The article by Stephen Grossberg and Ennio Mingolla explores the neural dynamics of perceptual grouping, focusing on how the visual system segments and groups textures, boundaries, and emergent segmentations. The authors use a real-time visual processing theory to explain various perceptual phenomena, including the grouping of textured images, randomly defined images, and periodic scenic elements. They highlight the role of both "local" feature processing and "emergent" features in segmenting scenes, and how segmentations can arise across image regions without luminance differences. The theory also addresses how segmentations can override local image properties in favor of global statistical factors and why powerful object recognition segmentations may be barely visible or invisible. The authors introduce the Boundary Contour (BC) System, Feature Contour (FC) System, and Object Recognition (OR) System to explain these phenomena. The BC System, defined by a hierarchy of orientationally tuned interactions, includes the OC filter and CC loop. The OC filter generates inputs to the CC loop, which contains spatially short-range competitive and long-range cooperative interactions. Feedback between these stages synthesizes context-sensitive segmentations from among many possible groupings of local featural elements. The article discusses the context sensitivity of perceptual units, the role of illusory contours, and the discounting of illuminant effects. It also examines the trade-off between boundary and feature contours, the importance of boundary completion via cooperative-competitive feedback signaling, and the relationship between form perception and object recognition. The authors provide a dynamical explanation for Beck's theory of textural segmentation, showing how emergent features contribute to perceptual grouping even if they are not visible. Overall, the article aims to provide a universal set of rules for preattentive perceptual grouping processes that feed into depth-form perception and object recognition, and to offer a computational framework for understanding how textural elements are grouped into easily separated figures and ground.The article by Stephen Grossberg and Ennio Mingolla explores the neural dynamics of perceptual grouping, focusing on how the visual system segments and groups textures, boundaries, and emergent segmentations. The authors use a real-time visual processing theory to explain various perceptual phenomena, including the grouping of textured images, randomly defined images, and periodic scenic elements. They highlight the role of both "local" feature processing and "emergent" features in segmenting scenes, and how segmentations can arise across image regions without luminance differences. The theory also addresses how segmentations can override local image properties in favor of global statistical factors and why powerful object recognition segmentations may be barely visible or invisible. The authors introduce the Boundary Contour (BC) System, Feature Contour (FC) System, and Object Recognition (OR) System to explain these phenomena. The BC System, defined by a hierarchy of orientationally tuned interactions, includes the OC filter and CC loop. The OC filter generates inputs to the CC loop, which contains spatially short-range competitive and long-range cooperative interactions. Feedback between these stages synthesizes context-sensitive segmentations from among many possible groupings of local featural elements. The article discusses the context sensitivity of perceptual units, the role of illusory contours, and the discounting of illuminant effects. It also examines the trade-off between boundary and feature contours, the importance of boundary completion via cooperative-competitive feedback signaling, and the relationship between form perception and object recognition. The authors provide a dynamical explanation for Beck's theory of textural segmentation, showing how emergent features contribute to perceptual grouping even if they are not visible. Overall, the article aims to provide a universal set of rules for preattentive perceptual grouping processes that feed into depth-form perception and object recognition, and to offer a computational framework for understanding how textural elements are grouped into easily separated figures and ground.