Biological motion point-light displays provide a powerful method for studying motion perception. Nonhuman animals are capable of discriminating point-light displays, but it remains unknown how they perceive biological motion in these displays. We trained two groups of pigeons to discriminate video stimuli using two different classification rules. The motion-congruent group was trained to discriminate full-detail and corresponding point-light displays of pigeons from full-detail and point-light displays of humans. The motion-incongruent group was trained to discriminate full-detail pigeons and point-light humans from the other displays. Both groups acquired the discrimination. When tested with novel displays, pigeons showed good transfer of learning. Transfer was poorest with the point-light displays in the motion-congruent group. The results indicate that the pigeons failed to make the connection between the full-detail displays and their point-light counterparts even when the common motion was available as a cue.