Sustained vigilance and animal performance

Psychological studies have established that the central nervous system cannot sustain vigilance for an extended period of time. The consequent vigilance decrement implies a gradual reduction in the ability to process information effectively. This may result in a reduced ability to detect hidden predators, locate cryptic food items or make successful behavioural decisions. A model incorporating parameters of vigilance decrement during foraging and vigilance recovery during rest predicts first, the optimal allocation of time between foraging activity and rest, and second, the optimal length of each foraging episode. For a given rate of vigilance recovery, the model predicts that both the optimal proportion of time devoted for foraging activity and the optimal length of a foraging bout should be decreasing functions of the rate of vigilance decrement. Because the rate of vigilance decrement is larger for more difficult tasks, both the total time spent foraging and the duration of each foraging episode should be smaller for more demanding foraging activities. It is suggested that vigilance decrement is a dominant factor determining temporal patterns of animal behaviour. One of the fundamental limitations of the central nervous system is its inability to sustain a high quality of information processing for an extended period of time. Consequently, performance on tasks such as the detection of cryptic targets degrades over time; this gradual reduction in performance is more pronounced for more difficult tasks (Nuechterlein et al. 1983; Warm 1984; Parasuraman & Mouloua 1987). In psychological terms, subjects cannot sustain vigilance, or attention, while engaged in a difficult cognitive task. Here ‘vigilance’ is used in its broad sense to imply a state of alertness directed towards certain stimuli; a higher level of vigilance enables the brain to better process and respond to these stimuli (Mackworth 1969; Mackie 1977; Parasuraman 1984). Theoretical, empirical and applied aspects of sustained vigilance have been widely studied for human performance (Mackworth 1948; Davis & Parasuraman 1982; Wickens 1984). In contrast, studies of non-human subjects have been restricted to neurobiological research (Foote et al. 1980; Aston-Jones et al. 1984). Nevertheless, the problem of sustained vigilance is highly relevant for the daily patterns and types of activity conducted by animals in their natural settings (Dimond & Lazarus 1974). To further evaluate the importance of sustained vigilance for animal performance, we present here a set of simple models. These models predict first, the optimal allocation of time between foraging activity and rest, and second, the optimal length of each foraging episode. MODELS AND PREDICTIONS Throughout this paper, we refer to ‘vigilance’ in its more general meaning usually used in the psychological literature (e.g. Davis & Parasuraman 1982). Vigilance is not only a state of behavioural alertness to predators (e.g. Lima & Dill 1990); rather, it is a general condition of enhanced ability to process information. Vigilance decrement therefore might cause either a gradual decrease in ability to notice an approaching camouflaged predator, a decline in capacity to detect cryptic prey, a reduced effectiveness in decision-making tasks, or more likely, some combination of all of the above. For brevity, we distinguish here only between activity and rest, and assume that all activity is devoted to foraging and detecting predators. Regardless, our general arguments are relevant for other kinds of activity such as mate choice, nest building, or care of offspring. 0003–3472/95/051259+09 $08.00/0 ? 1995 The Association for the Study of Animal Behaviour