Figure-ground Processing Research Articles

Functional recursion of orientation cues in figure-ground separation

Visual texture is an important cue to figure-ground organization. While processing of texture differences is a prerequisite for the use of this cue to extract figure-ground organization, these stages are distinct processes. One potential indicator of this distinction is the possibility that texture statistics play a different role in the figure vs. in the ground. To determine whether this is the case, we probed figure-ground processing with a family of local image statistics that specified textures that varied in the strength and spatial scale of structure, and the extent to which features are oriented. For image statistics that generated approximately isotropic textures, the threshold for identification of figure-ground structure was determined by the difference in correlation strength in figure vs. ground, independent of whether the correlations were present in figure, ground, or both. However, for image statistics with strong orientation content, thresholds were up to two times higher for correlations in the ground, vs. the figure. This held equally for texture-defined objects with convex or concave boundaries, indicating that these threshold differences are driven by border ownership, not boundary shape. Similar threshold differences were found for presentation times ranging from 125 to 500 ms. These findings identify a qualitative difference in how texture is used for figure-ground analysis, vs. texture discrimination. Additionally, it reveals a functional recursion: texture differences are needed to identify tentative boundaries and consequent scene organization into figure and ground, but then scene organization modifies sensitivity to texture differences according to the figure-ground assignment.

Resolving the Spatial Profile of Figure Enhancement in Human V1 through Population Receptive Field Modeling.

The detection and segmentation of meaningful figures from their background is one of the primary functions of vision. While work in nonhuman primates has implicated early visual mechanisms in this figure-ground modulation, neuroimaging in humans has instead largely ascribed the processing of figures and objects to higher stages of the visual hierarchy. Here, we used high-field fMRI at 7 Tesla to measure BOLD responses to task-irrelevant orientation-defined figures in human early visual cortex (N = 6, four females). We used a novel population receptive field mapping-based approach to resolve the spatial profiles of two constituent mechanisms of figure-ground modulation: a local boundary response, and a further enhancement spanning the full extent of the figure region that is driven by global differences in features. Reconstructing the distinct spatial profiles of these effects reveals that figure enhancement modulates responses in human early visual cortex in a manner consistent with a mechanism of automatic, contextually driven feedback from higher visual areas.SIGNIFICANCE STATEMENT A core function of the visual system is to parse complex 2D input into meaningful figures. We do so constantly and seamlessly, both by processing information about visible edges and by analyzing large-scale differences between figure and background. While influential neurophysiology work has characterized an intriguing mechanism that enhances V1 responses to perceptual figures, we have a poor understanding of how the early visual system contributes to figure-ground processing in humans. Here, we use advanced computational analysis methods and high-field human fMRI data to resolve the distinct spatial profiles of local edge and global figure enhancement in the early visual system (V1 and LGN); the latter is distinct and consistent with a mechanism of automatic, stimulus-driven feedback from higher-level visual areas.

On the relationship between mathematics and visuospatial processing in Turner syndrome

A common neurocognitive phenotype of Turner syndrome (TS) includes coincident deficits in math and visuospatial reasoning while overall IQ remains intact. However, research has highlighted disparities in the relationship between these properties in women with TS, suggesting that not all visuospatial domains are equally related to mathematics in this group. Here, we present findings from a longitudinal investigation of visuospatial processing and its relationship to math performance in adolescent girls with TS and age-matched healthy controls. Participants completed a standardized battery of math and visuospatial tests once a year for 4 years. Linear mixed effects modeling was used to examine the relationship between mathematics and each visuospatial domain over time. Our results indicate that math performance was related to visual tracking, visual-motor coordination, and figure-ground processing. Such visuospatial domains appear to be uniquely affected by TS and could contribute to their deficits in math performance. Furthermore, differences in math and visuospatial test performance between girls with TS and healthy controls remain stable over time. Our results have important implications for the role of visuospatial processing in early math performance and may inform the development of effective interventions aimed at improving math education in children with TS.

Figure-Ground Processing: A Reassessment of Gelb and Granit.

In 1923, Adhemar Gelb and Ragnar Granit, two prominent researchers in early Gestalt perceptual theory, reported a lower threshold for detection of a target (a small colored dot) on the ground region of an image than on an adjacent figural region. Although their results had a wide influence on the understanding of figure-ground perception, they are at odds with more recent investigations in which figural regions appear to have a processing advantage over ground regions. The two present studies replicated Gelb and Granit's experiment using a similar figure-ground stimulus albeit with a two-alternative forced choice procedure rather than their original method of adjustment. Experiment 1 found that, contrary to Gelb and Granit's findings, a detection advantage was found for the figural over the ground region. Experiment 2 indicated that explicit contours might have played a role in detection.

Figure-Ground Processing during Fixational Saccades in V1: Indication for Higher-Order Stability

In a typical visual scene we continuously perceive a "figure" that is segregated from the surrounding "background" despite ongoing microsaccades and small saccades that are performed when attempting fixation (fixational saccades [FSs]). Previously reported neuronal correlates of figure-ground (FG) segregation in the primary visual cortex (V1) showed enhanced activity in the "figure" along with suppressed activity in the noisy "background." However, it is unknown how this FG modulation in V1 is affected by FSs. To investigate this question, we trained two monkeys to detect a contour embedded in a noisy background while simultaneously imaging V1 using voltage-sensitive dyes. During stimulus presentation, the monkeys typically performed 1-3 FSs, which displaced the contour over the retina. Using eye position and a 2D analytical model to map the stimulus onto V1, we were able to compute FG modulation before and after each FS. On the spatial cortical scale, we found that, after each FS, FG modulation follows the stimulus retinal displacement and "hops" within the V1 retinotopic map, suggesting visual instability. On the temporal scale, FG modulation is initiated in the new retinotopic position before it disappeared from the old retinotopic position. Moreover, the FG modulation developed faster after an FS, compared with after stimulus onset, which may contribute to visual stability of FG segregation, along the timeline of stimulus presentation. Therefore, despite spatial discontinuity of FG modulation in V1, the higher-order stability of FG modulation along time may enable our stable and continuous perception.

Face inversion reduces the persistence of global form and its neural correlates.

Face inversion produces a detrimental effect on face recognition. The extent to which the inversion of faces and other kinds of objects influences the perceptual binding of visual information into global forms is not known. We used a behavioral method and functional MRI (fMRI) to measure the effect of face inversion on visual persistence, a type of perceptual memory that reflects sustained awareness of global form. We found that upright faces persisted longer than inverted versions of the same images; we observed a similar effect of inversion on the persistence of animal stimuli. This effect of inversion on persistence was evident in sustained fMRI activity throughout the ventral visual hierarchy, including the lateral occipital area (LO), two face-selective visual areas—the fusiform face area (FFA) and the occipital face area (OFA)—and several early visual areas. V1 showed the same initial fMRI activation to upright and inverted forms but this activation lasted longer for upright stimuli. The inversion effect on persistence-related fMRI activity in V1 and other retinotopic visual areas demonstrates that higher-tier visual areas influence early visual processing via feedback. This feedback effect on figure-ground processing is sensitive to the orientation of the figure.

The effect of recognizability on figure–ground processing: Does it affect parsing or only figure selection?

Though figure-ground assignment has been shown to be probably affected by recognizability, it appears sensible that object recognition must follow at least the earlier process of figure-ground segregation. To examine whether or not rudimentary object recognition could, counterintuitively, start even before the completion of the stage of parsing in which figure-ground segregation is done, participants were asked to respond, in a go/no-go fashion, whenever any out of 16 alternative connected patterns (that constituted familiar stimuli in the upright orientation) appeared. The white figure of the to-be-attended stimulus-target or foil-could be segregated from the white ambient ground only by means of a frame surrounding it. Such a frame was absent until the onset of target display. Then, to manipulate organizational quality, the greyness of the frame was either gradually increased from zero (in Experiment 1) or changed abruptly to a stationary level whose greyness was varied between trials (in Experiments 2 and 3). Stimulus recognizability was manipulated by orientation angle. In all three experiments the effect of recognizability was found to be considerably larger when organizational quality was minimal due to an extremely faint frame. This result is argued to be incompatible with any version of a serial thesis suggesting that processing aimed at object recognition starts only with a good enough level of organizational quality. The experiments rather provide some support to the claim, termed here "early interaction hypothesis", positing interaction between early recognition processing and preassignment parsing processes.

Rapid visual grouping and figure–ground processing using temporally structured displays

We examine the time course of visual grouping and figure–ground processing. Figure (contour) and ground (random-texture) elements were flickered with different phases (i.e., contour and background are alternated), requiring the observer to group information within a pre-specified time window. It was found this grouping has a high temporal resolution: less than 20 ms for smooth contours, and less than 50 ms for line conjunctions with sharp angles. Furthermore, the grouping process takes place without an explicit knowledge of the phase of the elements, and it requires a cumulative build-up of information. The results are discussed in relation to the neural mechanism for visual grouping and figure–ground segregation.

Grouping puts figure-ground assignment in context by constraining propagation of edge assignment

Figure-ground organization involves the assignment of edges to a figural shape on one or the other side of each dividing edge. Established visual cues for edge assignment primarily concern relatively local rather than contextual factors. In the present article, we show that an assignment for a locally unbiased edge can be affected by an assignment of a remote contextual edge that has its own locally biased assignment. We find that such propagation of edge assignment from the biased remote context occurs only when the biased and unbiased edges are grouped. This new principle, whereby grouping constrains the propagation of figural edge assignment, emerges from both subjective reports and an objective short-term edge-matching task. It generalizes from moving displays involving grouping by common fate and collinearity, to static displays with grouping by similarity of edge-contrast polarity, or apparent occlusion. Our results identify a new contextual influence on edge assignment. They also identify a new mechanistic relation between grouping and figure-ground processes, whereby grouping between remote elements can constrain the propagation of edge assignment between those elements. Supplemental materials for this article may be downloaded from http://app.psychonomic-journals.org/content/supplemental.

The reference frame of figure-ground assignment

Figure-ground assignment involves determining which visual regions are foreground figures and which are backgrounds. Although figure-ground processes provide important inputs to high-level vision, little is known about the reference frame in which the figure's features and parts are defined. Computational approaches have suggested a retinally based, viewer-centered reference frame for figure-ground assignment, but figural assignment could also be computed on the basis of environmental regularities in an environmental reference frame. The present research used a newly discovered cue, lower region, to examine the reference frame of figure-ground assignment. Possible reference frames were misaligned by changing the orientation of viewers by having them tilt their heads (Experiments 1 and 2) or turn them upside down (Experiment 3). The results of these experiments indicated that figure-ground perception followed the orientation of the viewer, suggesting a viewer-centered reference frame for figure-ground assignment.

Exogenous Spatial Attention Influences Figure-Ground Assignment

In a hierarchical stage account of vision, figure-ground assignment is thought to be completed before the operation of focal spatial attention. Results of previous studies have supported this account by showing that unpredictive, exogenous spatial precues do not influence figure-ground assignment, although voluntary attention can influence figure-ground assignment. However, in these studies, attention was not summoned directly to a region in a figure-ground display. In three experiments, we addressed the relationship between figure-ground assignment and visuospatial attention. In Experiment 1, we replicated the finding that exogenous precues do not influence figure-ground assignment when they direct attention outside of a figure-ground stimulus. In Experiment 2, we demonstrated that exogenous attention can influence figure-ground assignment if it is directed to one of the regions in a figure-ground stimulus. In Experiment 3, we demonstrated that exogenous attention can influence figure-ground assignment in displays that contain a Gestalt figure-ground cue; this result suggests that figure-ground processes are not entirely completed prior to the operation of focal spatial attention. Exogenous spatial attention acts as a cue for figure-ground assignment and can affect the outcome of figure-ground processes.

The time course of stereopsis: The delayed VEP component correlate of figure–ground processes

AbstractThe authors sought to analyze the time course of human stereopsis process, and focused on the measurement of VEP. The evoked response in stereopsis has been considered as a single component appearing with a latency of approximately 200 ms, but it has recently been revealed that the response is composed of two components with different properties. This paper discusses the relation between the second latency range in the two‐peaked response and the figure‐ground process. The characteristics of the figure‐ground stimulus are defined and examined by the size ratio of the single rectangular region (figure) of the random‐dot stereogram with binocular disparity to the background (ground). As a result, two successively generated negative components are observed. Of those, the initial component of the latency of approximately 150 ms increases the response with the absolute amount of stimulus, and is independent of the size ratio to the background. The second component with a latency of approximately 250 ms exhibits a stronger response when the figure region is narrowed. The response decreased with increase of the stimulus size, and disappeared when the whole stimulus was uniform. The above tendency of the response is verified, independently of the kind of local cue that defines the figure region. These results indicate that the second component has a strong correlation to the figure‐ground process, and also the latency range related to the binocular fusion is located before the occurrence of the second component. A discussion of the neural mechanism underlying the two‐peaked response is also presented, from the viewpoints of the relative delay of the second component and the local cue‐independent continuous response modulation. © 2003 Wiley Periodicals, Inc. Electron Comm Jpn Pt 2, 86(11): 72–86, 2003; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjb.10202

Divided attention in music

Two models have been advanced to account for the apparent ease with which attention can be divided in music: a “divided attention” model postulates that listeners effectively manage to follow two or more melodic lines played simultaneously. According to a “figure‐ground model,” the harmonic coherence of Western polyphonies allows a listener to focus on one melody while staying aware of the other melody, which acts as a background. This figure‐ground processing compensates for the inability to divide attention. The present study was designed to further investigate these two models. Participants were required to detect melodic errors in two familiar nursery tunes played simultaneously an octave apart. The divided‐attention model predicted that this task would be easily performed by participants, irrespective of the key of the nursery tune. The figure‐ground model predicted better performance when the keys of the tunes were identical or closely related. None of these predictions was fully supported by the data, leading us to propose a new “integrative model” of listening to polyphonic music.

Figure-ground organization and object recognition processes: an interactive account.

Traditional bottom-up models of visual processing assume that figure-ground organization precedes object recognition. This assumption seems logically necessary: How can object recognition occur before a region is labeled as figure? However, some behavioral studies find that familiar regions are more likely to be labeled figure than less familiar regions, a problematic finding for bottom-up models. An interactive account is proposed in which figure-ground processes receive top-down input from object representations in a hierarchical system. A graded, interactive computational model is presented that accounts for behavioral results in which familiarity effects are found. The interactive model offers an alternative conception of visual processing to bottom-up models.

Segmenting textures using cells with adaptive receptive fields.

Textural segmentation plays an important role in the figure-ground discrimination process. Evidence from neuroscience and psychophysics suggests that the segregation of texture patterns composed of oriented line segments is strongly influenced by the orientation contrast between the patterns (Nothdurft, Vision Res. 31, 1073-1078, 1991). In contrast to models available in the literature, this paper presents a neural network architecture for textural segmentation that can adaptively delimit the boundaries of uniformly textured regions. Cells with adaptive receptive fields encode uniformly textured regions by diffusively interpolating the estimates of feature orientations across the image. Orientation contrast boundaries are detected at gradients across the interpolated regions to produce the final segmentation. Consistent with neurophysiological data from simple and complex cells sensitive to static and moving textural patterns (Hammond and MacKay, Exp. Brain Res. 30, 275-296, 1977), the present model suggests that preattentive textural segregation can be performed by early visual processes localized to areas V1, V2, and perhaps V3 or V5. The computational results further support the idea that textural segmentation should not be thought of as a 'static' process, but as a system that employs cells with context-dependent response characteristics (Gilbert and Wiesel, Vision Res. 30, 1689-1701, 1990), here modeled as adaptive receptive fields.

On the nature and order of organizational processing: A reply to Peterson

Several points in the original article by Palmer and Rock (1994) are clarified in light of Peterson's (1994) commentary. First, we address several issues related to the nature and order of organizational processes in our theory. We argue that many of her objections result from misconceptions about our theory. We attempt to clarify the relevant issues, including the fact that we intended the theory to describe an ordering of processes in cascade rather than as a sequence of classical discrete stages. Second, we discuss Peterson's objections to our discussion of her findings on the relation between figure-ground processing and object recognition. We argue against her "prefigural recognition" proposal on logical grounds and provide an alternative account of her findings in terms of our theory.

Figure–ground discrimination from temporal information

The separation of figures from ground is achieved by the visual system based on differences in features such as luminance, colour, depth, orientation, texture or motion. Temporal information, namely phase differences between groups of spatially homogeneous points, can also lead to a clear discrimination of an object shape. The time difference needed to separate figure and ground is around 5 ms over a range of temporal frequencies between 1.3 Hz and 30 Hz, both for sharply focused and for blurred points. These short delays are clearly below the temporal integration time of the visual system. The results have implications for theories on temporal binding and object recognition.