Observation learning in day-old chicks using a one-trial passive avoidance learning paradigm

We tested the hypothesis that day-old chicks, Gallus gallus domesticus, can learn to avoid an aversive stimulus if they observe the responses of another chick. In experiment 1, one of a pair of chicks (the actor) was allowed to peck at a bead coated in the bitter-tasting substance methylanthranilate (MeA), while we prevented the other chick (the observer) from pecking the bitter-tasting bead by separating the chicks with a piece of wire mesh. Both chicks avoided pecking at a similar but dry bead 0.5, 3 and 24 h after the observer chick saw the actor chick peck at an MeA-coated bead. By contrast, when the actor chick had pecked at a water-coated bead, both chicks continued to peck at a dry bead at 0.5, 3 and 24 h after training. Experiment 2 investigated whether observer chicks showed avoidance if they were prevented (by the insertion of an opaque barrier) from observing their companion pecking at the MeA-coated bead during either training or testing. Observer chicks that could not see their companion during training but could observe the actor chicks at test showed no subsequent avoidance whereas chicks that observed the actor chick at training, but not during testing, showed high levels of avoidance. Although the sensory cues (visual, auditory or olfactory) or types of behaviour (i.e. levels of pecking or head shaking) that the observer chick used to maintain avoidance remain unclear, the results show that chicks can learn about an aversive object by observing the responses of a conspecific. (C) 1998 The Association for the Study of Animal Behaviour.