Roles of contour and surface processing in microgenesis of object perception and visual consciousness

Abstract

Developments in visual neuroscience and neural-network modeling indicate the existence of separate pathways for the processing of form and surface attributes of a visual object. In line with prior theoretical proposals, it is assumed that the processing of form can be explicit or conscious only as or after the surface property such as color is filled in. In conjunction with extant psychophysical findings, these developments point to interesting distinctions between nonconscious and conscious processing of these attributes, specifically in relation to distinguishable temporal dynamics. At nonconscious levels form processing proceeds faster than surface processing, whereas in contrast, at conscious levels form processing proceeds slower than surface processing. I mplications of separate form and surface processing for current and future psychophysical and neuroscientific research, particularly that relating cortical oscillations to conjunctions of surface and form features, and for cognitive science and philosophy of mind and consciousness are discussed.

Keywords: conscious visual processing; contour; nonconscious visual processing; surface color; surface contrast; temporal dynamics.

Figure 1.

Top panel: The spatial layout of stimuli used to study a target’s surface and contour processing during metacontrast. In the contrast matching task, the target and comparison stimulus were presented slightly above and symmetrically to the right and left of fixation, and only the target was followed by a mask. The observer on any trial had to indicate which of the two stimuli, target or comparison, appeared darker. In the contour discrimination task, the target and following mask were both centred slightly above and either to the left or else to the right of fixation. The target could be a disk with either an upper contour deletion (as shown), a similar lower contour deletion, or no deletion. Here the stimulus display position and target shape was randomized over trials. Using a three-alternative forced-choice procedure, on any trial the observer had to indicate which of the targets was presented. Bottom panel: The normalized target visibility functions, relative to a baseline visibility of 1.0 obtained when the targets were presented without the following mask, shown separately for surface contrast and for contours, as a function of the stimulus onset asynchrony (SOA) between the target and the mask. Note (a) the difference of 30 ms between the optimal masking obtained for the contour discrimination and the contrast matching tasks (dotted arrows) and (b) the dissociation between contour and contrast visibilities at the SOSOA of 10 ms (dashed arrow). Adapted from “Meta- and Paracontrast Reveal Differences Between Contour and Brightness-Processing Mechanisms” by B. G. Breitmeyer, H. Kafaligönül, H. Öğmen, L . Mardon, S. Todd, and R. Ziegler, 2006, Vision Research, 46, pp. 2646, 2647.

Figure 2.

Depictions of two original stimulus objects (top panel), the implicit representation of their contours by the Boundary Contour System (BCS, middle panel), and their explicit representation via filling in of surface color/contrast by the Feature Contour System (FCS, bottom panel).

Figure 3.

Schematic of the Boundary Contour System (BCS) and the Feature Contour System (FCS) in relation to two major, parvocellular (P) and magnocellular (M), visual pathways and their major projections, beginning respectively with retinal β and α cell, projecting via the lateral geniculate P and M layers to the primary visual cortex (V1). After V1 the M-pathway comprises the major, but not exclusive, dorsal projections to the parietal cortex; and similarly the P-pathway comprises the major, but not exclusive, ventral projections to the inferotemporal cortex. The dorsal projection is considered to comprise the “where” or the “vision-for-action” pathway; the ventral projection, the “what” or the “vision-for-perception” pathway. Note that in the ventral pathway, the FCS (outlined in red dashed lines) consists of the cortical P-pathway comprising the V1-blob / V2 (the secondary visual cortex)-thin stripe projections and beyond; the BCS (outlined in blue dashed lines) consists of the cortical P-pathway comprising the V1-interblob / V2-interstripe projections and beyond. The FCS processes only the wavelength and luminance properties (designated by the symbol and the symbol) of an object’s surfaces. The BCS processes the object’s contours (designated by the symbol) defined either by isoluminant wavelength differences (designated by the symbol) or by luminance differences (designated by the symbol). The dorsal pathway consists of the V1- layer 4B / V2-thick stripe projections and beyond. See text for further details. In all parts of the figure, the symbol designates ability to process motion direction, and the symbol designates ability to process binocular disparities. Adapted from “Concurrent Processing Streams in Monkey Visual Cortex” by E. A. DeYoe and D. C. Van Essen, 1988, Trends in Neuroscience, 11, p. 223.

Figure 4.

Upper panel: A schematic of several possible target stimuli followed, at an optimal masking SOA (the stimulus onset asynchrony) of 53 ms, by a surrounding metacontrast mask. The masked targets served as primes; the visible mask, as probe. The targets could either be presented in their entirety (whole), with only their vertices (corners), or only their side orientations (sides). Form features of the target could either be congruent or incongruent with those of the mask. The task of the observers was to respond as quickly and accurately as possible to the form of the mask, which could either be a square or a rhombus. Bottom panel: Priming effects for each of the three types of targets, obtained by subtracting the choice reaction time (RT) to the mask when the target and mask had congruent form features from the RTRT obtained when they had incongruent form features. Adapted from “Unconscious and Conscious Priming by Forms and Their Parts” by B. G. Breitmeyer, H. Öğmen, J. Ramon, and J. Chen, 2005, Visual Cognition, 12, pp. 722, 727.