Risk-sensitive foraging in coal tits

Current theory of risk-sensitive foraging predicts that forages should choo se feeding sites on the basis of variation in as well as mean reward rate w hen there is a shortfall in their food supply or a decrease in their. energ y budget. For a given mean reward delay, they should choose high variance f eeding sites if they al e running below energy requirement, but low valianc e sites if they are running above. It has been suggested that the smaller t he animal size, the stronger the preference reversion between high and low variable feeding sites. Previous tests of the energy budget rule when there was time variability have used bird species heavier than 80 g. Hence we te sted energy budget rule predictions with coal tits Parus ater, a bird of 9 g of body mass, foraging at two feeding sites with high or low variability in food delivering delay. We manipulated energy budgets by controlling air temperature in the laboratory. In one treatment (positive budget), individu als were allowed to cat at the level of their own ad-libitum,l daily consum ption and the air temperature was set to 24 degreesC, while for the other t reatment (negative budget), temperature was set to 14 degreesC, and food av ailability was limited to the maximum daily intake observed in the positive budget treatment. When ail temperature was low, daily intake increased but body mass decreased. Birds were also less active in the low temperature tr eatment, hopping less times every day. Latency to peck decreased as well, p eeking for food when it was available sooner than in the high temperature t reatment. These results show that coal tits were living in a negative energ y budget when air temperature was set to 14 degreesC. preference for the va riable feeding site was greatest in the negative energy budget, as it was p redicted by the energy budget rule. Contrary to the energy budget rule, coa l tits consistently preferred the variable option also in the positive ener gy budget. Possible explanations for these results are explored, including alternative foraging models to the energy budget rule.