Credit: M. W. RICHARDS/OXFORD SCIENTIFIC

It's party time. The hi-fi is booming, and people are chattering loudly. Yet you can readily hear a couple talking at the other side of the room. This is the ‘cocktail party effect’, a phenomenon familiar to attentional psychologists. And penguins.

Breeding colonies of king penguins (Aptenodytes patagonicus) on subantarctic islands can number up to 300,000 birds. Penguin chicks, of course, need feeding, and must find the parent that has been foraging at sea on the latter's return. It is known that the initial detection is by vocal cues rather than sight or smell. But how good are chicks at identifying the parent's call when it is masked by the background din of the colony and screened by intervening bodies?

T. Aubin and P. Jouventin have tackled the question (Proc. R. Soc. Lond. B 265, 1665-1673; 1998). They measured the amplitude and other acoustic properties of parental calls and the ambient noise of a king-penguin colony; they then assessed the propagation of the call in the centre of the feeding area, as compared to an open area, to quantify the screening effect of bodies. In playback experiments, a chick reacted only to calls made by its own parent, turning and then running towards the source. Further experiments involved ‘jamming’ the calls of one parent by mixing in the calls of other adults, mimicking the true situation in the feeding zone.

The authors' calculations suggest that the maximum distance at which the chick should be able to detect its parent against the background cacophony should not exceed 8-9 metres. But even when the parent's calls were jammed, chicks picked them out at nearly twice that distance. This remarkable feat of auditory discrimination is yet to be explained.