RESEARCH PAPERS Is Preening Behaviour Sexually Selected? An Experimental Approach Matteo Griggio & Herbert Hoi Konrad Lorenz Institute for Ethology, Vienna, Austria Introduction Comfort behaviour, in particular preening, is one aspect of self maintenance behaviour which has not attracted many investigations. Preening requires time which may result in a temporal trade-off between investment in preening and other activities (Redpath 1988; Cucco & Malacarne 1997). Further- more, preening produces public information which may vary depending on the receivers, i.e. the infor- mation transferred may differ for conspecific group members, predators or potential mates (Danchin et al. 2004). In birds for example, to maintain the plumage in good condition is essential for flight but also for mate choice. There is evidence for the importance of ornamental traits in mate choice and that females benefit by choosing a male based on the expression of elaborate or colourful plumage ornaments (for a review see Andersson 1994). Con- spicuous plumage signals are assumed to provide the choosy sex with information about potential mates. Many studies have demonstrated that conspicuous plumage ornamentation correlates with aspects of individual quality such us parasite resistance, immu- nocompetence, diet and social status (e.g. Zahavi 1975; Hill & Montgomerie 1994; Swaddle & Witter Correspondence Matteo Griggio, Dipartimento di Biologia, Universita ` di Padova, Via U. Bassi 58/B, I-35131 Padova, Italy. E-mail: matteo.griggio@unipd.it Received: March 27, 2006 Initial acceptance: April 4, 2006 Final acceptance: April 30, 2006 (J. Schneider) doi: 10.1111/j.1439-0310.2006.01270.x Abstract Elaborate or colourful feathers are important traits in female mate choice in birds but little attention has been given to potential costs of maintaining these traits in good condition with preening behaviour. Recent studies indicate that the time and energy required to maintain ornamental plumage in good condition reinforces the honesty of plu- mage trait. It has been proposed that some behaviours, whose primary function is not to transfer information, can also evolve as signalling components. Here we investigate whether the preening behaviour intensity has a signalling component: we hypothesized that if only high quality males can invest a lot of time in preening, this behaviour may be used by females as a quality signal (attractive preening hypothesis). We tested this hypothesis by using female budgerigars in mate-choice tests in captivity. We tried to experimentally manipulate the preening behaviour of two groups of budgerigar males (treatment and control group). The proportion of time in which treated males preened in front of females was statistically higher than for control males, however, females spent similar amounts of time with treated males and control males. Moreover, males did not show significant quantitative changes in preening (for both groups) when females were present, suggesting that male budgerigars did not use this behaviour to convey information. These results are inconsistent with the ‘attractive preening’ hypothesis which predicts that preening behaviour itself provides information on condition and is used in female choice. Ethology Ethology 112 (2006) 1145–1151 ª 2006 The Authors Journal compilation ª 2006 Blackwell Verlag, Berlin 1145