Vision and Light-Guided Behavior in Sea Urchins and Brittle Stars

Sea urchins and brittle stars lack eyes, yet nonetheless are capable of vision, or the detection and resolution of spatial images and detail. Their vision, according to what is known today, is mediated through a light-sensing system that extends across the body and is processed via a decentralized nervous system. This is different from two-eyed and even most multi-eyed animals, where light is collected via discrete organs (eyes or eye cups) and processed in a brain or central ganglion. As benthic marine invertebrates, vision may be useful to sea urchins and brittle stars for navigating, finding shelter, or identifying predators. Although photoreceptor cells have been identified in brittle stars, much remains unknown about vision and light responses in both groups and the echinoderms as a whole (sea urchins, brittle stars, sea stars, sea cucumbers, and feather stars). My dissertation examines some of the gaps in this field of inquiry. I investigate (1) the potential ecological correlates of a sea urchin trait thought to mediate spatial vision, (2) how various regions of the urchin body differ in their sensitivity to light, and (3) if brittle stars are capable learning to associate a darkness cue with the presentation of food.

First, I performed a comparative study on the density of spines on sea urchins. As stated previously, sea urchins do not have eyes yet they are capable of resolving coarse images. One suggestion as to the mechanism of this capability is that the spines shade off-axis light from reaching the photosensitive test (skeleton). Following this hypothesis, the density of spines across the body determines the resolution (or sharpness) of vision by restricting the incidence of light on the photosensitive skin of the animal, creating receptive areas of different minimum resolvable angles. Previous studies have shown that predicted resolutions in several species closely match behaviorally-determined resolutions, ranging from 10° to 33°. Here we present a comparative morphological survey of spine density with species representatives from 22 of the 24 families of regular sea urchins (Class Echinoidea) in order to better understand the relative influences of phylogenetic history and three visually-relevant environmental variables on this trait. We estimated predicted resolutions by calculating spine densities from photographs of spineless sea urchin tests (skeletons). Analyses showed a strong phylogenetic signal in spine density differences between species. Phylogenetically-corrected Generalized Least Squares (PGLS) models incorporating all habitat parameters were the most supported, and no particular parameter was significantly correlated with spine density. Spine density is subject to multiple, overlapping selective pressures and therefore it is possible that either: 1) spine density does not mediate spatial vision in echinoids, or 2) visual resolution via spine density is a downstream consequence of sea urchin morphology rather than a driving force of adaptation in these animals.

Second, I examined the sensitivity to light on different parts of the body of the urchin species Lytechinus variegatus. The sensitivity of an eye is important to understand because it not only determines the light levels under which an eye can function but also indirectly affects how sharp the vision can be. It is unknown how sensitivity maps across the body in urchins, which may have implications for how various parts of the body are used in visual tasks. I tested the behavioral sensitivity response of L. variegatus to light on different regions of the body, using positive or negative phototaxis as response criteria. I tested the ambulacral region first, because this has been shown to be more sensitive to light than the interambulacral region in other urchin species. Individuals of L. variegatus were negatively phototactic to the brightest light (10,000 lux) and exhibited positive phototaxis to any dimmer light, responding to as little as 10 lux (or about the amount of ambient light in late civil twilight). Next, I tested the relative sensitivity response of the ambulacrum and interambulacrum, the two regions of the body, and confirmed that the ambulacrum is the more sensitive of the two in L. variegatus. Finally, I tested the relative sensitivity response of different angular heights (elevations) on the urchin body, along the oral-aboral axis, as these may be ecologically meaningful to the animal. There was a behavioral shift as elevation increased. Bright (10,000 lux) light at 0° (the equator of the animal) caused positive phototaxis; at 30° above the equator, roughly an equal number of urchins moved towards and away from the light; and at 60° above the equator the light caused a negative phototaxis response. The negative phototaxis observed with the light at a 60° elevation on the animal may have ecological consequences or indicate that this region is less sensitive to light. The data from this study can inform which regions and structures future studies may want to target for sensitivity and vision studies in L. variegatus.

Third, I tested whether individuals of the brittle star species Ophiocoma echinata were able to associate a period of darkness with the presentation of a food reward. Like other members of Phylum Echinodermata, the ophiuroid nervous system is decentralized, consisting of five radially arranged ganglia joined by a central nerve ring. While operant and classical conditioning have been observed in asteroids in a limited number of studies, members of the other echinoderm classes remain relatively untested. A group of individually housed Ophiocoma in an experimental group were trained by only presenting food during a period of darkness, while control group animals were fed under regular daytime room lights many hours after a period of darkness of the same duration. After the training period, the experimental group demonstrated they had learned to associate the two cues by regularly emerging during the dark period even when no food was presented. The untrained control animals, as well as pre-training experimental animals, did not emerge during the dark periods, as no food was presented. There was, however, significant variation within the experimental group in terms of the number of times individuals displayed the learned behavior and how quickly animals learned the association. This study shows that classical conditioning is possible in a class of animals without centralized nervous systems.

These results contribute to greater understandings of resolution, sensitivity, and light-guided tasks in the echinoderms which have implications for the visual ecology of these species as well as the study of sensing and processing in decentralized systems.