New paper came out on responses of a tropical free-tailed bat to distress calls. The title is Distress Calls of a Fast-Flying Bat (Molossus molossus) Provoke Inspection Flights but Not Cooperative Mobbing. Here’s the story behind the paper.
A year ago, I took a great field course on animal communication sponsored by a collaboration between German and US labs. The goal of this course was to create teams with both German and US graduate students. Each team focused on studying acoustic communication in a different taxa (bird, frogs, bats, and bats). There were two bat teams because all of the instructors and about half of the students studied bats. Each team was expected to do conduct a small research project that would lead to a peer-reviewed publication. I thought that was pretty ambitious.
I was interested in studying responses of bats to distress calls. At the international bat conference the previous year, I met a M.Sc. student who did a fascinating playback study where he played distress calls of bats (a Rhinolophus) outside of caves, where the calls were recorded at the same or different caves. After talking a bit, he asked me for help with the statistical analysis. So I re-did all the statistics and sent it back. In the original drafts, he claimed that the bats were mobbing and attacking the speaker, but it seemed possible (and more likely to me) that the bats were simply making more passes to inspect the source of the distress calls, especially given the effect sizes. He also did not take video footage, so there was no direct evidence of what the bats were actually doing. I thought the conclusion of mobbing was therefore unwarranted. I suggested he tone down his interpretations and send the paper to PLOS One. Instead he stuck to the more exciting story and sent the paper to Biology Letters and Behavioral Ecology and Sociobiology… with my name on it, even though I did not approve or even see the manuscript. Thankfully, it was rejected from both journals. I have not heard back about the manuscript since then. I don’t blame the author that much. I think it was largely 1) a communication error, because he could not speak English very well, and 2) and example of how distorted the academic incentive structures have become.
But I still wanted to know about cooperative mobbing in bats. Is it real? If so, how common is it? This is something that is relatively common in birds. Where I used to live I would see crows and other birds mobbing raptors all the time:
There have been a number of papers suggesting that bats, like birds, will also mob if they hear distress calls. I’ve heard many people talk about this behavior, so it is something that at least some people believe bats do. But I have never seen it myself. There have been two studies that have actually observed bats mobbing predators, so we know some bats do mob. But we don’t know common it is, and I was a skeptical of reports in the literature based on playbacks of distress calls, including the recent one above that was not published.
In a playback experiment, it’s hard to know what is mobbing and what might be “predator inspection behavior”. Imagine you hear the an animal growling and a person screaming. Would you just run away immediately and blindly? Maybe. Or maybe you would try to see what exactly was going on. This would involve actually going towards the sounds. And you might even do it in pairs or small groups (like many prey animals do). That’s predator inspection. And that’s what I thought the bats might be doing when they fly over a speaker playing distress calls.
Getting back to the field course, Profs. Annette Denzinger and Hans Ulrich Schnitzler had been recording free-tailed bats (Molossus molossus) on Barro Colorado Island. They live under the eaves of the field station buildings. They fly out in fast direct flights, forage for 30-45 minutes and then come back to the roost. There is also evidence for group foraging in Molossus molossus. They also make many social calls. Dina Dechmann and others are doing great long-term studies on this bat’s social behavior, which I am excited to see published.
Annette served as the leader and advisor of our group, and she helped us develop the project throughout. We also got recordings of distress calls, advice, and feedback from Mirjam Knoernschild who has been studying the sac-winged bats there for years. The two German students in the group Diana and Marie-Kristin also recorded many echolocation calls and social calls of these bats in flight and in the roosts.
My interest was whether I could replicate the past studies that attracted bats with distress calls and see with my own eyes if these bats would do anything like mobbing. The experiment was very simple. I played distress calls from the bats as they exited the roosts. I played silence and noise as controls. I recorded responses using video and audio. What I observed and recorded was that the free-tailed bat and the sac-winged bat that also lived on the same buildings were both attracted to the distress calls, but they did not do anything that looked like mobbing. Usually one bat would fly nearby or circle nearby. Nothing like the mobbing Mirjam had observed in Phyllostomus hastatus.
The paper would have made a bit more sense if I could refer to the other study and data from Rhinolophus, but since that one was never published, I could not refer to it. This is an example of the importance of publishing negative results. If people do playbacks and find evidence for mobbing, that’s a neat paper. But if they do not, that’s seen as less publishable. So the literature on this topic is skewed.
One simple point from this study was that an increase in activity is not good evidence for cooperative mobbing. If not familiar with bats, many people in a room with a bat, or even in a place where bats are foraging, will often report that bats are attacking or “dive bombing” them. Likewise, the fact that bats approach a speaker playing distress calls does not indicate that they are cooperatively attacking or harrassing the speaker.
It would be interesting to studies like this with more bat species and with a predator model. Perhaps if the bats detected a predator instead of a metal box, they would have started mobbing. I just bought a life-like owl model for this purpose. It would make a neat undergraduate project.
Carter Lab 