The Verriest Lecture: Visual properties of metameric blacks beyond cone vision.

The generic framework of metamerism implies that the number of sensors is smaller than the dimension of the stimulus. The metameric black paradigm was introduced by Wyszecki [Farbe2, 39 (1953)] and developed by Cohen and Kappauf [Am. J. Psychol.95, 537 (1982)]. Within a multireceptor and multiprimary scheme, we investigate how far the choice of illumination can isolate a photoreceptor response. The spectral profiles of the fundamental metamers that correspond to a collection of (x,y) values over the chromaticity diagram are shown. When the luminance is set at a fixed value, the relative excitation of the melanopsin cells and of the rods elicited by the fundamental metamers varies over the chromaticity diagram. The range of excitation of the melanopsin cells and of the rods that could be achieved at a given chromaticity, by manipulating the metameric black content, is examined. When only the melanopsin excitation is manipulated, the range of melanopsin excitation that can be achieved is rather limited. On the chromaticity diagram, the largest range of variation of the rods and the melanopsin cells excitation is obtained for (x,y) chromaticity coordinates near (1/3,1/3). Extension of Cohen's procedure to rod and cone metamers is proposed. The higher the number of spectral bands, the wider the choice of metameric lights.

Domain of metamers exciting intrinsically photosensitive retinal ganglion cells (ipRGCs) and rods.

Any stimulus can be described as composed of two components-a fundamental color stimulus that controls the three cone responses and a metameric black that has no effect on cones but can drive photoreceptors other than cones [e.g., rods and melanopsin expressing retinal ganglion cells (ipRGCs)]. The Cohen and Kappauf [Am. J. Psychol. 95, 537 (1982)] method is extended to calculate the black metamer basis for a limited set of band spectra. Using seven colored LEDs, the method is exploited to produce real metamer illuminations that stimulate in parallel melanopsin expressing ipRGCs and rods, at most or at least. We have verified that the pupil diameter increases when the ipRGC and rod excitation is at a minimum. For 14 observers, the average relative increase is 12%.

Structural and view-specific representations for the categorization of three-dimensional objects.

It has been debated whether object recognition depends on structural or view-specific representations. This issue is revisited here using a paradigm of priming, supervised category learning, and generalization to novel viewpoints. Results show that structural representations can be learned for three-dimensional (3D) objects lacking generalized-cone components (geons). Metric relations between object parts are distinctive features under such conditions. Representations preserving 3D structure are learned provided prior knowledge of object shape and sufficient image input information is available; otherwise view-specific representations are generated. These findings indicate that structural and view-specific representations are related through shifts of representation induced by learning.

Temporal image fusion in human vision.

We studied temporal integration by presenting sequences of orthogonal high-contrast sinusoidal gratings. In rapid alternation (38 Hz) gratings appear fused whereas at lower rates they appear segregated. To test whether afterimages are the source of the fused and segregated appearance, we probed the decay characteristics of a vertical grating's afterimage by flashing an otherwise identical horizontal grating at varying times after the grating's offset. The results show a rapid decay function for a negative afterimage with a 50-ms time constant, much shorter than expected from other paradigms. From dichoptic presentations we conclude that the site of the interaction is before that of binocular integration.