Observers' numerosity judgments in binocular stereopsis were examined in four experiments, using random-dot stereograms (RDSs) that depicted a two-dimensional (2-D) stimulus side-by-side with a three-dimensional (3-D) stimulus. When the RDSs were correctly fused, a single surface and two (or three) transparent surfaces were observed for the 2-D and 3-D stimuli, respectively. Observers completed a numerosity discrimination task, where they judged which of the two stimuli had a greater number of dot elements. Results showed that (a) the 3-D stimulus was judged to contain more elements than the 2-D stimulus, even when both had the same number of elements, (b) the amount of overestimation increased as a function of the number of elements and the binocular disparity between the front and back surfaces of the 3-D stimulus, (c) the ratio of the physical number of elements in the front surface to that in the back surface of the 3-D stimulus had no effect on the magnitude of overestimation, and (d) when the number of elements for the two surfaces were judged separately, the ratio had more effect on the judged number of elements in the back surface than in the front surface. These results indicate that the extent of overestimation in the numerosity judgment of a set of elements in a stimulus depends on the number of depth layers in which the elements are embedded.

The number of elements in two stereo-surfaces parallelly overlapped in depth is overestimated compared to that in a single flat surface, even when both have the same number of elements. Using stereoscopic pairs of elements, we evaluated two hypotheses on the overestimation: one that a higher-order process, forming a background surface, increases the number of perceived elements, and the other that the number of elements potentially occluded by the elements on a front surface is taken accounted for. The data from four experiments showed that (a) when binocular disparity between (or among) stereoscopic elements was small, the overestimation occurred for the stimuli we used-a two-surface-overlapping stimulus, where the likelihood for the process to operate was manipulated by changing the averaged luminance of each surface, a volumetric stimulus, where the likelihood for the background surface to be formed would decrease, and a two-non-overlappingsurface stimulus, where the surfaces in depth were not overlapped-, and (b) when binocular disparity was large, the overestimation occurred for the two-surfaces-overlapping stimulus, when the averaged luminance of the two surfaces were the same, and for the volumetric stimulus, but diminished for the surface-overlapping stimulus, when the averaged luminance differed between the surfaces and for the surfaces-non-overlapping stimulus. These results cannot be explained either hypothesis only. We explain the results by postulating that the sensory system processing disparities of elements interferes with that estimating the number of elements, resulting in an overestimation of the elements in a stereostimulus, and the disparity range within which the interference occurs may depend on the stimulus depth structure.

According to the geometric relational expression of binocular stereopsis, for a given viewing distance the magnitude of the perceived depth of objects would be the same, as long as the disparity magnitudes were the same. However, we found that this is not necessarily the case for random-dot stereograms that depict parallel, overlapping, transparent stereoscopic surfaces (POTS). The data from five experiments indicated that (1) the magnitude of perceived depth between the two outer surfaces of a three- or a four-POTS configuration can be smaller than that for an identical pair of stereo surfaces of a two-POTS configuration for the range of disparities that we used (5.2-19.4 arcmin); (2) this phenomenon can be observed irrespective of the total dot density of a POTS configuration, at least for the range that we used (1.1-3.3 dots/deg(2)); and (3) the magnitude of perceived depth between the two outer surfaces of a POTS configuration can be reduced as the total number of stereo surfaces is increased, up to four surfaces. We explained these results in terms of a higher-order process or processes, with an output representing perceived depth magnitude, which is weakened when the number of its surfaces is increased.

Deep-learning workflows of microscopic image analysis are sufficient for handling the contextual variations because they employ biological samples and have numerous tasks. The use of well-defined annotated images is important for the workflow. Cancer stem cells (CSCs) are identified by specific cell markers. These CSCs were extensively characterized by the stem cell (SC)-like gene expression and proliferation mechanisms for the development of tumors. In contrast, the morphological characterization remains elusive. This study aims to investigate the segmentation of CSCs in phase contrast imaging using conditional generative adversarial networks (CGAN). Artificial intelligence (AI) was trained using fluorescence images of the Nanog-Green fluorescence protein, the expression of which was maintained in CSCs, and the phase contrast images. The AI model segmented the CSC region in the phase contrast image of the CSC cultures and tumor model. By selecting images for training, several values for measuring segmentation quality increased. Moreover, nucleus fluorescence overlaid-phase contrast was effective for increasing the values. We show the possibility of mapping CSC morphology to the condition of undifferentiation using deep-learning CGAN workflows.