Close

Social foraging in vampire bats is predicted by long-term cooperative relationships

Simon P. Ripperger, Gerald G. Carter

Preprint posted on April 23, 2021 https://www.biorxiv.org/content/10.1101/2021.04.23.441116v1

Blood sisters: vampire bats leave the roost alone but meet up with preferred roostmates to forage.

Selected by Lucy Nevard

Context

Access to food is a crucial benefit of cooperative relationships in group-living animals, ranging from birds to chimpanzees. Over the last few decades, the common vampire bat (Desmodus rotundus) has become a classic example of cooperation and reciprocal food-sharing – bats who fail to feed on a given night are provided for by other bats via regurgitation of blood meals. This food-sharing is based on social bonds formed through grooming in the roost and is not always driven by relatedness. Although we have detailed evidence of interactions and bond formation between roostmates, less is known about the impact of these relationships on activity outside the roost, i.e. bats’ nightly foraging behaviour. The authors of this work aimed to distinguish between six competing hypotheses regarding vampire bat foraging behaviour (Figure 1). They include the possibilities that socially bonded roostmates either prefer to or avoid feeding together, or that prior social association has no effect on foraging encounters.

A historical lack of data on foraging encounters is partly for practical reasons: it is technically challenging to track individual free-ranging bats. The authors chose to use custom-made proximity sensors, allowing them to continuously track 50 bats over several nights and observe both inside roosting associations and outside foraging encounters on herded cattle. Kinship estimates were calculated for all bats studied and used as predictors in the analysis of foraging encounters. Of the 50 bats tagged, 23 had previously been in captivity, and data on their food-sharing and social grooming interactions in captivity were also used as predictor variables for foraging behaviour.

 

Figure 1 (from preprint). Hypothetical scenarios for the association between within-roost relationships and foraging behaviour. Preferred roostmates are shown as pairs of light brown and dark brown bats.

 

Key findings

The authors of the study tracked 50 individual vampire bats over several days and nights, aiming to reveal the relationship between social networks in the roost and foraging behaviour at night. In other words: which roostmates forage together, if any?

  • The authors found that bats tend to leave the roost independently, but preferred roostmates often reunite while foraging, supporting Hypothesis C (Figure 1). Departures from the roost were based on tagged bats losing connection with the roost proximity sensor, and almost all departures were more than 5 seconds apart. The researchers similarly never saw large groups of bats leaving the nest together.
    However, bats were observed foraging on cattle together, and these dyadic interactions were more likely amongst close kin and bonded roostmates (both in the wild or previously in captivity). The authors suggest that joint foraging might be more ‘cost-effective’ than food-sharing at the roost, illustrating another way in which long-term ties between bats are important for individual (and roost) fitness.
  • Social traits of individual bats are stable: bats who made many dyadic associations in the roost during the day (bats with high social network centrality) were also likely to maintain this centrality during the foraging period. The basis for this consistent variation between individuals is not explored here but could have implications for overall roost social structure.
  • Time in captivity appears to affect bat foraging behaviour in the long term. Bats who had previously spent time in captivity departed the roost on average 1.6 hours earlier than wild bats, without an increase in time spent foraging. This effect was picked up thanks to the study design and is an intriguing aspect of bat behaviour to consider in future studies.
  • The authors observed a previously unknown call type in vampire bats – an “n-shaped call”, distinct from the known “sweeping” and “buzz” calls. N-shaped calls were made by bats interacting near cattle, suggesting that they may be involved specifically in foraging encounters between preferred roostmates.

 

Why I liked this preprint

I was initially intrigued by this paper because of my interests in animal foraging and the evolution of social behaviour. Tracking individual bats, and indeed many free-ranging animals, has been historically challenging, and I thought the approach here was well-suited to answering the research questions. I really liked the clear explanation and figure demonstrating alternative hypotheses for foraging associations between roostmates. I was especially intrigued by the discovery of a new type of call and I am curious about the role it plays in roostmate communication and foraging. Overall, the paper provides interesting findings on the social structure and foraging behaviour of vampire bats, with implications for the evolution of cooperation in these animals.

Questions for the authors

  1. Do you think bats differentiate between non-preferred roostmates (i.e. the “grey” bats in Figure 1) and conspecifics from another roost? If so, are they differentiated by different calls? Is there an advantage to favouring non-preferred roostmates over non-roostmates?
  2. Do you know why some individuals are more highly connected than others? Are highly connected individuals different from other bats in physical traits, for example size?
  3. You state that dominance hierarchies among female vampire bats are not necessarily clear and linear. How is social dominance determined and manifested amongst vampire bats and did you investigate dominance in the bats you studied? Do you think that dominance is also a stable trait within individuals?

Further reading

Wilkinson GS, 1984. Reciprocal food sharing in the vampire bat. Nature 308:181.

Carter GG, Farine DR, Crisp RJ, Vrtilek JK, Ripperger SP, Page RA, 2020. Development of new food-sharing relationships in vampire bats. Curr Biol 30:1275-1279. e1273.

Tags: bats, cooperation, foraging, sociality

Posted on: 28th May 2021

doi: https://doi.org/10.1242/prelights.29227

Read preprint (No Ratings Yet)




Author's response

Gerald Carter, Simon Ripperger shared

Q1.

Do you think bats differentiate between non-preferred roostmates (i.e. the “grey” bats in Figure 1) and conspecifics from another roost?  If so, are they differentiated by different calls? Is there an advantage to favouring non-preferred roostmates over non-roostmates?

 

Gerald Carter:  Great questions. We don’t know yet. I would guess that the bats could distinguish between slightly familiar and completely unfamiliar individuals, both by sound (calls) and scent. What we might label as “non-preferred roostmates” might actually be weak social preferences that are still greater than zero. For example, we humans have close friends, friends, associates, people we recognize but don’t know their names, and then… complete strangers. I am still more interested in interacting with someone I’ve seen repeatedly compared to a complete stranger. I would love to know if vampire bats have this same continuous spectrum of social familiarity. That’s a big question.

Simon Ripperger: In addition, it might be interesting to find out whether there is a cognitive limitation to distinguishing between conspecifics and whether this is somehow a challenging factor in respect to global change. How many unrelated individuals can a single bat distinguish and remember? I would assume that the average colony size in a mature rainforest is quite small (maybe 10 or 20? I actually don’t know but I have seen quite small colonies in areas without cattle). In the area where we work, a single tree or roost may host hundreds or more individuals because the food availability is basically unlimited in the presence of large herds of cattle. One may ask ‘where is the limit’? I would certainly believe that a bat that lives in a roost with another 10 bats can easily distinguish those from unfamiliar bats. However, in a roost with 300 conspecifics – I doubt it…it is actually quite likely that the colony we’ve been working on is not a single social group but several social units. This is actually one aspect of our ongoing research: What are the social dynamics within a single roost?

 

Q2.

Do you know why some individuals are more highly connected than others? Are highly connected individuals different from other bats in physical traits, for example size?  

 

GC: We have not found any physical traits (besides sex) that predict gregariousness or cooperativeness, however, we are currently looking at oxytocin receptor densities in the brain as one possible predictor.

SR: Another aspect of these automatically sensed social networks is that we collect data on associations, not on interactions. Apart from individual differences in hormone levels as mentioned by Gerry, we could also think of differences in activity levels or the degree of exploratory behavior. There might be individuals that move around more frequently within the roost or some individuals might be more curious and therefore more likely to approach others. That would not necessarily mean that those are more social or gregarious but in an association network, they would still appear more connected. It’s therefore always important to keep in mind which methods were used when interpreting animal behavior.

 

Q3.

You state that dominance hierarchies among female vampire bats are not necessarily clear and linear. How is social dominance determined and manifested amongst vampire bats and did you investigate dominance in the bats you studied? Do you think that dominance is also a stable trait within individuals?

 

GC: We measured female social dominance in another study (which should be out soon). We recorded the outcomes of >1000 competitive interactions over food (like one bat displacing another from a spout of blood) in the same captive colony. We found a weakly linear dominance hierarchy but the patterns of female dominance were relatively weak, and female social rank was not clearly predicted by body size, age, or reproductive status. The dominance relationships we observed were not very consistent in their directionality or linearity compared to other groups of female mammals. For example, A might displace B, then later B displaces A; or A might displace B, which displaces C, which displaces A (circularity), and both of these should not happen in a linear dominance hierarchy. This lack of structure could be real, or it might be that we need to measure social rank in a different way. There’s always more work to be done.

Have your say

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up to customise the site to your preferences and to receive alerts

Register here
Close