Recognition of human faces by dogs (Canis familiaris) requires visibility of head contour
Researchers have suggested that dogs are able to recognise human faces, but conclusive evidence has yet to be found. Experiment 1 of this study investigated whether dogs can recognise humans using visual information from the face/head region, and whether this also occurs in conditions of suboptimal visibility of the face. Dogs were presented with their owner’s and a stranger’s heads, protruding through openings of an apparatus in opposite parts of the experimental setting. Presentations occurred in conditions of either optimal or suboptimal visibility; the latter featured non-frontal orientation, uneven illumination and invisibility of outer contours of the heads. Instances where dogs approached their owners with a higher frequency than predicted by chance were considered evidence of recognition. This occurred only in the optimal condition. With a similar paradigm, Experiment 2 investigated which of the alterations in visibility that characterised the suboptimal condition accounted for dogs’ inability to recognise owners. Dogs approached their owners more frequently than predicted by chance if outer head contours were visible, but not if heads were either frontally oriented or evenly illuminated. Moreover, male dogs were slightly better at recognition than females. These findings represent the first clear demonstration that dogs can recognise human faces and that outer face elements are crucial for such a task, complementing previous research on human face processing in dogs. Parallels with face recognition abilities observed in other animal species, as well as with human infants, point to the relevance of these results from a comparative standpoint.
KeywordsDogs Human face recognition Illumination Outer face parts Owner Viewpoint
We are very grateful to Pamela Sambugaro and Carlo Poltronieri for their technical assistance and to all the dogs’ owners for volunteering to take part. This research was supported by a research grant from the University of Padova (to P.M., Grant Nr. CPDA144871/14), a postdoc grant awarded by the University of Padova (to A.S., Grant Nr. CPDR148844/14) and a study visit grant awarded by the Experimental Psychology Society (to R.S.S.K.).
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interests.
All procedures performed in the present study were in accordance with the ethical standards of the University of Padua and the Italian relevant legislation.
- Braje WL, Kersten D, Tarr MJ, Troje NF (1998) Illumination effects in face recognition. Psychobiology 26:371–380Google Scholar