Disentangling risk dilution and collective detection in the antipredator vigilance of semipalmated sandpipers in flocks

Collective detection and predation risk dilution are two nonexclusive mechanisms that can allow a reduction in antipredator vigilance in large animal groups. The two mechanisms make remarkably similar predictions in many contexts, and disentangling their relative contribution has proven empirically difficult thus far. We examined a situation where predators are more likely to attack from one side of the group. Provided that all foragers in the group can detect the predator equally well, the collective detection hypothesis predicts that all individuals will adopt the same vigilance level if this mechanism acted alone. In terms of risk dilution, individuals that are on the riskier side of the group, namely those on the attack side, are unlikely to benefit to the same extent as other groupmates from the presence of companions, and should have a higher level of vigilance than more buffered foragers further inside the group. We show that semipalmated sandpipers on the riskier side of flocks were more vigilant and thus pecked at a lower rate than those in the centre of the flocks and those on the less risky side. The proportion of successful pecks did not vary with spatial position within the flocks, suggesting that differences in food availability did not confound the above findings. We conclude that it is unlikely that collective detection acted alone and that risk dilution, acting alone or in combination with collective detection, shaped antipredator vigilance in an avian prey under predation risk.