Elsevier

Behavioural Processes

Volume 67, Issue 3, 30 November 2004, Pages 491-499
Behavioural Processes

Transfer between views of conspecific faces at different ages or in different orientations by sheep

https://doi.org/10.1016/j.beproc.2004.08.005Get rights and content

Abstract

Sheep are able to discriminate photographs of conspecific faces. The present study investigates adult ewe's recognition of faces of the same animal between different ages or between different orientations. Twelve adult sheep were first trained to discriminate between faces of two unfamiliar animals, one of which was associated with a food reward. Transfer of discrimination from this pair to the same pair but at a different age, or in a different orientation, was then evaluated (transfer test), and compared with a new pair of the same condition (control test). Learned discrimination of a frontal view of unfamiliar 3-month-old lambs’ faces improved subsequent discrimination of the same pair when they were 1-month-old in comparison to discrimination of new 1-month-old faces. Moreover, sheep that were trained to discriminate frontal views of unfamiliar adult individuals discriminated profile views of the same animals more accurately than that of novels. However, learned discrimination of the profile view of unfamiliar adult faces had no effect on subsequent discrimination of the frontal view of that same pair. These results suggest that to some extent sheep recognise faces of unfamiliar animals at different ages and in different orientations.

Introduction

Face recognition is perhaps the most commonly used process of individual recognition in primates, including humans (Parr et al., 2000, Pascalis and Bachevalier, 1998). This is demonstrated in natural conditions as well as in artificial conditions using two-dimensional pictures (for review: Bovet and Vauclair, 2000). Primates are not only capable of simple face identification but also can generalise over different facial orientations of the same individual, particularly on a front-profile dimension. In humans, transfer between full-face and three-quarter portraits had no effect upon recognition, suggesting that faces displayed in three-quarter are recognised as accurately as frontal views (Bruce et al., 1987, Davies et al., 1978, Logie et al., 1987). Facial identification after changing the orientation from front-to-profile, or from profile-to-front, is also possible, though it generally produces decrements in performance (Bruce et al., 1987, Laughery et al., 1971, Logie et al., 1987, Patterson and Baddeley, 1977, Seamon et al., 1978). Chimpanzees, rhesus monkeys and longtailed macaques are also able to match numerous novel face views of conspecifics previously used in training (Dasser, 1987, Parr et al., 2000, Rosenfeld and Van Hoesen, 1979). For instance, rhesus monkeys generalise from frontal views of monkey faces to views of the same faces if shown in three-quarter position (Rosenfeld and Van Hoesen, 1979). Moreover, longtailed macaques correctly identify faces regardless of whether they are rotated within the picture plane (Dittrich, 1990) or from frontal to profile views (Dittrich, 1994). Thus, it is clear that humans and other primates can generalise over different facial orientations of the same individual.

Face recognition is not limited to primates and recent evidence suggests that sheep are also able to recognise individual conspecifics visually. Sheep, like primates (Perret et al., 1987, Rolls, 1994), have cells in the temporal cortex, which respond preferentially to projected images of faces. These face-selective responses are influenced by different factors, such as species, breed and familiarity. Some cells discharge more in the presence of frontal views of sheep whereas other cells respond preferentially to human and dog faces (Kendrick and Baldwin, 1987, Kendrick, 1994, Kendrick, 1998, Peirce and Kendrick, 2002). Among cells responding best to sheep faces, a sub-group responds preferentially to the breed of the tested animal, and particularly to familiar animals (Kendrick and Baldwin, 1987). This face-recognition ability is also confirmed by behavioural studies. In a spontaneous choice test between two black-and-white photographs, sheep discriminate between frontal views of sheep and human faces, different breeds, and male and female sheep (Kendrick et al., 1995). In similar conditions, sheep learn significantly faster to discriminate between frontal views of familiar conspecific faces compared with unfamiliar ones (Kendrick et al., 1996, Peirce et al., 2001), and this familiarity effect is also demonstrated with human faces (Peirce et al., 2001). This face-recognition memory is very efficient, since sheep can remember 50 different conspecific faces over 2 years (Kendrick et al., 2001a). Moreover, sheep are able to transfer between different facial orientations. If sheep are trained to associate a frontal view of an individual with food reward they will also associate a profile view of that same individual with a reward without receiving any further training (Kendrick et al., 2001b).

However this transfer ability has not been investigated in other contexts. For instance, the ability of sheep to transfer across faces of the same animal at different ages remains to be elucidated. To address this question, we used slides of 1- and 3-month-old lamb faces since previous research demonstrated that adult sheep are able to discriminate between 1-month-old lamb faces (Kendrick et al., 1995). In a first step, sheep learned to discriminate between frontal views of the faces of two unfamiliar 3-month-old lambs. Then, we investigated if they associate the frontal views of 1-month-old faces of the same animals (transfer condition) more accurately than those of previously unknown 1-month-old lambs (control condition).

The ability of sheep to transfer from frontal to profile views was recently demonstrated in Dorset breed (Kendrick et al., 2001b). We first reinvestigated in another breed that learned discrimination of frontal views of two unfamiliar adult faces improves subsequent transfer tests using profile views of the same animals. The transfer from profile to frontal views was also suggested as possible in Dorset ewes but never demonstrated (Kendrick et al., 2001b). Therefore, we investigated whether sheep are able to perform the opposite operation, i.e. when profile views of conspecific faces are presented during discrimination learning and frontal views during subsequent transfer tests.

Section snippets

Animals and housing

Twelve adult multiparous Ile-de-France ewes, aged 3–5 years, were used. This is a hornless breed with high facial homogeneity. All tested sheep had been extensively trained in the discrimination procedure for 3 months. They were permanently housed in sheds and fed dehydrated lucerne, maize, straw and a supplement of vitamins and minerals, with free access to water.

Stimuli used

The test stimuli consisted of photographs of frontal and profile views of unfamiliar lambs (1 and 3 months old) and unfamiliar adult

Training

All 12 sheep rapidly learned to discriminate the frontal views of each pair of unfamiliar 3-month-old lamb faces over a period of 2–5 days (mean ± S.E.M. = 53.3 ± 5.5 trials to criterion). The day before the transfer test (1-month-old lamb faces), the average performance was 93.3 ± 1.2% correct choice. This percentage of correct choice was well above chance (one-sample Sign test, p < 0.001, Fig. 3A).

The discrimination of frontal views of unfamiliar adult faces required 2–5 days (65.1 ± 7.4

Discussion

Our results confirm that sheep are able to discriminate slides of frontal views of faces of unfamiliar lambs or adults and profile views of adult faces. More interestingly, sheep are capable to transfer from frontal views of 3-month-old lamb faces to 1-month-old faces, and from frontal to profile views of adult faces but not from profile to frontal views. These results demonstrate for the first time transfer between different ages of faces of the same individuals in a non-human mammalian

Acknowledgements

The authors thank E. Archer, I. Ngatumah, P. Vanbecelaere and G. Vénier for their assistance during the experiments and E. Archer for the care he provided to the animals. We also thank P. Poindron for comments on this manuscript and C. Dawson and R. Porter for the English of the final version. M. Keller was supported by a grant from French Ministère de la Recherche et de la Technologie and Université Paris XIII.

References (33)

  • O. Pascalis et al.

    Face recognition in primates: a cross-species study

    Behav. Proc.

    (1998)
  • J.W. Peirce et al.

    Configurational coding, familiarity and the right hemisphere advantage for face recognition in sheep

    Neuropsychologia

    (2000)
  • J.W. Peirce et al.

    Human face recognition in sheep: lack of configurational coding and right hemisphere advantage

    Behav. Proc.

    (2001)
  • E.T. Rolls

    Brain mechanisms for invariant visual recognition and learning

    Behav. Proc.

    (1994)
  • S.A. Rosenfeld et al.

    Face recognition in the rhesus monkey

    Neuropsychologia

    (1979)
  • V. Bruce et al.

    The basis of the 3/4 view advantage in face recognition

    Appl. Cogn. Psychol.

    (1987)
  • Cited by (29)

    • The effect of age, environment and management on social contact patterns in sheep

      2020, Applied Animal Behaviour Science
      Citation Excerpt :

      Social recognition of individuals plays an important role in development of social familiarization in sheep (Keller et al., 2011). Previous studies provided evidence that adult sheep can discriminate between familiar and unfamiliar conspecifics through the visual channel (Kendrick et al., 2001; Ferreira et al., 2004), and through olfactory and auditory signals (Keller et al., 2011). In our study, social familiarity, resulting from repeated interactions between ewes, may be an important factor influencing associations among individuals in this species.

    • Common and divergent psychobiological mechanisms underlying maternal behaviors in non-human and human mammals

      2015, Hormones and Behavior
      Citation Excerpt :

      More recent studies using photographs support the view that the lambs' faces are important for maternal recognition. Ewes can be trained to discriminate among photographs of faces of unfamiliar lambs (Ferreira et al., 2004), as well as between familiar versus unfamiliar lambs (Kendrick et al., 1996). Audition is crucial for distal communication, especially for bringing the mother and young together.

    View all citing articles on Scopus
    View full text