Limits to length asymmetry detection in starlings: implications for biological signalling

Fluctuating asymmetry has received considerable recent attention in evolutionary biology as these small developmental asymmetries can be related to biological fitness and, hence, could be used as a visual cue (or signal) of quality among individuals. The ability of signal receivers to detect and respond to small asymmetries is a fundamental assumption of the symmetry–signalling hypothesis, but has not been experimentally investigated. In this study I have investigated the perceptual threshold to detect and respond to paired–bar length asymmetry in a common bird, the European starling Sturnus vulgaris, by means of operant–learning techniques. The threshold indicates how large the length asymmetry must be to be reliably discriminated from symmetry; birds could not detect an asymmetry of 1.25%. In nature, many asymmetries can be smaller than 1.25%, hence this initial study suggests that caution should be used when trying to invoke symmetry–signalling in natural populations.