An 8-year longitudinal study of mirror self-recognition in chimpanzees (Pan troglodytes)

https://doi.org/10.1016/S0028-3932(02)00153-7Get rights and content

Abstract

In a previous cross-sectional study of mirror self-recognition involving 92 chimpanzees, Povinelli et al. [Journal of Comparative Psychology 107 (1993) 347] reported a peak in the proportion of animals exhibiting self-recognition in the adolescent/young adult sample (8–15 years), with 75% being classified as positive. In contrast, only 26% of the older animals (16–39 years) were classified as positive, suggesting a marked decline in self-recognition in middle to late adulthood. In the present study, all of the chimpanzees from the 8–15-year-old group in the Povinelli et al. study (n=12) were again tested for self-recognition, 8 years later. Using the same criteria, 67% of the animals were classified the same. Although a higher proportion of the adult animals in this study (50%) exhibited self-recognition than would be inferred on the basis of the previous study (25%), all changes in self-recognition status were in the negative direction. These results show that mirror self-recognition is a highly stable trait in many chimpanzees, but may be subject to decline with age. Connections with human research are briefly discussed.

Introduction

Gallup [6] demonstrated that chimpanzees (Pan troglodytes) are capable of what has been termed self-recognition in mirrors. Typically, chimpanzees initially react to their mirror image as if it were another chimpanzee, but after a period ranging from a few minutes to several days (see [11]), they use the mirror to explore otherwise difficult to view parts of their bodies. To validate the impression of self-recognition that arose from observing the animals engaging in self-exploratory behaviors, Gallup [6] designed a test where the animals were sedated and marked on the right eyebrow ridge and upper left ear. After recovery, the chimpanzees showed clear and selective attempts to touch and investigate these marks while in front of the mirror (for evidence of the specificity of these touches, see [12]).

In a cross-sectional study involving a sample of 92 chimpanzees, Povinelli et al. ([11], Exp. 1) obtained evidence that self-recognition typically emerges in young adolescence; relatively few animals younger than 8 years of age displayed compelling instances of self-exploratory behavior, whereas 75% of the subjects aged 8–15 years (n=12) did so. Povinelli et al. also reported the surprising finding that only 26% of the older adults (16–39 years, n=35) exhibited evidence of mirror-mediated patterns of self-exploration. Humans begin to show mirror self-recognition between 18 and 24 months of age (e.g. [1]) and do not typically lose the capacity until they enter the advanced stages of senility (for a review see [7]). Thus, on the basis of the Povinelli et al. sample, the onset of self-recognition in chimpanzees would appear to be delayed relative to humans, and the loss of self-recognition in later life by chimpanzees may be accelerated.

Povinelli et al. [11] also provided evidence for a dissociation between the two primary measures of mirror self-recognition: spontaneous self-exploratory behaviors and the results of mark tests. Thirty-six subjects who had been coded as either self-recognition positive (SR+, n=18), negative (SR−, n=12) or ambiguous (SR?, n=6) on the basis of their spontaneous patterns of mirror-mediated self-exploratory behavior, were administered mark tests. Positive results on the mark test came almost exclusively from animals that had been diagnosed as SR+ (82%, 9/11 cases), and animals the were coded as SR? or SR− exhibited only rare instances of passing the mark test (11%, 2/18 cases). However, only half of the SR+ animals passed the mark test (50%, 9/18 cases). These data add confidence to the scheme developed by Povinelli et al. for coding the spontaneous behavioral reactions of chimpanzees to their mirror images (see Section 2), by showing that the coding system does not lead to a substantial number of false negatives (i.e. SR− diagnosis based on spontaneous behavioral reactions to mirror, but SR+ diagnosis on mark test).

The main purpose of the present study was to examine the long-term stability of one of the measures of mirror self-recognition (self-exploratory behavior) in the adolescent subjects of Povinelli et al. [11]. To do so, we used a longitudinal design and re-tested the 12 chimpanzees from the 8–15-year-old group in the Povinelli et al. [11] study 8 years after the original study. These data can help to determine whether the apparent decline in mirror-mediated self-exploratory behavior among adult chimpanzees indicated by the cross-sectional data of Povinelli et al. is a robust phenomenon. A secondary objective was to re-examine data from the original study to further explore the relationship between measures of spontaneous self-exploratory behavior (the measure used here) and the results of the mark tests.

Section snippets

Subjects

Subjects were 12 chimpanzees (Pan troglodytes) housed in social groups (range=6–9 animals) at the University of Louisiana (enclosure dimensions: exterior, 8m×5.5m×4 m; interior, 4.5m×5.5m×3 m). Eight years earlier, when they were between 8 and 15 years of age (see Table 1), these subjects had participated in the mirror self-recognition experiments of Povinelli et al. [11]. At the time of the present tests, the subjects ranged in age from 16 to 23 years.

Materials

For testing, a one-way mirror (47cm×43 cm)

Results

The central focus of this investigation was the stability of the SR status of each animal from 1992 to 2000, and whether a substantially lower percentage of animals exhibited evidence of self-recognition from 1992 to 2000.

Discussion

Consistent with the possibility of an apparent decline with age, all of the changes in self-recognition status from the previous study [11] were in the negative direction. Two of the three chimpanzees that were SR+ in 1992 tested SR? in 2000, and the other tested SR−. Another animal moved from SR? to SR−. Finally, and of great interest concerning the stability of the trait, none of the animals that failed to show evidence of self-recognition in 1992 tested SR+ or SR? in 2000.

On the other hand,

Acknowledgements

We thank James Reaux and the caretakers and staff of the New Iberia Research Center for their assistance with this study. The research was supported by a Centennial Fellowship to DJP from the James S. McDonnell Foundation and by a travel grant from the Netherlands Organization for Scientific Research (NWO) and Utrecht University to MdV. We thank James Reaux for the variety of assistance he provided the design and implementation of this study. We thank Godelieve Kranendonk, Ryan Arnold and Sarah

References (13)

There are more references available in the full text version of this article.

Cited by (0)

1

Tel.: +31-302-53-4851; fax: +31-302-53-7727.

2

Tel.: +1-518-442-4852; fax: +1-518-442-4867.

View full text