for figure-ground segregation. To explore the underlying neural mechanisms of BO, we investigated stimulus configurations that activate BO circuitry through psychophysical investigation of the BO-dependent tilt aftereffect (BO-TAE). Specifically, we examined robustness of the border ownership signal by determining whether the BO-TAE is observed when gestalt factors are broken. The results showed significant BO-TAEs even when a global shape was not explicitly given due to the ambiguity of the contour, suggesting a contour-independent mechanism for BO coding.