Many crustaceans are sensitive to the polarization of light and use this information for object-based visually guided behaviors. For these tasks, it is unknown whether polarization and intensity information are integrated into a single-contrast channel, whereby polarization directly contributes to perceived intensity, or whether they are processed separately and in parallel. Using a novel type of visual display that allowed polarization and intensity properties of visual stimuli to be adjusted independently and simultaneously, we conducted behavioral experiments with fiddler crabs to test which of these two models of visual processing occurs. We found that, for a loom detection task, fiddler crabs process polarization and intensity information independently and in parallel. The crab’s response depended on whichever contrast was the most salient. By contributing independent measures of visual contrast, polarization and intensity provide a greater range of detectable contrast information for the receiver, increasing the chance of detecting a potential threat.

Crowding occurs when highly visible objects are rendered unrecognizable in the presence of nearby features. Several studies have demonstrated that crowding can be reduced by depth differences between target and flanking features, which challenges the assumption of real-world significance because real objects typically occur at differing depths. However, most previous studies tested only small differences in depth and/or used standard stereo displays in which disparity, accommodation, and defocus blur are inconsistent with the real world. For this study we developed a multi-depth plane display consisting of three screens at different distances from the observer to test how crowding is affected by large, real differences in target-flanker depth. Overall, perceptual error decreased with increasing target-flanker spacing, but increased when the target or flankers were displayed at a different depth from fixation and one another. Our findings demonstrate that depth can increase crowding when depth differences are large. Analysis of the perceived appearance of the stimuli suggest that outside Panum's Fusional Area, crowding is increased by suppression to maintain single vision rather than by an increase in clutter because of diplopia. These findings suggest that crowding may be a more significant problem in real scenes than is estimated with 2D displays.

No abstract

Polarization vision is used by a wide range of animals for navigating, orienting, and detecting objects or areas of interest. Shallow marine and semi-terrestrial crustaceans are particularly well known for their abilities to detect predator-like or conspecific-like objects based on their polarization properties. On land, some terrestrial invertebrates use polarization vision for detecting suitable habitats, oviposition sites or conspecifics, but examples of threat detection in the polarization domain are less well known. To test whether this also applies to crustaceans that have evolved to occupy terrestrial habitats, we determined the sensitivity of two species of land and one species of marine hermit crab to predator-like visual stimuli varying in the degree of polarization. All three species showed an ability to detect these cues based on polarization contrasts alone. One terrestrial species, Coenobita rugosus, showed an increased sensitivity to objects with a higher degree of polarization than the background. This is the inverse of most animals studied to date, suggesting that the ecological drivers for polarization vision may be different in the terrestrial environment.