Features

Animal vision: Eight-eyed marvels

Ilse Daly finds much appealing about the jumping spider and its visual system

While the name ‘jumping spider’ may conjure up terrifying visions of lethal attack by gargantuan, hairy brown arachnids of the kind beloved by horror movies, the reality is something quite different. Jumping spiders are, for want of a better word, cute. They are fluffy, brightly coloured and, at less than half the size of a human fingernail, they are not likely to pose a threat to anything bigger than a fruit fly. Oh, and they dance, but more on that later.

Fancy footwork aside, jumping spiders have the most acute vision of any land invertebrate. In fact, their acuity is similar to ours. So despite being only a fraction of the size of our entire eyeball, jumping spiders see the world in a similar level of detail to us. Actually, despite having a completely different ocular setup to ours (did I mention the eight eyes?) fundamentally their vision works in a similar way to our own. So how do they do it?

As is clear from the image above ( 1), jumping spiders have one pair of eyes that are much larger than the others. These are called the principal eyes, and they are what give jumping spiders such unexpectedly good vision. The other three pairs of eyes are referred to as secondary eyes and, while they may not be as spectacular as the principal eyes, they nonetheless play an important role in the spider’s vision.

Like us, spiders have camera-type eyes, which consist of a corneal lens that focuses light on a photo-sensitive retina. This is actually quite rare among invertebrate animals, most of whom have compound eyes. But that is where the similarity ends. Unlike us, they have no iris and consequently no pupil. Nor can they move their eyes like we can because the cornea is fixed in their carapace.

Light entering the principal eyes is focused by the corneal lens. Rather than being spherical, jumping spider eyes are effectively tubes, twice as long as they are wide. Their retina is also unusual. For a start, it is not flat. Instead, it forms a pit at the back of the eye, aligned with the ocular axis. Because the refractive index of the retina is higher than in the vitreous, this curved surface acts as a diverging lens, increasing the focal length of the system and magnifying the image by about one-and-a-half times. In fact, this is the same ‘telephoto’ optical arrangement that can be found in the eyes of birds of prey. Who knew that a jumping spider would have anything in common with a peregrine falcon?

Despite the telephoto-optics of the eye creating a magnified view of the world, the quality of the image reaching the back of the spider’s eye is degraded due to chromatic aberration; shorter wavelength light entering the principal eye is focused closer to the lens than longer wavelength light. However, the jumping spider has a solution for this. Not only is the retina not flat, but it is formed of four distinct tiers, each with a different photoreceptor type.

The tier closest to the front of the eye is sensitive to ultraviolet light, the tier behind to blue and then the two tiers at the back of the eye to green light. So light of each of those colours, despite being focused at different distances from the corneal lens, is still essentially in focus, and the magnified image remains sharp.

There is another problem these spiders have to overcome. Because their cornea is fixed, they cannot change the shape of their lens. Such accommodation is critical for being able to focus on objects at different distances. With a fixed lens, objects close to the eye are focused at a different depth than objects that are far away. Fortunately, the tiered retina solves this problem too. Light from distant and close-to objects are focused onto different layers of the tiered retina, a feature that the jumping spiders can make use of to accurately judge distance, which is critical to their lifestyle as you will see later.

The final unusual feature of the retina is its overall shape. It is long and ‘boomerang’ shaped (figure 2), probably due to the space constraints of the tube-like structure of the eye, with a narrow field of view. While the corneal lens has a viewing angle of approximately 90°, the retina’s horizontal field of view is only 2° to 5°. This clearly constrains how useful the eyes are. If something is out of its very narrow field of view, how can it see it? This is where the six other eyes come in.



Figure 2: The retinas of the principal eyes of jumping spiders are boomerang-shaped. They are capable of horizontal movement and torsional rotation clockwise and anti-clockwise


Structurally, the secondary eyes are much simpler than the principal eyes. Partly this is to conserve energy, but also because jumping spiders can only process a limited amount of information with their small brains. However, jumping spiders do not need their vision from the secondary eyes to be highly detailed. Instead, they use these eyes like we use the periphery of our vision; to detect object motion. So the secondary eyes, which due to their positioning around the spider’s head have an almost panoramic view, are on the lookout for moving objects. But, given that the principal eyes cannot move, what happens when they do spot something that may be of interest to the spider?

Well, it turns out that while jumping spiders cannot move their corneas, they can move their retinas. Just like someone looking through a telescope fixed at the far end, jumping spiders can move their retinas horizontally to give themselves a much bigger effective field of view. In effect the secondary eyes act like our peripheral vision and the principal eyes as our fovea. And jumping spiders really do foveate; they fixate both of their primary eyes on a target, rather than just pointing in its general direction. But what exactly are these spiders looking at and why do they need such good vision to see it?

Contrary to the stereotype, jumping spiders do not weave webs. Instead, they are active hunters that stalk their prey and pounce, like teeny-tiny cats. They do still use silk, but as a means of a safety line as they navigate through rough terrain, in much the same way that a climber might use a belay. Additionally, they have extremely elaborate courtship rituals that involve rhythmic body movement that (if we allow ourselves to anthropomorphize) looks a lot like dancing. I would highly recommend a web search (pun intended) on jumping spider courtship dances. I guarantee it will change your perspective on spiders as being terrifying and threatening.

So to facilitate its complex lifestyle, a jumping spider needs vision for three tasks: detection, identification and navigation. Detection is the easiest task out of the three. It just involves noticing that there is an object against the background. This is the role that the secondary peripheral eyes play. Once detected, the primary eyes fixate on the object just as a human would fixate their foveas on an object of interest. So despite having very different eyes, the way we move them is, in effect, rather similar.

Once detected, the object needs to be identified. Is it prey, predator or mate? Unfortunately for the spider, these three things can appear rather similar to one another. One of the characteristic behaviours of a jumping spider is their move-and-wait pattern. They quickly scuttle to a position and then wait for one to 10 seconds before deciding to take one of the following actions: creeping up to the target and capturing its prey, initiating a courtship ritual, walking away or running away.

So what is the spider doing during this wait (figure 4)? Well, it appears as though they are scanning the object and in particular looking for legs. Not only can jumping spiders move their retinas from side to side, but they can also rotate them. The narrow horizontal field of view of the retina, coupled with this ‘torsional’ rotation makes for a very efficient line (well, leg) detector. In an experiment, when jumping spiders were presented with small dark targets that had what would pass for legs (figure 3) the spiders were much less likely to jump on them than if the same target did not have legs. Instead, in the presence of ‘legs’ the spiders were much more likely to start courtship rituals (if they were male). So why would legs be so important?


Figure 3: When each of the shapes on the top row were presented to a jumping spider, they elicited courtship behaviour, while the shapes on the bottom row elicited a prey-capture response (ie the spider jumped on them). The common feature that seems to govern which behaviour is elicited is whether or not the shape had clearly discernible legs similar to those of a spider


Figure 4: A jumping spider plotting her next move


The courtship rituals of jumping spiders involve a lot of movement from the males and in particular a lot of leg-lifts (figure 5). These appear to be at specific angles that, we can only assume, are particularly attractive to a female jumping spider. So the rotation of the retinas, together with their saccadic ability, gives the females a tool that she can use to judge the fitness of her mate. Natural selection at work.


Figure 5: A male peacock jumping spider engaged in his courtship ritual, which looks a lot like dancing

Finally, there is navigation. It turns out that some species of jumping spider are every bit as intelligent as we feared. It seems as though they can actually plot a course through difficult terrain. As in, they look at the landscape and then plan beforehand where they are going to go, which is utterly amazing in an animal only a few millimetres across with a miniscule brain. Their detailed, telephoto vision helps them to plot a safe course through rough ground.

Jumping spiders are incredible. Their vision is comparable to ours, both in terms of eye movement and the level of detail they can see. This is all the more extraordinary given that they are minuscule invertebrates who took a very different evolutionary course. However, as nature (and these spiders) have shown, there is more than one way to solve an optics problem.

  • Dr Ilse Daly is a research associate at the School of Biological Sciences at the University of Bristol.