Distractible dogs, constant cats? A test of the distraction hypothesis in two domestic species

Dogs, Canis familiaris, and other domestic species are more skilled than great apes at following human communicative gestures in object choice tasks. Several hypotheses differentially emphasizing the role of domestication, socialization and experience have been put forward to account for this discrepancy. Recently, it has been suggested that the performance gap between apes and domestic species could be due instead to inconsistent methodologies across studies. Apes have generally been tested with the containers directly in front of them, and the experimenter performing the communicative cue behind the containers (central set-up). In contrast, domestic animals are usually placed at a larger distance with the human experimenter performing the cue in their direct line of sight and the containers peripheral on both sides of the experimenter (peripheral set-up). The distraction hypothesis posits that the close proximity of the food-associated containers in the central set-up proves too distracting for the test subjects, leading to an indiscriminate choice behaviour. Consistent with this idea, great apes are able to solve the peripheral version of the object choice task. To evaluate the distraction hypothesis further, we tested domestic dogs and cats, Felis catus, in the central as well as the peripheral set-up. As predicted, dogs' performance dropped significantly when tested in the central version of the task compared with the peripheral one. Cats performed as well as dogs in the peripheral version, but, intriguingly, their success levels did not decline significantly in the central version. We speculate that this might be due to cats cuing into the movement part of the pointing gesture. These findings partly support the distraction hypothesis and add to the evidence that domestic species do not necessarily have superior skills in reading human communicative cues.