Robot Lizard Push-ups

Have you ever been walking through the forest and thought to yourself, "Damn, its loud here...it's really, really hard to hear anything anybody else is saying"? Well, maybe that's what prompted Terry J. Ord and Judy A. Stamps, respectively from Harvard and UC Davis to investigate lizard exercise routines.

You ask: What do lizard calisthenics and communication have in common? Patience, grasshopper.

A noisy environment makes it particularly challenging for animals to detect communication signals from far away. In this case, noise doesn't mean audible noise, it means visual noise. Think about it this way. If you're a lizard, and you want to communicate to another lizard, you'd probably do so visually, with your body. If you're the lizard on the receiving end of the message, there might be a lot of movement in your field of vision - after all, you're in the forest, where there are lots of other animals, leaves and branches moving in the wind, and so on. And if you weren't already paying attention to the sending lizard, you'd need some way to realize that he was trying to get your attention, to pick out the movements of your buddy from all the rest of the motion in your visual field.

It's sort of like when you're at a party, lots of conversations are going on in the background, the music is playing, you're flirting with the hot bartender... The party host wants to make an announcement, so what does she do? She clinks her fork against a glass in an attempt to catch everyone else's attention and get them to ignore the ambient noise. Then she begins her speech.

Well, Anolis lizards, who live in the mountains of Puerto Rico, have evolved a similar mechanism to get the attention of other lizards. They might want to say "Hey! That's MY rock you're sitting on," or, "Your scaley body makes me hot!"

Scientists have often observed animals using an "announcement" call before beginning their true message. There are often introductory notes in territorial calls in bird species; frog mating calls begin with a low frequency ribbit, and coyotes often bark before they begin their characteristic howling. However, it is the case that the message is not ALWAYS preceded by the announcement. So when do animals use the announcement, and when do they just launch into their speech? It should be noted that it can be very dangerous for animals to use the announcement - whether visual or auditory - because it could also signal predators that lunch is ready. Like a visual dinner bell. (There's a species of bat that is really really sensitive to the mating call of a species of frog. Bad news for the male frogs when they're ready to get it on.)

Ord and Stamps observed that "male lizards...advertise territory ownership to neighbors and intruders via dynamic visual displays consisting of species-typical headbobs and extensions of a colored dewlap." Sometimes, they begin their displays with exaggerated 4-legged push-ups before beginning the typical headbob sequence.

The assumption is that the announcement facilitates signal detection in the receiving animals, like the clinking of the fork against the glass facilitates signal detection at a party. But this hypothesis has rarely been confirmed. Furthermore, given that animals only sometimes use the announcement, you might assume that announcement-inclusion might be correlated with increasing environmental noise. When it is "louder" (visually louder, in this case), you might expect the announcement to be used more often.

If it is the case that these lizard push-ups serve to attract the attention of inattentive lizards, and if there are costs to doing lizard push-ups (energy expenditure, attracting the attention of predators), then it should be the case that the lizards should only use the push-ups when necessary.

First, they did a regular old-school naturalistic observation study. Twenty-six male lizards with neighboring male lizards who all had line-of-sight to the sending lizard were observed. Out of all displays, how many began with push-ups, and when were push-ups used? Fluctuations in habitat light, increased visual noise, and the average distance to neighbors all predicted the inclusion of introductory push-ups (and accounted for 60% of the variance). Want to watch a video of a lizard doing push-ups and showing off his dewlap?

Then, they built LIZARD ROBOTS. Who did push-ups. That's right. EXERCISING LIZARD ROBOTS.

Okay, they actually looked like this:

Sometimes the robot lizards did the headbob thing without the push-ups. Sometimes they used push-ups before the headbob thing. Sometimes they did the dewlap thing before the headbob thing. Importantly, this particular species of lizard never uses the dewlap thing as an announcement (though other species do). The researchers included this condition to test the hypothesis that ANY high-speed movement at the beginning of a display could function as an announcement.

As expected, high-speed introductory movements greatly reduced the time required for receiving lizards to attend to the robot lizard. In poor light, lizards turned to face the robot more quickly when the display was preceded with an announcement. It did not matter whether or not they used push-ups or dewlaps - suggesting that alerts attract attention more generally, but do not affect the main message. Under bright light, there was no difference between displays with or without announcements.

One puzzling finding emerged from the results: why did lizards attend to the robots more quickly, when using alerting signals (push-ups, dewlaps) in low light than in bright light? Even though the pre-message announcement would enhance signal detection in low light compared to messages without the alert, you would still expect that under more favorable conditions, messages with alerts would be detected even quicker. Yet this was not the case. Ord and Stamps hypothesized that perhaps lizards monitor their conspecifics more closely under conditions of poor light, perhaps because it is harder to notice behavior changes in low light.

A follow-up experiment was conducted to address this question. A non-moving robot lizard was placed in view of other lizards, as was a control object (a bunch of leaves clumped together with roughly the same mass and surface area as a lizard). Would the lizards look at the stationary lizard more often than the control object? The answer was yes. And this pattern was exaggerated under low-light conditions. Lizards monitor their conspecifics more often under poor light conditions than under bright light conditions.

So the evolution of alert signals is potentially prevalent, but remains largely uninvestigated. More research is necessary, and in other animals. If alerting signals are used in other species in a similar way (increased under increased noise or poor environmental conditions), then it would be a pretty fantastic example of functional convergence in animal communication crossing signal modalities (auditory and visual), as well as taxonomic boundaries (e.g. frogs, lizards, birds, mammals).

Ord TJ, & Stamps JA (2008). Alert signals enhance animal communication in "noisy" environments. Proceedings of the National Academy of Sciences of the United States of America, 105 (48), 18830-5. PMID: 19033197