Unpredictability of vigilance in two sympatric Tibetan ungulates

Using a behaviour random permutation model to identify displacement grooming in ungulates

Behaviour affects an individual’s life in all aspects, e.g., enhancing fitness, leveraging predation risk, and reducing competition with conspecifics. However, the sequential distribution of behaviours received less attention and is unclear what the function of displacement behaviour is. Displacement activities can be found in vertebrate species but there is no formal method to determine whether a behaviour is expressed as a displaced or normal activity. Analysing the sequential distributions of behaviours in a natural setting may allow researchers to identify unexpected distributions as a possible signature of displacement activities. In this study, we used a behaviour random permutation model to detect the presence of a displacement activity in the Tibetan antelope Pantholops hodgsonii and the Tibetan gazelle Procapra picticaudata. The results showed that grooming in both ungulates tended to be accompanied with vigilance, and the frequency of grooming after vigilance was significantly higher than before vigilance. A significant positive correlation between the scan rate and grooming rate in the two ungulates was obtained. We suggest that grooming could sometimes be expressed as a displacement activity in ungulates. In addition to providing a general method for further research on displacement activities in a variety of animal species, this study sheds light on the importance of a spectral analysis of sequential distribution of animal behaviours. Behaviour random permutation models can be used to explore the relevance between any two behaviours in a specific sequence, especially to identify a myriad of unexpected behaviours relative to their normal context of occurrence.

Nonrandom pattern of vigilance by preening black-headed gulls

Classic models of vigilance assume instantaneous and sequential randomness in the scanning process, implying negative exponential distribution of interscan durations and no interdependence among successive interscans. We examined whether vigilance pattern by preening black-headed gulls, Chroicocephalus ridibundus, meets these assumptions. Out of 54 behavioural sequences, 50 departed from the expected negative exponential distribution, whereas the focal interscan duration was significantly affected by the interaction of the preceding scan and the interscan interval. These results reveal departures from randomness in the scanning process by gulls, which may be a consequence of the hunting strategies of their predators or due to the trade-off between the needs for feather maintenance and antipredator vigilance.