Trade-off between foraging and vigilance in the Red-legged Seriema

Technical Summary: Trade-off between Foraging and Vigilance in the Red-legged Seriema

Problem Statement
Foraging inherently increases exposure to predators, necessitating a trade-off where individuals must allocate time to vigilance alongside food acquisition. While these behaviors are often assumed to be mutually exclusive, disentangling them is complex due to the application of multiple sensory modalities. In social foraging contexts, the "many-eyes" hypothesis suggests that group members can share vigilance duties, potentially reducing individual costs and increasing net energetic gains. However, it remains unclear how this trade-off manifests in pair-living species, particularly when foraging involves complex manipulation that impairs specific sensory modalities (e.g., vision). Specifically, it is unknown whether pairs of red-legged seriemas (Cariama cristata) exhibit a division of labor where one individual specializes in vigilance while the other specializes in foraging, and whether these individuals coordinate or synchronize their behaviors to maximize predator detection.

Methodology
The study was conducted at the University of Viçosa (UFV) campus in Florestal, Minas Gerais, Brazil, during the macauba palm (Acrocomia aculeata) fruiting season (mid-November to end of December). Researchers utilized wildlife video cameras to record foraging events in four distinct areas where pairs of seriemas were known to nest.

• Data Collection: Approximately 2,700 hours of daylight footage were recorded, with 48 hours containing interactions with macauba fruit. To facilitate the cracking of the hard fruit shells, researchers placed rocks (anvils) at camera sites and supplemented the area with 15–20 macauba fruits every other day.
• Individual Identification: As the species is sexually monomorphic and individuals were not ringed, individual identification across different recordings was not possible. Analysis was restricted to behavioral differences within single recording sessions (i.e., comparing two individuals present in the same frame).
• Behavioral Scoring: A custom ethogram classified six behaviors: pecking, picking up, eating, throwing, dropping fruit, and vigilance. Videos were filtered to include only instances where seriemas interacted with fruit.
• Detection and Analysis: A "Seriema Detection Model" (YOLOv8) was developed to assist in video selection, achieving 93.7% precision and 89.5% recall. Behavioral data were extracted using BORIS.
• Statistical Approach: Generalized Linear Mixed Models (GLMMs) were employed to test for differences in time allocation between individuals within a pair. Variables included vigilance proportion (VP), eating proportion (EP), and throwing rate (TR). The study also calculated predicted versus observed probabilities of vigilance and eating to test for synchronization (behavior overlapping more than chance) or coordination (behavior minimizing overlap).

Key Contributions and Results
The study provides the first large-scale behavioral dataset on wild red-legged seriemas, revealing a significant asymmetric effort allocation within foraging pairs:

1. Division of Labor: Within a given recording, one individual consistently spent significantly more time vigilant (mean 38%) than its partner (mean 22%). Conversely, the partner with lower vigilance spent approximately twice as much time eating (29% vs. 13%) and threw fruit at a rate roughly six times higher (0.74 vs. 0.13 throws per minute).
2. Negative Correlation: Statistical models confirmed a negative relationship between vigilance and foraging metrics. Individuals that were more vigilant than their partner threw fewer fruits and spent less time eating. Specifically, a 10 percentage point increase in vigilance relative to a partner correlated with approximately 1.2 fewer throws and a 4.2 percentage point reduction in eating time.
3. Lack of Coordination/Synchronization: Contrary to the hypothesis that pairs would coordinate to ensure continuous group vigilance, the study found no evidence of synchronization or coordination. The observed probability of at least one individual being vigilant (or eating) did not differ significantly from the predicted probability assuming independent behavior. This indicates that individuals do not actively time their vigilance to avoid simultaneous gaps in alertness during a feeding bout.
4. Throwing vs. Eating: No significant relationship was found between the difference in throwing rates and the difference in eating proportions, suggesting that individuals may consume fruit opened by their partners without necessarily throwing the fruit themselves.

Significance and Claims
The authors claim that these results offer insight into the foraging ecology of red-legged seriemas and the specific trade-offs between foraging and vigilance in social contexts. The findings support the existence of a trade-off where one individual in a pair assumes a higher vigilance burden, allowing the other to focus on the energetically costly task of manipulating and cracking macauba fruit.

However, the paper maintains a modest stance regarding the mechanisms driving this asymmetry:

• Role of Reciprocity: While the lack of coordination within a session suggests the birds are not actively managing group alertness in real-time, the authors suggest that long-term pair bonds (monogamy) could facilitate role alternation between feeding sessions via direct reciprocity. This hypothesis remains untested due to the inability to identify individuals across different days.
• Predation Risk: The lack of coordination is attributed to potentially low predation risk in the study area, suggesting that the energetic cost of coordinating vigilance may outweigh the benefits when predator density is low.
• Limitations: The study explicitly notes that without individual identification, it cannot confirm if these roles are fixed (e.g., one bird is always the "vigilant" one) or if they switch roles over time. Furthermore, sex differences could not be analyzed, though the authors acknowledge that in many monogamous birds, males often invest more in vigilance.

In conclusion, the paper demonstrates that while red-legged seriemas exhibit a clear division of labor during specific foraging events, they do not coordinate their vigilance behaviors in a synchronized or complementary manner within those events.