Does sentinel presence improve foraging success? A study in a cooperative breeder

Ostreiher, Roni 1 ; Mundry, Roger2 ; Heifetz, Aviad1

Published Jun 13, 2024 on Dryad. https://doi.org/10.5061/dryad.kwh70rzcz

Data files

Jun 13, 2024 version files 276.37 KB

code_experiment.r
52.04 KB
code_natural_observations.r
39.60 KB
experiment_data.txt

36.15 KB
functions.r
100.28 KB
nat_obs_data.txt

41.55 KB
read_me.txt

3.16 KB
README.md
3.61 KB

Abstract

In various cooperative breeding species, a group member often sentinels from a prominent position that enables early detection of approaching predators and then utters alarm calls to which the other group members react immediately. We tested in the cooperatively breeding Arabian babbler (Argya squamiceps), whether in the presence of a sentinel the other group members forage for longer or more effectively. In natural observations, at different distances from the shelter and in different group sizes, we found no effect of sentinel presence on foraging time. Likewise, in a giving-up density (GUD) experiment with artificial food patches containing mealworms at each of three different distances from the shelter, fewer mealworms were consumed further from the shelter, irrespective of whether a sentinel was present or not during foraging. While foragers react to a sentinel’s alarm calls when a predator approaches, our findings indicate that Arabian babbler foragers do not improve their foraging efficiency due to the sentinel's presence. Combined with previous studies on the Arabian babbler, these findings reinforce the conclusion that in this species the sentinel is 'selfish' and benefits from early detection of predators. These results contrast with those from other cooperative breeders in which sentinels utter calming calls and foraging efficiency increases when sentinels are present. We, therefore, suggest that similar sentinel behaviour may serve different sentinel systems and functions in different cooperative breeders.

by Roni Ostreiher, Roger Mundry & Aviad Heifetz

Files:

experiment_data.txt and nat_obs_data.txt are tab-separated text files (i.e., plain asci)
nat_obs_data.txt contains the data collected during the natural observations
experiment_data.txt contains the data collected during the experiment
code_natural_observations.r and code_experiment.r contains the R code for the analysis

Columns in nat_obs_data.txt:

row.id: consecutive integers with row numbers;
subject: ID of an individual (nested in the group, i.e., the same ID in different group means different individual);
obs.id.within.group: ID of the trial within a group;
date: the date at which a trial was conducted [day, month, year; year with 2 or 4 digits];
foraging_within_sentinel_yes: total time foraging within the shelter when sentinel was present [seconds];
foraging_within_sentinel_no: total time foraging within the shelter when sentinel was absent [seconds];
foraging_near_sentinel_yes: total time foraging near the shelter when sentinel was present [seconds];
foraging_near_sentinel_no: total time foraging near the shelter when sentinel was absent [seconds];
foraging_far_sentinel_yes: total time foraging far from the shelter when sentinel was present [seconds];
foraging_far_sentinel_no: total time foraging far from the shelter when sentinel was absent [seconds];
preening_sentinel_yes: total time preening when sentinel was present [seconds];
preening_sentinel_no: total time preening when sentinel was absent [seconds];
interactions_sentinel_yes: total time spent with social interactions when sentinel was present [seconds];
interactions_sentinel_no: total time spent with social interactions when sentinel was absent [seconds];
moving_sentinel_yes: total time moving when sentinel was present [seconds];
moving_sentinel_no: total time moving when sentinel was absent [seconds];
resting_sentinel_yes: total time resting when sentinel was present [seconds];
resting_sentinel_no: total time resting when sentinel was absent [seconds];
total_time_sentinel: total time a sentinel was present (obsolete and not used);
total_time_foraging: total time foraging (obsolete and not used) [seconds];
total_time_other_behavs: total time spent with other behaviours (obsolete and not used) [seconds];
obs_dur: total observation duration (obsolete and not used) [seconds];
group: ID of the group;
tot.time.sentinel.yes: total time a sentinel was present [seconds];
tot.time.sentinel.no: total time a sentinel was absent [seconds].

Columns in experiment_data.txt:

group: ID of the group;
repetition: index number of the experiment (i.e., order of experiments, per group) [consecutive integer per group];
date: the date at which the experiment was conducted [D.M.YYYY, DD.M.YYYY, D.MM.YYYY, or DD.MM.YYYY];
seconds: duration of feeding [seconds];
mealworms: number mealworms eaten [count];
distance: distance from the shelter in which the tray with the mealworms was placed [factor with levels close, middle, and far];
context: dominance (dominant or subordinate) and sex of the sentinel if there was one present and otherwise ‘Without sentinel’;
r.date: date in nr. days elapsed since 1970-01-01 [integer];
group.size: size of the group [nr. individuals];
winterID: ID of the testing winter (extends over the change of years);
day.in.season: nr. days elapsed since the first experiment of a given testing winter [days];sentinel: sentinel status (absent or present).