"The authors have declared that no competing interests exist."
The effects of age on neophobia and exploration are best described in birds and primates, and broader comparisons require reports from other taxa. Here we present data showing age-dependent exploration in a long-lived social species, the common vampire bat (
In many birds and mammals, immature individuals appear to be more exploratory and curious relative to their adult conspecifics [
If an early period of enhanced exploration increases opportunities for learning, this could be particularly adaptive for long-lived animals with flexible foraging strategies [
Common vampire bats (
During a previous experiment [
To determine if young males, or young bats in general, are more likely than older bats to explore novel objects, we conducted tests of novel object exploration first in the same captive group of vampire bats described above, which had eight young males and two young females, and then in a second mixed-age captive colony which had six young female and six young male vampire bats.
This work was approved by the University of Maryland Institutional Animal Care and Use Committee (Protocol R-10–63). The first captive colony contained two young females, eight young males, 12 adult females, and 20 adult males. All individuals were born in captivity, and this colony is described in more detail elsewhere [
We exposed the captive bat colony to two different kinds of stimuli: (1) a conspecific groupmate trapped in a small cage and (2) an inanimate novel object. We first conducted 17 one-hour trials, during which we placed a bat inside a small cage on the floor of the larger flight cage and recorded infrared-illuminated video using a Sony Nightshot camcorder (see [
We counted the number of unique visitors as the response. To get a conservative minimum estimate of this response for the unmarked young males, we counted the greatest number of young males seen to visit the object together at one time in the trial (i.e. the minimum number of visitors in that trial). For example, if seven unmarked young males were observed crawling on the cage in one video frame, then the number of visiting young males was scored as seven. This approach is highly conservative because we should underestimate visits by young males, and we tested the alternative hypothesis of more visits by young males.
To test if young males were more likely than other bats to explore novel objects, and if they were more likely to visit trapped groupmates than novel inanimate objects, we first used Fisher’s exact test to compare the number of visits summed across all trials to each stimulus type (trapped bat or novel object) by category (young male or other). Second, we used permutation tests applied to generalized linear models (binomial distribution, logit link function) to test for the effect of stimulus type on the probability of visits by young males and by others. These tests compare the observed coefficients to those expected under the null hypothesis, simulated by randomizing the stimulus type across visits (5000 randomizations). To calculate 95% confidence intervals (CI) around the mean proportion of bats of each category that visited each stimulus type, we used nonparametric bootstrapping via the BCa method in the boot package in R [
The response towards trapped conspecifics versus novel objects depended on the category of bat. Most notably, all the observed visits to the novel objects were by the same eight young males. The 17 trapped bats resulted in at least 106 unique visits by the eight young males (six per trial) but only 18 unique visits from the 34 other bats (1 per trial), whereas the eight novel objects resulted in at least 59 unique visits by the young males (7 per trial) but zero visits from all other bats (Fisher’s exact test: odds ratio ~ 0, p < 0.001, permutation test interaction effect: p = 0.002,
Dots of each color show responses in each trial. The lowest observed proportion of visiting young males was higher than the highest observed proportion of visiting bats that were female or adult male. Novel objects were always visited by more than five of the eight young males, whereas the other bats never visited a novel object.
This work was approved by the Smithsonian Tropical Research Institute Animal Care and Use Committee (#2015-0915-2018-A9), and by the Panamanian Ministry of the Environment (#SE/A-76-16). We worked with a second captive colony of 24 wild-caught female common vampire bats and their 12 captive-born offspring (six females, six males), which ranged in age from 3 to 15 months at the time of the tests. The colony had been housed for one year in a 1.7 x 2.1 x 2.3 m outdoor flight cage in Gamboa, Panama. All subjects were individually marked with visible unique forearm-bands.
From November 3 to 11, 2016, we conducted the first 12 trials starting at sunset (1750–1825 h) and then conducted eight additional trials from March 19 to April 12, 2017, later at night (2000 h). In each trial, a novel object roughly the size of a shoebox was placed on the flight cage floor for one hour and filmed with infrared illumination and a Sony Nightshot camcorder. Examples of novel objects were a 2-liter water bottle, a cardboard box, a tripod, and a rubber boot. The novel object and location was different for each trial. We used the statistical methods described above for experiment 1 to test if observed visits were disproportionately made by young bats, and to test for a difference in the mean proportion of bats that visited for each of three categories: young females, young males, and adult females.
We only analyzed visits in which bats could be reliably identified. In the first 12 trials conducted at sunset, we observed only a single visit, and the visiting bat could not be identified. In the next eight trials that we conducted later in the night at 2000 h, we could reliably identify the visiting bat in 23 of 54 total visits. Therefore, the visitation rate differed dramatically the first and second set of trials (permutation test: p = 0.005). In the second set, the identified visitors included two adult females, three young females, and three young males. As in experiment 1, age category influenced the proportion of bats that visited the novel object long enough to be identified; 19 of these 23 visits were by young bats (Fisher’s exact test: odds ratio = 0.09, p < 0.001; permutation test, p = 0.005,
Dots of each color show responses for trials in which the individual could be reliably identified. The majority of the proportions of visiting young bats were higher than the highest observed proportion of visiting adults.
As previously demonstrated in other mammals and birds [
By explicitly testing responses to novelty, our results also corroborate and extend previous evidence of age-dependent exploration in vampire bats [
Taken together, these observations suggest that, in comparison to adults, both young male and young female common vampire bats are more exploratory than are adult vampire bats of either sex. The age difference in our results corroborates other observations that young mammals might be more inclined to explore novel objects and situations as they are learning about their environment. In some species, immatures or adolescents peak in explorative behavior compared to younger infants and older adults [
Age-dependent “neophobia” might result from mechanisms other than fear. In experiment 1, adults responded to trapped conspecifics and not novel objects, but it is unclear whether this difference is due to greater fear or less interest in the inanimate objects. More experienced adults might simply be faster to classify a novel object as irrelevant or uninteresting, rather than dangerous [
The social environment is also likely to influence reactions to novelty because risk is reduced in groups. In social species, such as vampire bats, younger individuals are expected to use social cues for when and what to explore in their environment [
In birds, dietary specialists are reported to be less exploratory and more neophobic, compared to opportunistic foragers that require increased behavioral flexibility [
Novel object exploration rates can vary with time of day and with situational risk [
Finally, it would be interesting to test if younger vampire bats are more likely to feed on host species that are unfamiliar to them or that are in novel locations. If younger vampire bats are more likely than adults to feed on atypical hosts, such as dogs or humans, rather than cattle, this could have implications for pathogen spillover because younger vampire bats appear to experience higher exposure to pathogens such as rabies virus [
Experiments were approved by the Smithsonian Tropical Research Institute Animal Care and Use Committee (#2015-0915-2018-A9) and by the Panamanian Ministry of the Environment (#SE/A-76-16).
(R)
We thank Jana Nowatzki, Ellen Jacobs, and Katharina Eggert for help with data collection. We thank Rob Mies, Jessica Fabian, Amanda Felk, and the Smithsonian Tropical Research Institute for logistical support.